132 4

SENSATION AND PERCEPTION

These unusual perceptual events are varieties of synaesthesia – the perceptual experience

DAVID REDFERN/REDFERNS/GETTY IMAGES

of one sense that is evoked by another sense (Hubbard and Ramachandran, 2003). For
some synaesthetes, musical notes evoke the visual sensation of colour. Other people with
synaesthesia see printed letters (see FIGURE 4.1) or numbers in specific, consistent col-
ours, for example always seeing the digit 2 as pink and 3 as green. Still others experience
specific tastes when certain sounds are heard.

A B C D E FIGURE 4.1 Synaesthesia Most of

us see letters printed in black as
(a) Usual appearance
they appear in (a). Some people with
synaesthesia link their perceptions
of letters with certain colours
and perceive letters as printed in

A B C D E different colours, as shown in (b). In

synaesthesia, brain regions for different
sensory modalities cross-activate one
(b) Appearance to a person with synaesthesia another.

For those of us who don’t experience synaesthesia, the prospect of tasting sounds or
hearing colours may seem unbelievable or the product of some hallucinogenic experience.
Maybe creative types really do have Indeed, for many years scientists dismissed synaesthesia either as a rare curiosity or a case
different brains. Successful musicians, of outright faking. But recent research indicates that synaesthesia is far more common
actors and artists are common in the list of
individuals who report well-documented
than previously believed. British neuroscientist Jamie Ward (2008) estimates that every-
experiences of synaesthesia. Or maybe it is one will be closely acquainted with at least six or seven synaesthetes without necessarily
much more common in the general public knowing who they are. As noted above, this was true for one of the authors who only dis-
than previously thought and we simply hear covered his wife had always had synaesthesia after telling her about the phenomenon.
about it because we are fascinated by the
private lives of celebrities. Jazz musician
Recent research has documented the psychological and neurobiological reality of
and composer Duke Ellington (1899–1974) synaesthesia. For example, a synaesthete who sees the digits 2 and 4 as pink and 3 as
perceived timbre as colour while performing green will find it easier to pick out a 2 from a bunch of 3s than from a bunch of 4s,
on stage. whereas a nonsynaesthete will perform these two tasks equally well (Palmieri et al.,
2002). This indicates that people’s subjective reports map onto their perceptual experi-
SYNAESTHESIA The perceptual ence. Moreover, brain-imaging studies also show that in some synaesthetes, areas of the
experience of one sense that is evoked brain involved in processing colours are more active when they hear words that evoke
by another sense. colour than when they hear tones that don’t evoke colour; no such differences are seen
MODALITIES Sensory brain regions that among people in a control group (Nunn et al., 2002).
process different components of the So, synaesthesia is neither an isolated curiosity nor the result of faking. In fact, it may
perceptual world.
indicate that in some people, the brain is ‘wired’ differently than in most, so that brain
regions for different sensory modalities cross-activate one another. Indian-born neurolo-
© [Link]/PEDERK

gist Vilayanur ‘Rama’ Ramachandran’s hypothesis (Ramachandran and Hubbard, 2003)

is that synaesthesia arises because the separate modalities, sensory brain regions that pro-
cess different components of the perceptual world, are initially all interconnected during the
early postnatal months as part of the spurt of typical brain growth that takes place during
the first year of life. This explosion of interconnectivity is followed by a period when the
brain streamlines these connections that are reinforced though real-world experience
(see Chapter 11 for further details). According to Ramachandran, whereas most of us
lose interconnections between different modalities, for some reason, synaesthetes main-
tain many, which is why they experience cross-talk between modalities.
One hypothesis is that synaesthesia is an inherited ability that facilitates creativity
(Ramachandran and Hubbard, 2003, p. 29):
If some genetic factor were to cause excess connections between different brain maps,
then depending on where and how widely in the brain the trait was expressed, it could
lead to both synesthesia and to a propensity towards linking seemingly unrelated con-
A study of synaesthetes found that 24%
cepts and ideas.
were employed in artistic professions
compared to the national average of 2% This proposal is supported by an Australian study of 192 synaesthetes, which found that
(Rich et al., 2005). 24% were employed in artistic professions compared to the national average of 2% (Rich
 SENSATION AND PERCEPTION 4 133

et al., 2005). Another intriguing possibility is that synaesthesia enhances memory.

Synaesthetes typically report that they have better memory for learning lists when they
can engage their synaesthesia to enrich the learning experience – a claim borne out in
experiments that compare them with control subjects (Yaro and Ward, 2007). Whatever
the ultimate explanations for this fascinating phenomenon, the recent wave of research
shows that synaesthesia is a mindbug that can shed new light on how the brain is organ-
ized and how we sense and perceive the world.
In this chapter we’ll explore key insights into the nature of sensation and perception.
These experiences are basic for survival and reproduction; we wouldn’t last long without
the ability to accurately make sense of the world around us. Indeed, research on sensation
and perception is the basis for much of psychology, a pathway towards understanding
more complex cognition and behaviour such as memory, emotion, motivation or deci-
sion making. Yet sensation and perception also sometimes reveal mindbugs, ranging
from the complexities of synaesthesia to various kinds of perceptual illusions that you
might see at a science fair or in a novelty shop. These mindbugs are reminders that the
act of perceiving the world is not as simple or straightforward as it might seem. This is
because there are multiple stages of sensory processing that must take place before we
become consciously aware of the world around us. Our sensory organs may provide us
with the raw data but these data have to be number crunched and interpreted to provide
meaningful experience.
We’ll look at how physical energy in the world around us is encoded by our senses,
sent to the brain and enters conscious awareness. Vision is predominant among our
senses; correspondingly, we’ll devote a fair amount of space to understanding how the
visual system works. Then we’ll discuss how we perceive sound waves as words or music
or noise, followed by the body senses, emphasizing touch, pain and balance. We’ll end
with the chemical senses of smell and taste, which together allow you to savour the foods
you eat. But before doing any of that, we will provide a foundation for examining all the
sensory systems by reviewing how psychologists measure sensation and perception in the
first place.

Our senses encode the information our brains

perceive
From the vantage point of our own consciousness, sensation and perception appear to be
one seamless event. Information comes in from the outside world, gets registered and inter-
preted and triggers some kind of action: no breaks, no balks, just one continuous process.
However, psychologists have known for some time now that sensation and perception are
two separate activities. This was one of the first discoveries made by Hermann von
Helmholtz, German pioneer of experimental psychology, who we read about in Chapter 1.
He systematically studied the limits and nature of our sensory capacities and realized that
additional processing was undertaken by the brain after the signals had been registered in
the central nervous system. So, sensation is the simple awareness due to the stimulation of a SENSATION Simple awareness due to the
sense organ. It is the basic registration of light, sound, pressure, odour or taste as parts of stimulation of a sense organ.
your body interact with the physical world. After a sensation registers in your central nerv- PERCEPTION The organization,

ous system, perception, the organization, identification and interpretation of a sensation in identification and interpretation of a
sensation in order to form a mental
order to form a mental representation, takes place at the level of your brain. representation.
As an example, your eyes are coursing across these sentences right now. The sensory
receptors in your eyeballs are registering different patterns of light reflecting off the page.
Your brain, however, is integrating and processing that light information into the mean-
ingful perception of words, such as ‘meaningful’, ‘perception’ and ‘words’. Your eyes – the
sensory organ – aren’t really seeing words, they’re simply encoding different shapes and
patterns of ink on a page. Your brain – the perceptual organ – is transforming those
shapes into a coherent mental representation of words and concepts.
If all this sounds a little peculiar, it’s because from the vantage point of your conscious
experience, it seems as if you’re reading words directly; again, sensation and perception
feel like one single event. If you think of the discussion of brain damage in Chapter 3,
134 4
SENSATION AND PERCEPTION

however, you’ll recall that sometimes a person’s eyes can work just fine, yet the individual
is still ‘blind’ to faces they have seen for many years. Damage to the visual processing
centres in the brain can interfere with the interpretation of information coming from the
eyes: the senses are intact, but perceptual ability is compromised. Sensation and percep-
tion are related – but separate – events.
We all know that sensory events involve vision, hearing, touch, taste and smell. Arguably,
we possess several more senses besides these five. Touch, for example, encompasses distinct
body senses, including sensitivity to pain and temperature, joint position and balance, and
even the state of the gut – perhaps to sense nausea via the autonomic nervous system.
TRANSDUCTION What takes place when Despite the variety of our senses, they all depend on the process of transduction, which
many sensors in the body convert occurs when many sensors in the body convert physical signals from the environment into
physical signals from the environment
into neural signals sent to the central
neural signals sent to the central nervous system.
nervous system. In vision, light reflected from surfaces provides the eyes with information about the
shape, colour and position of objects. In audition, vibrations (from vocal chords or a
guitar string, perhaps) cause changes in air pressure that propagate through space to a
listener’s ears. In touch, the pressure of a surface against the skin signals its shape, texture
and temperature. In taste and smell, molecules dispersed in the air or dissolved in saliva
reveal the identity of substances that we may or may not want to eat. In each case, physi-
cal energy from the world is converted to neural energy inside the central nervous sys-
tem. We’ve already seen that synaesthetes experience a mixing of these perceptions;
however, even during synaesthesia, the processes of transduction that begin those per-
ceptions are the same. Despite ‘hearing colours’, your eyes simply can’t transduce sound
waves, no matter how long you stare at your stereo speakers!

Psychophysics
It’s intriguing to consider the possibility that our basic perceptions of sights or sounds
might differ fundamentally from those of other people. One reason we find synaesthetes
fascinating is because their perceptual experiences are so different from most of ours. But
we won’t get very far in understanding such differences by simply relying on casual self-
reports. As you learned in Chapter 2, to understand a behaviour researchers must first
operationalize it, and that involves finding a reliable way to measure it.
Any type of scientific investigation requires objective measurements. Measuring the
physical energy of a stimulus, such as the colour and brightness of a light, is easy enough:
you can probably buy the necessary instruments online to do that yourself. But how do
you quantify a person’s private, subjective perception of that light? It’s one thing to know
that a torch produces ‘100 candelas’ or gives off ‘8,000 lumens’, but it’s another matter
entirely to measure a person’s psychological experience of that light energy.
The structuralists, led by Wilhelm Wundt and Edward Titchener, tried
GETTY IMAGES/ISTOCKPHOTO THINKSTOCK IMAGES

using introspection to measure perceptual experiences (see Chapter 1). They

failed miserably at this task. After all, you can describe your experience to
another person in words, but that person cannot know directly what you
perceive when you look at a sunset. You both may call the sunset ‘orange’ and
‘beautiful’, but neither of you can directly perceive the other’s experience of
the same event. Evoked memories and emotions intertwine with what you
are hearing, seeing and smelling, making your perception of an event – and
therefore your experience of that event – unique. We return to this fascinat-
ing issue in Chapter 8 when we talk about consciousness and why it is
impossible to know what it would be like to be a different animal such as bat.
Given that perception is different for each of us, how could we ever hope
to measure it? This question was answered in the mid-1800s by German sci-
entist and philosopher Gustav Fechner (1801–87). Fechner was originally
Two people might both describe this trained as a physicist but developed strong interests in philosophy and psy-
sunset as ‘beautiful’ and ‘orange’, but each chology, especially the study of perception. He began conducting informal studies of
individual’s perception of the moment
will be different, based on a cocktail of
visual perception on himself during the 1830s. However, he got a bit carried away with
emotions and memories evoked in the his research and temporarily blinded himself while staring at the sun for a prolonged
instant. time. Fechner’s eyes became so sensitive to light that he had to bandage them before
 SENSATION AND PERCEPTION 4 135

leaving the house, and they bothered him for the rest of his life. Limited in his abilities,
Fechner took on extra work to help support his family, such as translating works from
French to German and even writing much of an encyclopedia of household knowledge
(Watson, 1978). His workload and eye problems resulted in a psychological breakdown
and severe depression, leading him to resign his professorship at the University of Leipzig
and go into seclusion.
Although this was a difficult period in his life, it was of great importance to psychol-
ogy. In his isolation, Fechner was free to think deeply about the issues that interested
him the most, especially how it might be possible to link psychology and physics. His
efforts led him to develop an approach to measuring sensation and perception called
psychophysics – methods that measure the strength of a stimulus and the observer’s sensi- PSYCHOPHYSICS Methods that measure

tivity to that stimulus (Fechner, [1860]1966). In a typical psychophysics experiment, the strength of a stimulus and the
observer’s sensitivity to that stimulus.
researchers ask people to make a simple judgement – whether or not they saw a flash of
light, for example. The psychophysicist then relates the measured stimulus, such as the
brightness of the light flash, to each observer’s yes-or-no response.

Measuring thresholds
Psychophysicists begin the measurement process with a single sensory signal to deter-
mine precisely how much physical energy is required to evoke a sensation in an observer.

Absolute threshold
The simplest quantitative measurement in psychophysics is the absolute threshold – the ABSOLUTE THRESHOLD The minimal

minimal intensity needed to just barely detect a stimulus. A threshold is a boundary. The intensity needed to just barely detect a
stimulus.
doorway that separates the inside from the outside of a house is a threshold, as is the
boundary between two psychological states, for example ‘awareness’ and ‘unawareness’.
In finding the absolute threshold for sensation, the two states in question are sensing and TABLE 4.1 Approximate sensory
not sensing some stimulus. TABLE 4.1 lists the approximate sensory thresholds for each of thresholds
the five senses. Sense Absolute threshold
To measure the absolute threshold for detecting a sound, for example, an observer sits
Vision A candle flame 50 km away on
in a soundproof room wearing headphones linked to a computer. The experimenter pre- a clear, dark night
sents a pure tone (the sort of sound made by striking a tuning fork) using the computer Hearing A clock’s tick 6 m away when
to vary the loudness or the length of time each tone lasts and recording how often the all is quiet
observer reports hearing that tone under each condition. The outcome of such an experi- Touch A fly’s wing touching the cheek
ment is graphed in FIGURE 4.2. Notice from the shape of the curve that the transition from 1 cm away
from not hearing to hearing is gradual rather than abrupt. Investigators typically define Smell A single drop of perfume
the absolute threshold as the loudness required for the listener to say they have heard the diffused through an area
tone on 50% of the trials. equivalent to the volume of six
If we repeat this experiment for many different tones, we can observe and record the rooms

thresholds for tones ranging from very low pitch to very high. It turns out that people Taste A teaspoon of sugar dissolved
in 7.56 L of water
tend to be most sensitive to the range of tones corresponding to human conversation. If
Source: Based on Galanter, 1962
the tone is low enough, such as the lowest note on a pipe organ, most humans cannot
hear it at all, we can only feel it. If the tone is high
enough, we cannot hear it, but dogs and many Absolute threshold for sound FIGURE 4.2 Absolute

100 threshold Some of us

other animals can. are more sensitive than
others, and we may
Difference thresholds even detect sensory
stimulation below our
The absolute threshold is useful for assessing how own absolute threshold.
50% respond,
sensitive we are to faint stimuli, but most every- Percentage ‘Yes, I hear it.’ Absolute threshold
day perception involves detecting differences correct 50 is graphed here as
among stimuli that are well above the absolute the point where the
increasing intensity of
threshold. Most people are pretty adept at notic- the stimulus enables
ing that a sofa is red, but they’re likely to want to an observer to detect it
know if the sofa is redder than the curtains they’re in 50% of the trials. As
considering. Similarly, parents can usually detect its intensity gradually
0 increases, we detect
their own infant’s cry from the cries of other Absolute threshold
the stimulation more
babies, but it’s probably more useful to be able to Physical intensity of stimulus tone frequently.
136 4
SENSATION AND PERCEPTION

differentiate the ‘I’m hungry’ cry from the ‘I’m cranky’ cry from the ‘something is biting
my toes’ cry. In short, the human perceptual system excels at detecting changes in stimula-
tion rather than the simple onset or offset of stimulation.
JUST NOTICEABLE DIFFERENCE ( JND ) The As a way of measuring this difference threshold, Fechner proposed the just noticeable
minimal change in a stimulus that can difference (JND) – the minimal change in a stimulus that can just barely be detected. The
just barely be detected.
JND is not a fixed quantity; rather, it depends on how intense the stimuli being meas-
ured are and the particular sense being measured. Consider measuring the JND for a
bright light. An observer in a dark room is shown a light of fixed intensity, called the
standard (S), next to a comparison light that is slightly brighter or dimmer than the
standard. When S is very dim, observers can see even a very small difference in brightness
between the two lights: the JND is small. But if S is bright, a much larger increment is
needed to detect the difference: the JND is larger.
In fact, the JND can be calculated for each sense. It is roughly proportional to the
magnitude of the standard stimulus. This relationship was first noticed in 1834 by Ernst
Weber, a German physiologist who taught at the University of Leipzig around the time
that Fechner was a student there and probably influenced Fechner’s thinking (Watson,
1978). Fechner applied Weber’s insight directly to psychophysics, resulting in a formal
WEBER ’ S LAW The just noticeable relationship called Weber’s law, which states that the just noticeable difference of a stimu-
difference of a stimulus is a constant lus is a constant proportion despite variations in intensity.
proportion despite variations in intensity.
As an example, the JND for weight is about 2%. If you picked up a 25-g envelope, then
a 50-g envelope, you’d probably notice the difference between them. But if you picked up a
2.27-kg package, then a 2.30-kg package, you’d probably detect no difference at all between
them. In fact, you’d probably need about a 2.60-kg package to detect a JND. When calcu-
lating a difference threshold, it is the proportion between stimuli that is important; the
measured size of the difference, whether in brightness, loudness or weight, is irrelevant.

Signal detection
Measuring absolute and difference thresholds requires a critical assumption: that a
threshold exists. But much of what scientists know about biology suggests that such a
discrete, all-or-none change in the brain is unlikely. Humans don’t suddenly and rapidly
switch between perceiving and not perceiving; in fact, recall that the transition from not
sensing to sensing is gradual (see Figure 4.2). The same physical
© [Link]/NIKADA

stimulus, such as a dim light or a quiet tone, presented on several

different occasions may be perceived by the same person on some
occasions but not on others. Remember, an absolute threshold is
operationalized as perceiving the stimulus 50% of the time, which
means that the other 50% of the time it might go undetected.
Our accurate perception of a sensory stimulus, then, can be
somewhat haphazard. Whether in the psychophysics lab or out in
the world, sensory signals face a lot of competition, or noise, which
refers to all the other stimuli coming from the internal and external
environment. Memories, moods and motives intertwine with what
you are seeing, hearing and smelling at any given time. This internal
‘noise’ competes with your ability to detect a stimulus with perfect,
focused attention. Other sights, sounds and smells in the world at
Cluttered environments such as this music
large also compete for attention; you rarely have the luxury of
festival present our visual system with a attending to just one stimulus apart from everything else. As a consequence of noise, you
challenging signal detection task. may not perceive everything that you sense, and you may even perceive things that you
haven’t sensed.
To see how these mismatches might happen, imagine measuring the electrical activity
of a single neuron sending signals from the eye to the brain. As a dim spot of light is
flashed onto an observer’s eye, the number of subsequent action potentials fluctuates
from one presentation to the next even when the light is exactly the same brightness each
time. Occasionally, the neuron might fire even if no light is presented – a spontaneous
action potential has occurred. Sensory systems are noisy; when the signals are very small,
dim or quiet, the senses provide only a ‘fuzzy’ indicator of the state of the world.
 SENSATION AND PERCEPTION 4 137

This variability among neural responses helps explain why Figure 4.2 shows a gradual
rise in the likelihood of hearing a tone. For a fixed tone intensity, the evoked neural
response varies a little from one presentation to the next. On some presentations, the
auditory neurons’ responses will be a bit greater than average, and the listener will be
more likely to detect the tone. On other presentations, the neural response will, by
chance, be a bit less than average, and the listener will be less likely to detect the tone. On
still other occasions, the neurons might produce spontaneous action potentials, leading
the observer to claim that a tone was heard when none was presented.
Given the variability in neural responses, observers are faced with a decision. If they
say ‘Yes, I heard a tone’ anytime there is any activity in the auditory system, they will
often respond ‘yes’ when no tone is presented. So observers might adopt a more con-
servative response criterion, deciding to say ‘Yes, I heard a tone’ only when the sensory
experience is quite obvious. The problem now is that an observer will often miss fainter
tones that were actually presented. Think of the last time you had a hearing test. You no
doubt missed some of the quiet beeps that were presented, but you also probably said
you heard beeps that weren’t really there.
An approach to psychophysics called signal detection theory holds that the response SIGNAL DETECTION THEORY An

to a stimulus depends on a person’s sensitivity to the stimulus in the presence of noise and on observation that the response to
a stimulus depends on a person’s
a person’s response criterion. That is, observers consider the sensory evidence evoked by sensitivity to the stimulus in the presence
the stimulus and compare it to an internal decision criterion (Green and Swets, 1966; of noise and on a person’s response
Macmillan and Creelman, 2005). If the sensory evidence exceeds the criterion, the criterion.
observer responds by saying ‘Yes, I detected the stimulus’, and if it falls short of the crite-
rion, the observer responds by saying ‘No, I did not detect the stimulus.’
Signal detection theory allows researchers to quantify an observer’s response in the
presence of noise. In a signal detection experiment, a stimulus, such as a dim light, is
randomly presented or not (see FIGURE 4.3). If you’ve ever taken an eye test that checks
your peripheral vision, you have an idea about this kind of setup: lights of varying inten-
sity are flashed at various places in the visual field, and your task is to respond anytime
you see one. Observers in a signal detection experiment must decide whether they saw
the light or not, leading to the four possible outcomes shown in FIGURE 4.3a. If the light
is presented and the observer correctly responds ‘Yes’, the outcome is a hit. If the light is
presented and the observer says ‘No’, the result is a miss. However, if the light is not pre- FIGURE 4.3 Signal detection criteria
sented and the observer nonetheless says it was, a false alarm has occurred. Finally, if the Sensation depends not only on our
sensitivity to stimulation but also on how
light is not presented and the observer responds ‘No’, a correct rejection has occurred: the we make decisions. Of the four possible
observer accurately detected the absence of the stimulus. outcomes on the grid, in (a), we may
Observers can adopt a very liberal response criterion, saying ‘Yes’ at the slightest hint correctly report the presence (a hit) or
of evidence for the stimulus (see FIGURE 4.3b). Notice that this strategy will produce a lot absence (correct rejection) of a stimulus,
fail to detect it (a miss), or say we detect
of hits but also a lot of false alarms. Conversely, adopting a very conservative criterion – it when it’s not there (false alarm). People
saying ‘Yes’ only when the stimulus is clear, strong and unambiguous – should minimize may be equally sensitive to stimulation
the rate of false alarms but increase the proportion of misses (see FIGURE 4.3c). but adopt very different decision criteria.
Signal detection theory is a more sophisticated approach than was used in the early Those who tend to say they detect a signal
produce many false alarms as well as many
days of establishing absolute thresholds. Back then, it might have been assumed that hits (b). Those who tend to say they detect
everyone (or at least a majority of observers) heard a tone or saw a flickering candle no signal minimize false alarms but often
miss the stimulus (c). Decision criteria have
wide application in areas as diverse as drug
trials and dating.
Yes No Yes No Yes No

Light Light Light

Hit Miss 80% 20% 35% 65%
presented presented presented

Light not False Correct Light not Light not

66% 34% 20% 80%
presented alarm rejection presented presented

(a) Possible outcomes on each trial (b) Purely liberal criterion response (c) Purely conservative criterion response
138 4
SENSATION AND PERCEPTION

flame with equal facility. Signal detection theory, in contrast, explicitly takes into
account observers’ response tendencies, such as liberally saying ‘Yes’ or reserving iden-
tifications only for obvious instances of the stimulus. For example, some events are
likely to lead to more variability in responses because they are less discernible, such as
whether or not you briefly saw a camouflaged mouse in the forest. Others are going to
be more obvious and should lead to greater certainty, such as whether or not there is a
stuffed elephant in the corner of the room. Describing these events can be expressed in
D - PRIME ( D ’) A statistic that gives a a measure known as d-prime (d’) – a statistic that gives a relatively pure measure of the
relatively pure measure of the observer’s observer’s sensitivity or ability to detect signals based on the relative proportion of hits
sensitivity or ability to detect signals.
to misses and the group variability in detecting the phenomenon under consideration.
If an event has a high d’ value, people are more certain when it is present or absent. It’s
interesting, then, to learn that the ideas behind signal detection theory were developed
first by none other than Fechner ([1860]1966). It’s not clear why such important ideas
were not grasped by later psychologists who appreciated other aspects of Fechner’s
work, but it’s possible that most of those researchers lacked the mathematical training
required to appreciate Fechner’s insights (Link, 1994). In short, Fechner anticipated
long ago that both the characteristics of the stimulus and the characteristics of the
observer need to be taken into account, producing a better understanding of the per-
ceptual process.
Signal detection theory proposes a way to measure perceptual sensitivity – how effec-
tively the perceptual system represents sensory events – separately from the observer’s
decision-making strategy. Two observers with opposite decision criteria and correspond-
ingly distinct hit rates and false alarm rates may exhibit similar levels of sensitivity. In
other words, even though one person says ‘Yes’ much more often than another, both may
be equally accurate in distinguishing between the presence or absence of a stimulus.
Although the purely conservative and liberal strategies represent two poles on a long
continuum of possible decision criteria, signal detection theory has practical applications
at home, school, work and even while driving.
For example, a radiologist may have to decide whether a mammogram shows that a
patient has breast cancer. The radiologist knows that certain features, such as a mass of a
particular size and shape, are associated with the presence of cancer. But noncancerous
features can have a very similar appearance to cancerous ones. The radiologist may decide
on a strictly liberal criterion and check every possible case of cancer with a biopsy. As
shown in Figure 4.3, this decision strategy minimizes the possibil-
GETTY

ity of missing a true cancer but leads to many false alarms. A strictly
conservative criterion will cut down on false alarms but will miss
some treatable cancers (see Figure 4.3c).
As another example, imagine that the police are on the lookout
for a suspected criminal who they have reason to believe will be at a
crowded football match. Although the police provided a fairly good
description – 1.82-m tall, sandy brown hair, beard, glasses – there
are still thousands of people to scan. Rounding up all men between
1.65 m and 1.95 m would probably produce a hit (the criminal is
caught) but at the expense of an extraordinary number of false
alarms (many innocent people are detained and questioned).
These different types of errors have to be weighed against one
another in setting the decision criterion. Signal detection theory
offers a practical way to choose among criteria that permit decision
Organisms adapt to conditions with
continued exposure. The sharp iciness
makers to take into account the consequences of hits, misses, false alarms and correct rejec-
of the water will fade as this diver keeps tions (McFall and Treat, 1999; Swets et al., 2000). (For an example of a common everyday
swimming. task that can interfere with signal detection, see the real world box.)

Sensory adaptation
When you walk into a bakery, the aroma of freshly baked bread overwhelms you, but after
a few minutes, the smell fades. When you wake up in the middle of the night for a drink
of water, the bathroom light blinds you, but after a few minutes, you no longer squint.
 SENSATION AND PERCEPTION 4 139

These are all examples of sensory adaptation, the observation that sensitivity to pro- SENSORY ADAPTATION Sensitivity to
longed stimulation tends to decline over time as an organism adapts to current conditions. prolonged stimulation tends to decline
Imagine that while you are studying in a quiet room, your neighbour in the flat next door over time as an organism adapts to
turns on the stereo. That gets your attention, but after a few minutes, the sounds fade from current conditions.

your awareness as you continue your studies. But remember that our perceptual systems
emphasize change in responding to sensory events: when the music stops, you notice.

the real world

COURTES Y SARAH SHOMSTEIN

Superior temporal lobe Fusiform gyrus

Multitasking
By one estimate, using a mobile phone while driving makes having
an accident four times more likely (McEvoy et al., 2005). In response
to road safety experts and statistics such as this, some countries are
passing laws that restrict, and sometimes ban, using mobile phones
while driving. You might think that’s a good idea … for everyone
else on the road. But surely you can manage to punch in a number
on a phone, carry on a conversation, or maybe even send a text
message while simultaneously driving in a safe and courteous
manner. Right?
In a word, wrong. The issue here is selective attention, perceiving (a) (b)
only what’s currently relevant to you. Try this. Without moving a
muscle, think about the pressure of your skin against your chair right Shifting attention Participants received fMRI scans as they performed
now. Effortlessly, you shifted your attention to allow a sensory signal tasks that required them to shift their attention between visual and
to enter your awareness. This simple shift shows that your auditory information. (a) When focusing on auditory information, a
perception of the world depends both on what sensory signals are region in the superior (upper) temporal lobe involved in auditory
present and your choice of which signals to attend to and which to processing showed increased activity (yellow/orange). (b) In striking
ignore. Perception is an active, moment-to-moment exploration for contrast, a visual region, the fusiform gyrus, showed decreased activity
when participants focused on auditory information (blue).
relevant or interesting information, not a passive receptacle for
whatever happens to come along.
Talking on a mobile phone while driving demands that you juggle rear-end collision. Whether the phone was handheld or hands-free
two independent sources of sensory input – vision and audition – at made little difference. This suggests that laws requiring drivers to
the same time. Normally, this kind of multitasking works rather well. use hands-free phones may have little effect on reducing accidents.
It’s only when you need to react suddenly that your driving Other researchers have measured brain activity using fMRI while
performance may suffer. Researchers have tested experienced people were shifting attention between visual and auditory
drivers in a highly realistic driving simulator, measuring their information. The strength of visual and auditory brain activity was
response times to brake lights and stop signs while they listened to affected. When attention was directed to audition, activity in visual
the radio or carried on phone conversations about a political issue, areas decreased compared to when attention was directed to vision
among other tasks (Strayer et al., 2003). (Shomstein and Yantis, 2004). It was as if the participants could
These experienced drivers reacted significantly slower during adjust a mental ‘volume knob’ to regulate the flow of incoming
phone conversations than during the other tasks. This is because a information according to which task they were attending to at the
phone conversation requires memory retrieval, deliberation and moment.
planning what to say and often carries an emotional stake in the So how well do we multitask in several thousand kilograms of
conversation topic. Tasks such as listening to the radio require far metal hurtling down the road? Experienced drivers can handle
less attention or none at all. divided attention to a degree, yet most of us have to acknowledge
The tested drivers became so engaged in their conversations that we have had close calls due to driving while distracted. Unless
that their minds no longer seemed to be in the car. Their slower you have two heads with one brain each – one to talk and one to
braking response translated into an increased stopping distance concentrate on driving – you might do well to keep your eyes on
that, depending on the driver’s speed, would have resulted in a the road and not on the phone.

stats facts evidence from real-world examples, such as driving and

answering a telephone, for gender differences, so why would one
expect or indeed look for one in a lab (Strayer et al., 2013)? This
Are women better at multitasking? First ask is the Bayesian approach to accepting or rejecting significant
a Bayesian findings by establishing the prevalence or incidence of the
How often have you heard that women are better than men at phenomenon under consideration in the general population.
multitasking? Before you accept any folk wisdom, one needs to The two sets of scientists disagree about the importance of the
consider the evidence. One recent Swedish study reported that difference found in the lab but it does raise the interesting
men were in fact better than women in an experiment where they question about the relevance of interpreting findings obtained
had to perform two tasks simultaneously (Mäntylä, 2013). under experimental conditions in a limited situation and
However, critics pointed out that there was little statistical extrapolating to the real world.
140 4
SENSATION AND PERCEPTION

The fact that the perceptual system is attuned to change is no coincidence. Remember
that the central nervous system has evolved to respond to the environment and so the
significant events in the world that are most likely going to require some response involve
something changing. It makes sense (literally) to have our senses looking for changes
rather than trying to estimate absolute levels. However, the flip side is that we have to be
able to adapt to changes otherwise we would be in a continual state of alert.
This is why sensory adaptation is a useful process for most organisms. Imagine what
your sensory and perceptual world would be like without it. When you put on your jeans
in the morning, the feeling of rough cloth against your bare skin would be as noticeable
hours later as it was in the first few minutes. The stink of rubbish in your flat when you
first walk in would never dissipate. If you had to constantly be aware of how your tongue
feels while it is resting in your mouth, you’d be driven to distraction. Our perceptual sys-
tems respond more strongly to changes in stimulation rather than to constant stimula-
tion. A stimulus that doesn’t change usually doesn’t require any action; your car probably
emits a certain hum all the time that you’ve got used to. But a change in stimulation often
VISUAL ACUITY The ability to see fine signals a need for action. If your car starts making different kinds of noises, you’re not
detail.
only more likely to notice them, but you’re also more likely to do something about it.

In summary, sensation and perception are critical to survival. that stimulus. Psychophysicists have developed procedures for
Sensation is the simple awareness that results from stimulation of a measuring an observer’s absolute threshold, the smallest intensity
sense organ, whereas perception organizes, identifies and interprets needed to just barely detect a stimulus, and the just noticeable
sensation at the level of the brain. All sensory modalities depend on difference (JND), the smallest change in a stimulus that can just
the process of transduction, which converts physical signals from the barely be detected. Signal detection theory allows researchers to
environment into neural signals carried by sensory neurons into the distinguish between an observer’s perceptual sensitivity to a stimulus
central nervous system. In the 19th century, German researchers and criteria for making decisions about the stimulus. Sensory
developed psychophysics, an approach to studying perception that adaptation occurs because sensitivity to lengthy stimulation tends to
measures the strength of a stimulus and an observer’s sensitivity to decline over time.

Vision: more than meets the eye

You might be proud of your 20/20 vision, even if it is corrected by glasses or contact
lenses. Here, 20/20 refers to a measurement associated with a Snellen chart, named
BRAND X

after Hermann Snellen (1834–1908), the Dutch ophthalmologist who developed it as

a means of assessing visual acuity, the ability to see fine detail, which is the smallest line
of letters that a typical person can read from a distance of 20 ft (the metric equivalent is
6/6 vision where the distance is 6 m). But if you dropped into the birds of prey oph-
thalmologic office, your visual pride would wither. Hawks, eagles, owls and other rap-
tors have much greater visual acuity than humans; in many cases, about eight times
greater, or the equivalent of 20/2 vision. That’s handy if you want to spot a mouse from
a couple of kilometres away, but if you simply need to see where your flatmate left the
box of chocolates, you can probably live with the fact that no one ever calls you ‘Ol’
Eagle Eye’.
Although you won’t win any I Spy contests against a hawk, your sophisticated visual
system has evolved to transduce visual energy in the world into neural signals in the
brain. Humans have sensory receptors in their eyes that respond to wavelengths of light
energy. When we look at people, places and things, patterns of light and colour give us
information about where one surface stops and another begins. The array of light
reflected from those surfaces preserves their shapes and enables us to form a mental rep-
resentation of a scene (Rodieck, 1998). Understanding vision, then, starts with under-
standing light.

The Snellen chart is commonly used Sensing light

to measure visual acuity. Chances
are you’ve seen one yourself on Visible light is simply the portion of the electromagnetic spectrum that we can see, and it
more than one occasion. is an extremely small slice. You can think about light as waves of energy. Like ocean
 SENSATION AND PERCEPTION 4 141

waves, light waves vary in height and in the distance between their peaks, or wavelengths,
as TABLE 4.2 shows.

TABLE 4.2 Properties of light waves

Physical dimension Psychological dimension
Length Hue or what we perceive as colour

Amplitude Brightness

Purity Saturation or richness of colour

There are three properties of light waves, each of which has a physical dimension that
produces a corresponding psychological dimension:
1 The length of a light wave determines its hue, or what humans perceive as colour.
2 The intensity or amplitude of a light wave – how high the peaks are – determines
what we perceive as the brightness of light.
3 Purity is the number of wavelengths that make up the light, which corresponds to
what humans perceive as saturation, or the richness of colours (see FIGURE 4.4).

FIGURE 4.4 Electromagnetic spectrum

The sliver of light waves visible to humans
as a rainbow of colours from violet to red
is bounded on the short end by ultraviolet
400 nm 500 nm 600 nm 700 nm
rays, which honeybees can see, and on
Visible light the long end by infrared waves, on which
night-vision equipment operates. Someone
Shorter waves Longer waves wearing night-vision goggles, for example,
can detect another person’s body heat in
Gamma rays X-ray Ultraviolet Infrared Microwaves Radio waves
complete darkness. Light waves are minute,
but the scale along the bottom of this chart
offers a glimpse of their varying lengths,
10–4 10–3 10–2 10–1 1 10 102 103 104 105 106 107 108 109 1010 1011 1012 1013 1014 measured in nanometres (nm; 1 nm = 1
Wavelength in nanometres (nm) billionth of a metre).

In other words, light doesn’t need a human to have the properties it does: length, ampli-
tude and purity are properties of the light waves themselves. What humans perceive from
those properties are colour, brightness and saturation.
To understand how the properties of waves affect how we sense light, it’s helpful to
understand how our eyes detect light in the first place.
The human eye
FIGURE 4.5 shows the human eye in cross-section. Light that reaches the eyes passes first
through a clear, smooth outer tissue called the cornea, which bends the light wave and
sends it through the pupil, a hole in the coloured part of the eye. This coloured part is the
iris, which is a translucent, doughnut-shaped muscle that controls the size of the pupil
and hence the amount of light that can enter the eye.
When you move from the dim illumination of a cinema into the bright sunshine out-
side, your irises contract, reducing the size of the pupils and the amount of light passing
through them. You may still have to shade your eyes until their light-sensitive cells adapt
to the brighter light level. This process is a type of sensory adaptation called light
adaptation.
Immediately behind the iris, muscles inside the eye control the shape of the lens to
bend the light again and focus it onto the retina – light-sensitive tissue lining the back of RETINA Light-sensitive tissue lining the
the eyeball. The muscles change the shape of the lens to focus objects at different dis- back of the eyeball.
tances, making the lens flatter for objects that are far away or rounder for nearby objects. ACCOMMODATION The process by which
This is called accommodation – the process by which the eye maintains a clear image on the the eye maintains a clear image on the
retina.
retina. FIGURE 4.6 shows how accommodation works.
142 4
SENSATION AND PERCEPTION

FIGURE 4.5 Anatomy of the human eye Muscles to

Light reflected from a surface enters the move eye
eye via the transparent cornea, bending to
pass through the pupil at the centre of the
coloured iris. Behind the iris, the thickness
and shape of the lens adjust to focus Retina
the light on the retina, where the image Lens
appears upside down and backwards. Muscles to
Basically, this is how a camera lens works. Pupil adjust lens
Light-sensitive receptor cells in the retinal
surface, excited or inhibited by spots of Fovea
light, influence the specialized neurons that
convey nerve impulses to the brain’s visual
centres through their axons, which make up
the optic nerve. Iris
Blind
spot
Cornea

Optic nerve
to brain

CONES Photoreceptors that detect

colour, operate under normal daylight
conditions, and allow us to focus on FIGURE 4.6a shows normal vision. However, if your eyeballs are a little too long or a little
fine detail. too short, the lens will not focus images properly on the retina. If the eyeball is too long,
RODS Photoreceptors that become images are focused in front of the retina, leading to nearsightedness (myopia), shown in
active only under low light conditions for FIGURE 4.6b. If the eyeball is too short, images are focused behind the retina, and the result is
night vision.
farsightedness (hyperopia), as shown in FIGURE 4.6c. Glasses (spectacles), contact lenses and
surgical procedures can correct either condition. For example, glasses and contacts both
provide an additional lens to help focus light more appropri-
ately, and procedures such as laser treatment physically reshape
(a) Normal vision the eye’s existing lens. For many of us, the natural ageing pro-
cess means that the mechanisms for bending and focusing
light onto the retina need a helping hand as we get older.
Phototransduction in the retina
The retina is the interface between the world of light outside
the body and the world of vision inside the central nervous
system. Two types of photoreceptor cells in the retina con-
tain light-sensitive pigments that transduce light into neural
impulses. Cones detect colour, operate under normal daylight
conditions, and allow us to focus on fine detail. Rods become
(b) Nearsighted active only under low light conditions for night vision (see
FIGURE 4.7).
Rods are much more sensitive photoreceptors than
cones, but this sensitivity comes at a cost. Because all rods
contain the same photopigment, they provide no informa-
tion about colour and sense only shades of grey. Think about
this the next time you wake up in the middle of the night
and make your way to the bathroom for a drink of water.
Using only the moonlight from the window to light your
way, do you see the room in colour or in shades of grey?
(c) Farsighted
FIGURE 4.6 Accommodation Inside the eye, the lens changes
shape to focus nearby or faraway objects on the retina. (a) People
with normal vision focus the image on the retina at the back of
the eye, both for near and far objects. (b) Nearsighted people
see clearly what’s nearby, but distant objects are blurry because
light from them is focused in front of the retina, a condition called
myopia. (c) Farsighted people have the opposite problem: distant
objects are clear, but those nearby are blurry because their point
of focus falls beyond the surface of the retina, a condition called
hyperopia.
 SENSATION AND PERCEPTION 4 143

STEVE GSCHMEISSNER/SCIENCE PHOTO LIBRARY

Cones

Rods

Fovea

Fovea FIGURE 4.7 Close-up of the retina The surface

of the retina is composed of photoreceptor
cells, the rods and cones, beneath a layer of
transparent neurons, the bipolar and retinal
ganglion cells, connected in sequence. Viewed
Optic close up in this cross-sectional diagram is the
nerve area of greatest visual acuity, the fovea, where
most colour-sensitive cones are concentrated,
Retinal Bipolar Cone Rod allowing us to see fine detail as well as colour.
ganglion cell cell Rods, the predominant photoreceptors
activated in low light conditions, are distributed
Retina everywhere else on the retina.

Rods and cones differ in several other ways as well, most notably in their numbers.
About 120 million rods are distributed more or less evenly around each retina except in
the very centre, the fovea – an area of the retina where vision is the clearest and there are FOVEA An area of the retina where
no rods at all. The absence of rods in the fovea decreases the sharpness of vision in reduced vision is the clearest and there are no
rods at all.
light, but it can be overcome. For example, when amateur astronomers view dim stars
through their telescopes at night, they know to look a little off to the side of the target so
that the image will fall not on the rod-free fovea but on some other part of the retina that
contains many highly sensitive rods.
In contrast to rods, each retina contains only about 6 million cones, which are densely
packed in the fovea and much more sparsely distributed over the rest of the retina, as you
can see in Figure 4.7. The high concentration of cones in the fovea directly affects visual
acuity and explains why objects off to the side, in your peripheral vision, aren’t so clear. The
light reflecting from those peripheral objects has a difficult time landing in the fovea,
making the resulting image less clear. The more fine detail encoded and represented in the
visual system, the clearer the perceived image. The process is analogous to the quality of
photographs taken with a six-megapixel digital camera versus a two-megapixel camera.
Rods and cones also differ in the way their sensitivity changes when the overall light
level changes. Remember that the pupil constricts when you move from dim to bright
© [Link]/ZIUTOGRAF

The full-colour image on the left is what

you’d see when your rods and cones were
fully at work. The greyscale image on the
right is what you’d see if only your rods
were functioning.
144 4
SENSATION AND PERCEPTION

FIGURE 4.8 Blind spot demonstration To illumination. Now consider the

find your blind spot, close your left eye and
reverse. When you enter a dark cinema
stare at the cross with your right eye. Hold
the book 15–30 cm away from your eyes after being outside on a sunny day, your
and move it slowly towards and away from pupil enlarges to let in more light, but
you until the dot disappears. The dot is at first you will be almost blind to the
now in your blind spot and so is not visible.
At this point, the vertical lines may appear
+ seating layout. Gradually, however,
your vision adapts. This form of sen-
as one continuous line because the visual
system fills in the area occupied by the sory adaptation is called dark adapta-
missing dot. To test your left-eye blind spot, tion (Hecht and Mandelbaum, 1938).
turn the book upside down and repeat with
your right eye closed.
Cones adapt to the dark within about
8 minutes but aren’t too sensitive at
low light levels. Rods require about 30 minutes to completely adapt to the dark, but they
provide much better sensitivity in dim light, at the cost of colour vision.
BLIND SPOT An area of the retina that The retina is thick with cells. Among the different neuron types that occupy the reti-
contains neither rods nor cones and
therefore has no mechanism to sense
na’s three distinct layers, the photoreceptor cells (rods and cones) form the innermost
light. layer. The middle layer contains bipolar cells, which collect neural signals from the rods
RECEPTIVE FIELD The region of the and cones and transmit them to the outermost layer of the retina, where neurons called
sensory surface that, when stimulated, retinal ganglion cells (RGCs) organize the signals and send them to the brain. In fact, the
causes a change in the firing rate of that wiring of the eye is back to front, with the photoreceptors furthest away from the incom-
neuron.
ing light that has to pass through the tangle of blood vessels that supply the retina.
The axons and dendrites of photoreceptors and bipolar cells are relatively short (just a
few microns long, or millionths of a metre), whereas the axons of the retinal ganglion
cells span several centimetres. RGCs are the sensory neurons that connect the retina to
various centres within the brain. The bundled RGC axons – about
Receptive field 1.5 million per eye – form the optic nerve, which leaves the eye through
a hole in the retina called the blind spot, which contains neither rods nor
cones and therefore has no mechanism to sense light. Try the demonstra-
tion in FIGURE 4.8 to find the blind spot in each of your own eyes.
Receptive 昀椀elds and lateral inhibition
Each axon in the optic nerve originates in an individual retinal ganglion
Cone cell, as shown at the bottom of FIGURE 4.9. Most RGCs respond to input
patch not from a single retinal cone or rod but from an entire patch of adja-
cent photoreceptors lying side by side, or laterally, in the retina. A par-
ticular RGC will respond to light falling anywhere within that small
To retina patch, which is called its receptive field – the region of the sensory surface
that, when stimulated, causes a change in the firing rate of that neuron.
Although we’ll focus on vision here, the general concept of receptive
fields applies to all sensory systems. For example, the cells that connect
to the touch centres of the brain have receptive fields, which are the part
of the skin that, when stimulated, causes that cell’s response to change in
some way.
Within a receptive field, neighbouring photoreceptors respond to
Bipolar cells stimulation differently: some cells are excited, whereas some are inhib-
ited. These opposing responses interact, which means that the signals
they send through the bipolar cells to the RGC are based on differing
levels of receptor activation, a process called lateral inhibition. Moving
from top to bottom in Figure 4.9, a spot of light that covers any or all of

FIGURE 4.9 Receptive 昀椀eld of a retinal ganglion cell The axon of a retinal
ganglion cell, shown at the bottom of the figure, joins with all other RGC axons
to form the optic nerve. Moving back towards the surface of the retina in this side
view, each RGC connects to a cluster of five or six bipolar cells. The responses
conveyed to the ganglion cell by each bipolar cell depend on the combination of
Retinal ganglion excitatory or inhibitory signals transduced by the larger group of photoreceptors
cell (RGC) Axon
connected to that bipolar cell. The entire grouping, from photoreceptors to RGC,
forms a receptive field, shown at the top of the figure. The RGC responds to a
To optic nerve spot of light falling on any or all of the photoreceptors within its receptive field as
a result of lateral inhibition.
 SENSATION AND PERCEPTION 4 145

(a) On-centre ganglion cell (b) Off-centre ganglion cell

Receptive field Response Receptive field Response

Light on
Light on

Spot in Spot in
centre centre
On Off

Off On

Spot in Spot in
surround surround
On Off

Off On

FIGURE 4.10 RGC receptive 昀椀elds viewed

end-on Imagine that you’re looking
the cones will activate one or more bipolar cells, which in turn causes the ganglion cell to down on the receptive field represented
change the rate at which it sends action potentials. at the top of Figure 4.9. (a) An on-centre
A given RGC responds to a spot of light projected anywhere within a small, roughly cir- ganglion cell increases its firing rate when
the receptive field is stimulated by light
cular patch of retina (Kuffler, 1953). Most receptive fields contain either a central excitatory in the central area, but decreases its firing
zone surrounded by a doughnut-shaped inhibitory zone, which is called an on-centre cell, or rate when the light strikes the surrounding
a central inhibitory zone surrounded by an excitatory zone, which is called an off-centre cell area. Both neural response levels are shown
(see FIGURE 4.10). The doughnut-shaped regions represent patches of retina, as if the top of in the right column. (b) The off-centre
ganglion cell decreases its firing rate when
the diagram in Figure 4.9 were tilted forwards so we could look at the cones end-on. its receptive field is stimulated by light in
Think about the response of an on-centre RGC when its receptive field is stimulated the central area, but increases its firing rate
with spots of light of different sizes (FIGURE 4.10a). A small spot shining on the central when the light strikes the surrounding area.
Both responses are shown at the right.
excitatory zone increases the RGC’s firing rate. When the spot exactly fills the excitatory
zone, it elicits the strongest response, whereas light falling on the surrounding inhibitory
zone elicits the weakest response or none at all. The response of an off-centre cell, shown
in FIGURE 4.10b, is just the opposite. A small spot shining on the central inhibitory zone
elicits a weak response, and a spot shining on the surrounding excitatory zone elicits a
strong response in the RGC.
If a spot of light ‘spills over’ into the inhibitory zone of either receptive field type, the
cell’s response decreases somewhat, and if the entire receptive field is stimulated, excita-
tory and inhibitory activations cancel out due to lateral inhibition and the RGC’s
response will look similar to its response in the dark. Why would the RGC respond the
same way to a uniformly bright field as to a uniformly dark field? The answer is related to
a central notion that we keep coming back to, namely that the central nervous system has
evolved to detect and process difference thresholds. The visual system encodes differences
in brightness or colour. In other words, the RGC is a kind of ‘spot detector’, recording
the relative changes in excitation and inhibition of receptive fields.
Lateral inhibition reveals how the visual system begins to encode the spatial structure
of a scene and not merely the point-by-point light intensity sensed at each location in the
retina. The retina is organized in this way to detect edges – abrupt transitions from light
to dark or vice versa. Edges are of supreme importance in vision. They define the shape of
objects, and anything that highlights such boundaries improves our ability to see an
object’s shape, particularly in low light situations.

Perceiving colour
We thrill to the burst of colours during a fireworks display, ‘ooh’ and ‘aah’ at nature’s pal-
ette during sunset, and marvel at the vibrant hues of a peacock’s tail feathers. Colour
indeed adds zest to the visual world, but it also offers fundamental clues to an object’s
146 4
SENSATION AND PERCEPTION

S-cone M-cone L-cone identity. A black banana or blue lips are colour-coded
(blue) White (green) (red)
419 496 531 559 calls to action – to avoid or sound the alarm, as the case
100 might be.

75 Seeing colour
Maximum
sensitivity (%) Sir Isaac Newton pointed out around 1670 that colour
50
is not something ‘in’ light. In fact, colour is nothing but
25 our perception of light’s wavelengths (see FIGURE 4.11).
We perceive the shortest visible wavelengths as deep
400 450 500 550 600 650
purple. As wavelengths increase, the colour perceived
Wavelength (nanometres) changes gradually and continuously to blue, then green,
FIGURE 4.11 Seeing in colour We perceive
yellow, orange, and, with the longest visible wavelengths,
a spectrum of colour because objects red. This rainbow of hues and accompanying wavelengths is called the visible spectrum,
selectively absorb some wavelengths of illustrated in Figure 4.11.
light and reflect others. Colour perception You’ll recall that all rods contain the same photopigment, which makes them ideal for
corresponds to the summed activity of the
three types of cones. Each type is most
low light vision but bad at distinguishing colours. Cones, by contrast, contain any one of
sensitive to a narrow range of wavelengths three types of pigment. Each cone absorbs light over a range of wavelengths, but its pig-
in the visible spectrum – S-cones process ment type is especially sensitive to visible wavelengths that correspond to red (long-
short (bluish) light, M-cones process medium wavelength), green (medium-wavelength) or blue (short-wavelength) light. Red, green
(greenish) light and L-cones process long
(reddish) light. Rods, represented by the
and blue are the primary colours of light, and the idea that colour perception relies on
white curve, are most sensitive to the three components in the retina dates to the 19th century, when it was first proposed
medium wavelengths of visible light but do by English scientist Thomas Young (1773–1829). Young produced staggering
not contribute to colour perception. accomplishments – he was a practising doctor and a distinguished physicist, and in his
spare time he contributed to solving the mystery of the Rosetta Stone (a tablet that
allowed archaeologists to translate ancient languages). He knew so much about so many
topics that a recent biographer called him ‘the last man who knew everything’ (Robinson,
2006). Happily for psychology, Young had some pretty good ideas about how colour
vision works. But it was von Helmholtz who more fully developed Young’s idea that
colour perception results from different combinations of the three basic elements in the
retina that respond to the wavelengths corresponding to the three primary colours of
light. This insight has several implications and applications.
In one of his early experiments in the darkened room of his study at Trinity College,
Cambridge, Newton discovered that white light is made up of all the visible colours of
the spectrum. He demonstrated that if one holds a prism into a beam of natural daylight
streaming through a shuttered window, it can be decomposed into the full colour spec-
trum projected on the back wall, as illustrated in FIGURE 4.12. This demonstrates that a
white surface is reflecting all the visible wavelengths of light. Lighting designers have
since used this principle to create colours by combing various amounts of primary col-
ours in a process called additive colour mixing.

Window
White shade
sunlight

Spectrum White Hole

wall

Red
Orange
Yellow
Green Glass prism
Blue
FIGURE 4.12 The colour spectrum In his Indigo
early experiments with prisms, Sir Isaac Violet
Newton demonstrated that natural white
light is made up of different wavelengths of
light that correspond to the full spectrum of
visible colours.
 SENSATION AND PERCEPTION 4 147

Centuries before Newton first experimented with light, Renaissance painters in Italy TRICHROMATIC COLOUR REPRESENTATION
had learned that they could re-create any colour found in nature simply by mixing only The pattern of responding across the
three colours: red, blue and yellow. You may have discovered this process for yourself by three types of cones that provides a
mixing paints. Mixing paint works in the opposite way to mixing light as these pigments unique code for each colour.

absorb different colours from the visible spectrum. This subtractive colour mixing works
by removing light from the mix, such as when you combine all the different coloured

PIXTAL
paints in the right proportion, you end up with black. The darker the colour, the less
light it contains, which is why black surfaces reflect no light.
When you perceive colour, then, the cone receptors in your retina encode the wave-
lengths of light reflected from a surface. But colour processing in the human visual sys-
tem occurs in two stages. The first stage – encoding – occurs in the retina, whereas the
second stage – processing – requires the brain (Gegenfurtner and Kiper, 2003).
Trichromatic colour representation in the cones
Light striking the retina causes a specific pattern of response in the three cone types
(Schnapf et al., 1987). One type responds best to short-wavelength (bluish) light, the
second type to medium-wavelength (greenish) light, and the third type to long-
wavelength (reddish) light. Researchers refer to them as S-cones, M-cones and L-cones, Many people inherit conditions in which
respectively (see Figure 4.11). either the red or the green photoreceptors
do not transduce light properly. Such
This trichromatic colour representation means that the pattern of responding across people have difficulty distinguishing hues
the three types of cones provides a unique code for each colour. Researchers can ‘read out’ that to typical individuals appear as red or
the wavelength of the light entering the eye by working backwards from the relative fir- green. Unfortunately, in many countries,
ing rates of the three types of cones. A genetic disorder in which one of the cone types is traffic signals use red and green lights
to indicate whether cars should stop or
missing – and, in some rare cases, two or all three – causes a colour deficiency. Around 4% continue at a junction. Why do drivers with
of the population of Western Europe and the US have a congenital colour deficiency red-green blindness not risk accidents every
( Jacobs, 1997). This trait is sex-linked, affecting men much more often than women. time they approach a junction?
Colour deficiency is often referred to as colour blindness, but in fact, in the most common
form of the condition, people missing only one type of cone can still distinguish many col-
ours, just not as many as someone who has the full complement of three cone types. Like Colour blindness has implications for
synaesthetes, people whose vision is colour deficient often do not realize that they experience which jobs people can hold. One of
colour differently from others. Colour blindness is usually linked to the X chromosome, the first examples of this was a
Swedish train crash in 1875, which
which accounts for why it is found in approximately 8% of males and less than 1% of females. was attributed to the colour blindness
Trichromatic colour representation is well established as the first step of encoding of the driver who could not
colour in the visual system (Abromov and Gordon, 1994). Sensory adaptation helps to distinguish signals. However, there
explain the second step. are still many individuals who are
unaware that they are colour blind. In
Colour-opponent representation into the brain 2010, Robert Law lost his job after 36
years as a driver of freight trains when
Recall that sensory adaptation occurs because our sensitivity to prolonged stimulation it was discovered that he had trouble
tends to decline over time. Just like the rest of your body, cones need an occasional break distinguishing colours. He was
too. Staring too long at one colour fatigues the cones that respond to that colour, pro- responsible for transporting
ducing a form of sensory adaptation called colour afterimage. To demonstrate this effect hazardous goods to a Scottish nuclear
plant.
for yourself, follow these instructions for FIGURE 4.13:
• Stare at the small cross between the two colour patches for about one minute. Try to
keep your eyes as still as possible.
COLOUR - OPPONENT SYSTEM Pairs of
• After a minute, look at the lower cross. You should see a vivid colour aftereffect that
visual neurons that work in opposition.
lasts for a minute or more. Pay particular attention to the colours in the afterimage.
Were you puzzled that the red patch produces a green afterimage and the green patch
produces a red afterimage? This result may seem like nothing more than a curious mind-
bug, but in fact it reveals something important about colour perception. The explanation +
stems from the second stage of colour representation, the colour-opponent system,
where pairs of visual neurons work in opposition – red-sensitive against green-sensitive
cells (as in Figure 4.13) and blue-sensitive against yellow-sensitive cells (Hurvich and

+
FIGURE 4.13 Colour afterimage demonstration Follow the accompanying
instructions in the text, and sensory adaptation will do the rest. When the
afterimage fades, you can get back to reading the chapter.
148 4
SENSATION AND PERCEPTION

Jameson, 1957). How do opponent pairs of four colours make

Left Right
sense if we have just three cone types?
visual field visual field It may be that opponent pairs evolved to enhance colour per-
ception by taking advantage of excitatory and inhibitory stimula-
tion. Red-green cells are excited (they increase their firing rates) in
response to wavelengths corresponding to red and inhibited (they
decrease their firing rates) in response to wavelengths correspond-
ing to green. Blue-yellow cells increase their firing rate in response
to blue wavelengths (excitatory) and decrease their firing rate in
response to yellow wavelengths (inhibitory). The colour pairs are
linked to each other as opposites.
The colour-opponent system explains colour aftereffects. When
you view a colour, let’s say, green, the cones that respond most strongly
Optic Optic nerve to green become fatigued over time. Fatigue leads to an imbalance in
chiasm the inputs to the red-green colour-opponent neurons, beginning with
Lateral
geniculate the RGCs: the weakened signal from the green-responsive cones
nucleus leads to an overall response that emphasizes red. A similar explana-
in the Optic
thalamus tract tion can be made for other colour aftereffects; find a bright blue circle
of colour and get ready to make your flatmate see yellow spots!
Working together, the trichromatic and colour-opponent sys-
tems begin the process of colour perception. S-, M- and L-cones
Superior
colliculus connect to colour-opponent RGCs with excitatory and/or inhibi-
tory connections that produce the colour-opponent response.
Colour-opponent, excitatory-inhibitory processes then continue
Area V1 down the visual pathways to the brain, first to neurons in the thala-
mus and then to the occipital cortex, as mapped in FIGURE 4.14 (de
FIGURE 4.14 Visual pathway from eye Valois et al., 1966).
through brain Objects in the right visual
field stimulate the left half of each retina,
and objects in the left visual field stimulate The visual brain
the right half of each retina. The optic
nerves, one exiting each eye, are formed A great deal of visual processing takes place within the retina itself, including the encod-
by the axons of RGCs emerging from the ing of simple features such as spots of light, edges and colour. More complex aspects of
retina. Just before they enter the brain
at the optic chiasm, about half the nerve
vision, however, require more powerful processing, and that enlists the brain.
fibres from each eye cross. The left half of Streams of action potentials containing information encoded by the retina travel to the
each optic nerve, representing the right brain along the optic nerve. Half of the axons in the optic nerve that leave each eye come
visual field, runs through the brain’s left from RGCs that code information in the right visual field, whereas the other half code
hemisphere via the thalamus, and the
right halves, representing the left visual information in the left visual field. These two nerve bundles link to the left and right
field, travel this route through the right hemispheres of the brain, respectively (see Figure 4.14). The optic nerve travels from each
hemisphere. So, information from the right eye to the lateral geniculate nucleus, located in the thalamus. As you will recall from
visual field ends up in the left hemisphere
Chapter 3, the thalamus receives inputs from all the senses except smell. From there, the
and information from the left visual field
ends up in the right hemisphere. visual signal travels to the back of the brain, to the primary visual cortex, the rear part of
the occipital lobe, which is about the size of a credit card in humans. Here, the informa-
tion is systematically mapped into a representation of the visual scene. The initial process-
AREA V 1 The initial processing region of ing region is known as area V1 and has a topographic visual organization, which means
the primary visual cortex. that adjacent neurons process adjacent portions of the visual field. In fact, the topographical
TOPOGRAPHIC VISUAL ORGANIZATION organization is somewhat distorted as the central portion of the visual field which corre-
Adjacent neurons process adjacent sponds to the input from foveal cells is greatly enlarged in area V1, which is exactly what
portions of the visual field.
you would expect, given the number and densely packed photoreceptors in this part of
the retina. After area V1, there are thought to be about 30–50 additional brain areas spe-
cialized for vision, located mainly in the occipital lobe at the back of the brain and in the
temporal lobes on the sides of the brain (Orban et al., 2004; van Essen et al., 1992).
Neural systems for perceiving shape
One of the most important functions of vision involves perceiving the shapes of objects;
our day-to-day lives would be a mess if we couldn’t distinguish individual shapes from
one another. Imagine not being able to reliably differentiate between a warm doughnut
with glazed icing and a straight stalk of celery and you’ll get the idea; breakfast could
 SENSATION AND PERCEPTION 4 149

become a traumatic experience if you couldn’t distinguish shapes. Perceiving shape

depends on the location and orientation of an object’s edges. It is not surprising, then,
that area V1 is specialized for encoding edge orientation.
As you read in Chapter 3, neurons in the visual cortex selectively respond to bars and
edges in specific orientations in space (Hubel and Wiesel, 1962, 1998). In effect, area V1
contains populations of neurons, each ‘tuned’ to respond to edges oriented at each posi-
tion in the visual field. This means that some neurons fire when an object in a vertical
orientation is perceived, other neurons fire when an object in a horizontal orientation is
perceived, still other neurons fire when objects in a diagonal orientation of 45° are per-
ceived and so on (see FIGURE 4.15). By combining the output from this population of
neurons, the resultant pattern of activation can code for the varying features of objects
and distinguish where a doughnut ends and celery begins.
Pathways for what, where and how
In Chapter 2, you learned how brain researchers have used transcranial magnetic stimula-
tion (TMS) to demonstrate that a person who can recognize what an object is may not be
able to perceive that the object is moving. This observation implies that one brain system
identifies people and things and another tracks their movements, or guides our move-
ments in relation to them. Two functionally distinct pathways, or visual streams, project
from the occipital cortex to visual areas in other parts of the brain (see FIGURE 4.16):
1 The ventral (‘below’) stream travels across the occipital lobe into the lower levels of
the temporal lobes and includes brain areas that represent an object’s shape and iden-
tity, in other words, what it is.
2 The dorsal (‘above’) stream travels up from the occipital lobe to the parietal lobes
(including some of the middle and upper levels of the temporal lobes), connecting
with brain areas that identify the location and motion of an object, in other words,
where it is. Because the dorsal stream allows us to perceive spatial relations, researchers
originally dubbed it the ‘where pathway’ (Ungerleider and Mishkin, 1982). More
recently, neuroscientists have argued that because the dorsal stream is crucial for guid-
ing movements, such as aiming, reaching or tracking with the eyes, the ‘where pathway’
should more appropriately be called the ‘how pathway’ (Milner and Goodale, 1995).
Some of the most dramatic evidence for two distinct visual streams comes from study-
ing the mindbugs that result from brain injury. A patient known as D. F. suffered perma-
nent brain damage following exposure to toxic levels of carbon monoxide (Goodale et al.,

Stimulus Neuron’s responses

Frontal Parietal lobe
lobe

Do Occipital
rsa
l st lobe
rea
m

Area V1
Ventral stream

Temporal
lobe

FIGURE 4.15 Single neuron feature detectors Area V1 contains FIGURE 4.16 Visual streaming One interconnected visual system forms
neurons that respond to specific orientations of edges. Here, the a pathway that courses from the occipital visual regions into the lower
responses of a single neuron are recorded as the bars are viewed at temporal lobe. This ventral pathway enables us to identify what we see.
different orientations. The response rate on the right reveals that this Another interconnected pathway travels from the occipital lobe through
particular neuron fires continuously when the bar is pointing to the right the upper regions of the temporal lobe into the parietal regions. This
at 45°, less often when it is vertical, and not at all when it is pointing to dorsal pathway allows us to locate objects, track their movements and
the left at 45°. move in relation to them.
150 4
SENSATION AND PERCEPTION

1991). A large region of the lateral occipital cortex was destroyed, an area in the ventral
stream that is very active when people recognize objects. D. F.’s ability to recognize
objects by sight was greatly impaired, although her ability to recognize objects by touch
was normal. This suggests that the visual representation of objects, and not D. F.’s memory
for objects, was damaged. D. F.’s brain damage belongs to a category called visual form
agnosia – the inability to recognize objects by sight (Goodale and Milner, 1992, 2004).
Oddly, although D. F. could not recognize objects visually, she could accurately guide
her actions by sight. D. F. was shown a display board with a slot in it, as in FIGURE 4.17. The
researchers could adjust the orientation of the slot. In one version of the task, shown at the
top in the figure, they asked D. F. to report the orientation of the slot by holding her hand
up at the same angle as the slot. D. F. performed very poorly at this task, almost randomly,
suggesting that she did not have a reliable representation of visual orientation.
In another version of the task, shown at the bottom in Figure 4.17, D. F. was asked to
insert a flat block into the slot, as if she were posting a letter into a letter box. Now, she
performed the task almost perfectly. The paradox is that D. F.’s explicit or conscious
understanding of what she was seeing was greatly impaired, but her ability to use this
same information nonconsciously to guide her movements remained intact. When D. F.
was scanned with fMRI, researchers found that she showed normal activation of regions
within the dorsal stream during guided movement ( James et al., 2003).
Other patients with brain damage to the parietal section of the dorsal stream have
FIGURE 4.17 Testing visual form agnosia difficulty using vision to guide their reaching and grasping movements, a condition
When researchers asked patient D. F. to termed optic ataxia (Perenin and Vighetto, 1988). However, these patients’ ventral
orient her hand to match the angle of the
slot in the testing apparatus, as shown at streams are intact, meaning they recognize what objects are. We can conclude from these
the top, she was unable to comply. Asked to two patterns of impairment that the ventral and dorsal visual streams are functionally
insert a card into the slot at various angles, distinct; it is possible to damage one while leaving the other intact.
as shown at the bottom, D. F. accomplished Researchers are starting to examine how the two streams must work together during
the task virtually to perfection.
visual perception in order to integrate ‘what’ and ‘where’. One intriguing possibility is
suggested by recent fMRI research indicating that some regions within the dorsal stream
VISUAL FORM AGNOSIA The inability to are sensitive to properties of an object’s identity, responding differently, for example, to
recognize objects by sight. line drawings of the same object in different sizes or viewed from different vantage points
(Konen and Kastner, 2008). The sensitivity of some regions within the dorsal stream to
aspects of object identity may allow the dorsal and ventral streams to exchange informa-
tion and thus promote integration of ‘what’ and ‘where’ (Farivar, 2009; Konen and
Kastner, 2008).

In summary, light initially passes through several layers in the eye, medium-wavelength (greenish) light, and long-wavelength (reddish)
with the retina linking the world of light outside and the world of light. The overall pattern of response across the three cone types
visual perception inside the central nervous system. Two types of results in a unique code for each colour, known as its trichromatic
photoreceptor cells in the retina transduce light into neural impulses: colour representation. Information encoded by the retina travels to
cones, which operate under normal daylight conditions and sense the brain along the optic nerve, which connects to the lateral
colour, and rods, which are active only under low light conditions for geniculate nucleus in the thalamus and then to the primary visual
night vision. cortex, area V1, in the occipital lobe.
The retina contains several layers, and the outermost consists of Two functionally distinct pathways project from the occipital lobe
retinal ganglion cells (RGCs) that collect and send signals to the brain. to visual areas in other parts of the brain. The ventral stream travels
A particular RGC will respond to light falling anywhere within a small into the lower levels of the temporal lobes and includes brain areas
patch that constitutes its receptive field. Light striking the retina that represent an object’s shape and identity. The dorsal stream goes
causes a specific pattern of response in each of three cone types that from the occipital lobes to the parietal lobes, connecting with brain
are critical to colour perception: short-wavelength (bluish) light, areas that identify the location and motion of an object.

Integrating visual features

As we’ve seen, specialized feature detectors in different parts of the visual system analyse
each of the multiple features of a visible object – orientation, colour, size, shape and so on.
But how are different features combined into single, unified objects? What allows us to
perceive so easily and correctly that the girl with the blue bucket in the photo is wearing a
red top and the girl with the blue top is not holding a blue bucket? Why don’t we see
free-floating patches of blue and red, or even incorrect combinations, such as each girl
 SENSATION AND PERCEPTION 4 151

wearing different tops? These questions refer to what researchers call the
binding problem in perception, which concerns how features are linked
together so that we see unified objects in our visual world rather than free-
floating or miscombined features (Treisman, 1998, 2006). Recent research
indicates that the visual system (in birds at least) may already be set up to
bind shape and colour together from the very beginning (Wood, 2014).
Specifically, newly hatched chicks presented with a virtual object were able
to build an integrated representation of it, binding its colour and shape fea-
tures into integrated colour-shape representations in memory. For example,
they reliably distinguished an object defined by a purple circle and yellow
triangle from an object defined by a purple triangle and yellow circle.

Illusory conjunctions: perceptual mistakes

In everyday life, we correctly combine features into unified objects so auto-
matically and effortlessly that it may be difficult to appreciate that binding is

BRUCE HOOD
ever a problem at all. However, researchers have discovered errors in binding
that reveal important clues about how the process works. One such error is
known as an illusory conjunction – a perceptual mistake where features from
multiple objects are incorrectly combined. In a pioneering study of illusory conjunctions,
BINDING PROBLEM How features are
Treisman and Schmidt (1982) briefly showed study participants visual displays in which
linked together so that we see unified
black digits flanked coloured letters, then instructed them to first report the black digits and objects in our visual world rather than
second to describe the coloured letters. Participants frequently reported illusory conjunc- free-floating or miscombined features.
tions, claiming to have seen, for example, a green A or a purple X instead of the purple A ILLUSORY CONJUNCTION A perceptual
and the green X that had actually been shown (see FIGURE 4.18). These illusory conjunctions mistake where features from multiple
were not just the result of guessing; they occurred more frequently than other kinds of objects are incorrectly combined.
errors, such as reporting a letter or colour that was not present in the display (Figure 4.18). FEATURE INTEGRATION THEORY A theory
that proposes that attention binds
Illusory conjunctions look real to the participants, who were just as confident they had seen
individual features together to comprise
them as they were about the actual coloured letters they perceived correctly. a composite stimulus.
Why do illusory conjunctions occur? Treisman and her colleagues have tried to
explain them by proposing a feature integration theory (Treisman, 1998, 2006;
Treisman and Gelade, 1980; Treisman and Schmidt, 1982), which proposes that attention
binds individual features together to comprise a composite stimulus. From this perspective,
Presented A X
attention provides the ‘glue’ necessary to bind features together, and illusory conjunc-
tions occur when it is difficult for participants to pay full attention to the features that
need to be glued together. For example, in the experiments we just considered, partici-
pants were required to process the digits that flank the coloured letters, thereby reducing Illusory conjunction
attention to the letters and allowing illusory conjunctions to occur. When experimental
conditions are changed so that participants can pay full attention to the coloured letters,
Reported A X
and they are able to correctly bind their features together, illusory conjunctions disap-
pear (Treisman, 1998; Treisman and Schmidt, 1982).
Feature integration theory also helps to explain some striking effects observed when
people search for targets in displays containing many items. When searching through a Misreported letter Misreported colour

display containing green Xs and Os (see FIGURE 4.19), it does not require much focused
attention to spot a target item defined by a unique feature such as a purple X. The purple
A T A X
X seems to simply ‘pop out’ of the display, and people can find it just as quickly when
there are many other nontarget items in the display as when there are only a few nontar-
get items in the display. But when searching through a display of green Xs and purple Os,
FIGURE 4.18 Illusory conjunctions Illusory
a target purple X is no longer defined by a unique feature in relation to the nontarget conjunctions occur when features such as
items; instead, it contains a conjunction of two features, ‘X’ like the green Xs and ‘pur- colour and shape are combined incorrectly.
ple’ like the purple Os. Now, the purple X requires focused attention to pick out, and it For example, when participants are shown
takes more time to find when the display contains many nontargets than when it con- a purple A and green X, they sometimes
report seeing a green A and purple X.
tains few nontargets (Treisman, 1998; Treisman and Gelade, 1980) (Figure 4.19). Other kinds of errors, such as a misreported
letter (reporting ‘T’ when no T was
The role of the parietal lobe presented) or misreported colour (reporting
‘yellow’ when no yellow was presented),
The binding process makes use of feature information processed by structures within the occur rarely, indicating that illusory
ventral visual stream, the ‘what pathway’ (Seymour et al., 2010) (see Figure 4.16, above). conjunctions are not the result of guessing
But because binding involves linking together features processed in distinct parts of the (based on Robertson, 2003).
152 4
SENSATION AND PERCEPTION

FIGURE 4.19 Visual search If you are Feature search display Conjunction search display
asked to try to find the purple X in these
displays, it is easy to find when the purple
X differs from the surrounding items in
either colour or shape (feature search); the
target just pops out at you. It is harder to
find the purple X when it is surrounded by
both purple Os and green Xs (conjunction
search), because focused attention is now
required to spot the target (based on
Robertson, 2003).

ventral stream at a particular spatial location, it also depends critically on the parietal
lobe in the dorsal stream, the ‘where pathway’ (Robertson, 1999). For example, Treisman
and others studied a patient, R. M., who had suffered strokes that destroyed his left and
right parietal lobes. Although many aspects of his visual function were intact, he had
severe problems attending to spatially distinct objects. When presented with stimuli
such as those in Figure 4.18, R. M. perceived an abnormally large number of illusory
conjunctions, even when he was given as long as 10 seconds to look at the displays
(Friedman-Hill et al., 1995; Robertson, 2003). More recent studies of similar patients
suggest that damage to the upper and posterior portions of the parietal lobe is likely to
produce problems with focused attention, resulting in binding problems and increased
illusory conjunctions (Braet and Humphreys, 2009; McCrea et al., 2006). Neuroimaging
studies indicate that these same parietal regions are activated in healthy individuals when
they perform the kind of visual feature binding that patients with parietal lobe damage
are unable to perform (Shafritz et al., 2002), as well as when they search for conjunction
features (Corbetta et al., 1995; Donner et al., 2002).
These findings fit neatly with recent TMS studies in which researchers attempted to
temporarily ‘turn off ’ the posterior parietal lobe while participants performed a feature
binding task involving colours and letters (Braet and Humphreys, 2009). Applying TMS
to the posterior parietal lobe during the task resulted in an increased number of illusory
conjunctions but not in other kinds of perceptual errors. When TMS was applied to the
occipital lobe, it had no effect on illusory conjunctions. Interestingly, the effects of pari-
etal TMS on illusory conjunctions were seen mainly when TMS was applied after pres-
entation of the target item, suggesting that the attentional process supported by the
parietal region serves to ‘lock in’ or consolidate perceived features. These findings help to
refine the original suggestion from feature integration theory that feature binding
depends critically on attentional processes (Treisman, 1998).
Binding and attention in synaesthesia
According to a 2013 report from the
The findings and ideas we’ve just considered turn out to be highly relevant to synaesthe-
British Security Industry Association,
there are up to 5.9 million CCTV sia. We considered examples of synaesthesia at the outset of this chapter, such as consist-
cameras in the UK. That’s one camera ently perceiving particular letters in a particular colour (see Figure 4.1). Some researchers
for every 11 people. When a culprit have characterized synaesthesia as an instance of atypical feature binding. Normal bind-
has been captured clearly on a CCTV ing of colours and letters, for example, is a response to actual features of the external
recording and the identity is in
stimulus, but in synaesthesia, the colour feature is not present in the external stimulus.
question, juries in the UK (and
Australia) can be invited to compare Surprisingly, recent research shows that some of the same processes involved in normal
the recorded images with the feature binding also occur in synaesthesia. fMRI studies of synaesthetic individuals have
defendant in court. In 2009, 95% of revealed that the parietal lobe regions are implicated in the normal binding of colour and
Scotland Yard murder cases used shape and become active during the experience of letter colour synaesthesia (Weiss et al.,
CCTV footage as evidence. However,
2005). Further, applying TMS to these parietal regions interferes with synaesthetic per-
experimental studies show that
approximately 1 in 5 adults ceptions (Esterman et al., 2006; Muggleton et al., 2007). Consistent with the idea that
misidentify the culprit when asked to parietal activity is related to attentional processes needed for binding, other experiments
make this comparison (Davis and have shown that synaesthetic bindings, such as seeing a particular digit in a particular col-
Valentine, 2009). We may have our, depend on attention (Mattingly, 2009; Robertson, 2003; Sagiv et al., 2006). For
specialized areas for processing faces,
instance, when dots are quickly presented to a synaesthetic individual together with digits,
but humans are not infallible when it
comes to matching recordings with for example ‘7’, that induce a synaesthetic perception of green, the synaesthete names the
real people. colour of the dots more quickly when they are green than when they are orange, that is,
when the dot colour matches the colour of the synaesthetic perception. But when
 SENSATION AND PERCEPTION 4 153

synaesthetes are instructed to ignore the numbers, there is little difference in the amount
of time taken to name the colour of the green and orange dots, suggesting that attention is
required to bind the synaesthetic colour to the digit (Robertson, 2003; Sagiv et al., 2006).
Although our perceptual experiences differ substantially from those of synaesthetes, they
rely on the same basic mechanisms of feature binding.

Recognizing objects by sight

Take a quick look at the letters in the accompanying illustration. Even though they’re
quite different from one another, you probably effortlessly recognized them as all being A quick glance and you recognize all these
examples of the letter G. Now consider the same kind of demonstration using your best letters as G, but their varying sizes, shapes,
angles and orientations ought to make this
friend’s face. Your friend might have long hair, but one day she decides to get it cut dra- recognition task difficult. What is it about
matically short. Although your friend now looks strikingly different, you still recognize the process of object recognition that allows
that person with ease. Add glasses, a dye job, producing reddish hair. Maybe your friend us to perform this task effortlessly?
uses coloured contact lenses, or gets piercings to accommodate a nose ring. Any or all of
these elements have the effect of producing a distinctly different looking face, yet just like
the variability in Gs, you are somehow able to extract the underlying features of the face
that allow you to accurately identify your friend.
This thought exercise may seem trivial, but it’s no small perceptual feat. If the visual
system were somehow stumped each time a minor variation occurred in an object being
perceived, the inefficiency of it all would be overwhelming. We’d have to effortfully pro-
cess information just to perceive our friend as the same person from one meeting to
another, not to mention labouring through the process of knowing when a G is really a
G. In general, though, object recognition proceeds fairly smoothly, in large part due to
the operation of the feature detectors we discussed earlier.
How do feature detectors help the visual system get from a spatial array of light hitting
the eye to the accurate perception of an object, such as your friend’s face? Some research-
ers argue for a modular view: that specialized brain areas, or modules, detect and represent
faces or houses or even body parts. Using fMRI to examine visual processing in healthy
young adults, researchers found a subregion in the temporal lobe that responds selectively
to faces compared to just about any other object category, while a nearby area responds
selectively to buildings and landscapes (Kanwisher et al., 1997). This view suggests that
we not only have feature detectors to aid in visual perception but also ‘face detectors’,
‘building detectors’ and possibly other types of neurons specialized for particular types of
object perception (Kanwisher and Yovel, 2006). How this modularization, the process of MODULARIZATION The process of
relatively encapsulated function, comes about is still a matter of controversy. For example, relatively encapsulated function.
some perceptual categories such as faces may have a degree of pre-specified organization
in the brain – nature has wired regions to expect facelike patterns and seek these out. As
we will examine in Chapter 12, some theorists argue that we have built-in perceptual tem-
plates for recognizing faces, which is why newborn babies prefer to look at faces shortly
after birth ( Johnson and Morton, 1991). Others argue that modularization emerges as a
consequence of exposure and expertise (Elman et al., 1997). As you will discover, the
answer appears to be a combination of both built-in propensity and experience.
Psychologists and researchers who argue for a more distributed representation of
object categories challenge the modular view. Researchers have shown that although a
subregion in the temporal lobes does respond more to faces than to any other category,
parts of the brain outside this area may also be involved in face recognition. In this view,
it is the pattern of activity across multiple brain regions that identifies any viewed object,
including faces (Haxby et al., 2001). Each of these views explains some data better than
others, and researchers continue to debate their relative merits.

Representing objects and faces in the brain

Investigations of how the brain responds to complex objects and to faces began in the
1980s with experiments using primates as research subjects. Researchers recorded from
single cells in the temporal lobes of macaque monkeys and found that different neurons
respond selectively to different object shapes (Tanaka, 1996). Other investigators found
neurons that respond best to other monkey faces or to human faces.
154 4
SENSATION AND PERCEPTION

In the mid-1990s, neuroscientists began using fMRI to investigate whether special-

ized neurons like these operate in the human brain. They showed healthy participants
photographs of faces, houses and other object categories, for example shoes, tools or
dogs. During the past decade, fMRI studies have revealed that some brain regions in the
occipital and temporal lobes do respond selectively to specific object categories
(Downing et al., 2006).
Another perspective on this issue is provided by experiments designed to measure
precisely where seizures originate; these experiments have provided insights on how sin-
gle neurons in the human brain respond to objects and faces (Quiroga et al., 2005).
Electrodes were placed on the temporal lobes of people who suffer from epilepsy. Then
the volunteers were shown photographs of faces and objects as the researchers recorded
their neural responses. The researchers found that neurons in the temporal lobe respond
to specific objects viewed from multiple angles and to people wearing different clothing
and facial expressions and photographed from various angles. In some cases, the neurons
also respond to the words for the objects they prefer. For example, a neuron that
responded to photographs of the Sydney Opera House also responded when the words
Sydney Opera were displayed but not when the words Eiffel Tower were displayed
(Quiroga et al., 2005).
PERCEPTUAL CONSTANCY A perceptual Taken together, these experiments demonstrate the principle of perceptual constancy –
principle stating that even as aspects even as aspects of sensory signals change, perception remains consistent. Think back once again
of sensory signals change, perception
remains consistent.
to our discussion of difference thresholds early in this chapter. Our perceptual systems are
sensitive to relative differences in changing stimulation and make allowances for varying
sensory input. This general principle helps explain why you still recognize your friend
despite changes in hair colour or style or the addition of facial jewellery. It’s not as though
your visual perceptual system responds to a change with, ‘Here’s a new and unfamiliar face
to perceive.’ Rather, it’s as though it responds with, ‘Interesting, here’s a deviation from the
way this face usually looks.’ Perception is sensitive to changes in
© [Link]/OLGAECAT

stimuli, but perceptual constancies allow us to notice the differences

in the first place.
Principles of perceptual organization
Before object recognition can even kick in, the visual system must
perform another important task: to group the image regions that
belong together into a representation of an object. The idea that
we tend to perceive a unified, whole object rather than a collec-
tion of separate parts is the foundation of Gestalt psychology,
which you read about in Chapter 1. Gestalt principles character-
ize many aspects of human perception. Among the foremost are
the Gestalt perceptual grouping rules (see FIGURE 4.20), which gov-
ern how the features and regions of things fit together (Koffka,
1935). Here’s a sample:
• Simplicity: A basic rule in science is that the simplest explana-
tion is usually the best. This is the idea behind the Gestalt
grouping rule of Pragnanz, which translates as ‘good form’.
When confronted with two or more possible interpretations of
an object’s shape, the visual system tends to select the simplest
or most likely interpretation (FIGURE 4.20a).
Our visual system allows us to identify • Closure: We tend to fill in missing elements of a visual scene, allowing us to perceive
people as the same individual even when
they change features such as their hairstyle, edges that are separated by gaps as belonging to complete objects (FIGURE 4.20b).
hair colour, facial hair or jewellery. • Continuity: Edges or contours that have the same orientation have what the Gestaltists
called ‘good continuation’, and we tend to group them together perceptually
(FIGURE 4.20c).
• Similarity: Regions that are similar in colour, lightness, shape or texture are perceived
as belonging to the same object (FIGURE 4.20d).
• Proximity: Objects that are close together tend to be grouped together (FIGURE 4.20e).
• Common fate: Elements of a visual image that move together are perceived as parts of
a single moving object (FIGURE 4.20f).
 SENSATION AND PERCEPTION 4 155

FIGURE 4.20 Perceptual grouping rules

Principles first identified by Gestalt
psychologists and now supported by
experimental evidence demonstrate that
the brain is predisposed to impose order on
incoming sensations. One neural strategy
for perception involves responding to
patterns among stimuli and grouping like
patterns together.
(a) Simplicity (b) Closure (c) Continuity

(d) Similarity (e) Proximity (f) Common fate

Separating 昀椀gure from ground

Perceptual grouping is a powerful aid to our ability to recognize objects by sight.
Grouping involves visually separating an object from its surroundings. In Gestalt terms,
this means identifying a figure apart from the (back)ground in which it resides. For
example, the words on this page are perceived as figural: they stand out from the ground
of the sheet of paper on which they’re printed. Similarly, your lecturer is perceived as the
figure against the backdrop of all the other elements in the lecture hall. You certainly can
perceive these elements differently, of course: the words and the paper are all part of a
thing called ‘a page’, and your lecturer and the lecture hall can all be perceived as ‘your
learning environment’. Typically, though, our perceptual systems focus attention on
some objects as distinct from their environments.
Size provides one clue to what’s figure and what’s ground: smaller regions are likely to
be figures, such as tiny letters on a big paper. Movement also helps: your lecturer is (we
hope) a dynamic lecturer, moving around in a static environment. Another critical step
towards object recognition is edge assignment. Given an edge, or boundary, between fig-
ure and ground, which region does that edge belong to? If the edge belongs to the figure,
it helps define the object’s shape, and the background continues behind the edge.
Sometimes, though, it’s not easy to tell which is which.
Danish psychologist Edgar Rubin (1886–1951) capitalized on this ambiguity in devel-
oping a famous illusion called the Rubin vase or, more generally, a reversible figure–ground
relationship. You can view this ‘face–vase’ illusion in FIGURE 4.21 in two ways, either as a
vase on a black background or as a pair of silhouettes facing each other. Your visual system
settles on one or the other interpretation and fluctuates between them every few seconds. FIGURE 4.21 Ambiguous edges Here’s
This happens because the edge that would normally separate figure from ground is really how Rubin’s classic reversible figure–
part of neither: it equally defines the contours of the vase as it does the contours of the ground illusion works. Fix your eyes on the
centre of the image, and your perception
faces. Evidence from fMRIs shows, quite neatly, that when people are seeing the Rubin will alternate between a vase and facing
image as a face, there is greater activity in the face-selective region of the temporal lobe we silhouettes, even as the sensory stimulation
discussed earlier than when they are seeing it as a vase (Hasson et al., 2001). remains constant.

Theories of object recognition

Researchers have proposed two broad explanations of object recognition, one based on
the object as a whole and the other on its parts. Each set of theories has strengths and
weaknesses, making object recognition an active area of study in psychology.
According to image-based object recognition theories, an object you have seen before is
stored in memory as a template – a mental representation that can be directly compared to TEMPLATE A mental representation that

a viewed shape in the retinal image (Tarr and Vuong, 2002). Shape templates are stored can be directly compared to a viewed
shape in the retinal image.
along with name, category and other associations to that object. Your memory compares
156 4
SENSATION AND PERCEPTION

its templates to the current retinal image and selects the template that most closely

BANANASTOCK/PUNCHSTOCK
matches the current image. For example, supermarket scanners use a form of template
matching to identify the barcodes on labels.
Image-based theories are widely accepted, yet they do not explain everything
about object recognition. For one thing, the time it takes to recognize a familiar
object does not depend on its current orientation relative to the object’s standard
orientation: you can quickly recognize that a cup is a cup even when it is tilted on its
side. Correctly matching images to templates suggests that you’d have to have one
template for cups in a normal orientation, another template for cups on their side,
another for cups upside down and so on. This makes for an unwieldy and inefficient
system and therefore one that is unlikely to be effective, yet seeing a cup on its side
rarely perplexes anyone for long. Another limitation is that image-based theories can-
not account for objects you have never seen before. How can you correctly identify
an object by matching it to a template if you don’t have a template because you’ve
never seen the object before? This roundabout reasoning suggests that people would
be mystified when encountering unfamiliar objects, yet actually we make sense of
even unfamiliar objects quite readily.
Parts-based object recognition theories propose instead that the brain deconstructs
viewed objects into a collection of parts (Marr and Nishihara, 1978). One important
parts-based theory contends that objects are stored in memory as structural descriptions:
mental inventories of object parts along with the spatial relations among those parts
(Biederman, 1987). The parts’ inventories act as a sort of ‘alphabet’ of geometric ele-
ments called geons that can be combined to make objects, just as letters are combined to
form words (see FIGURE 4.22). For example, elements such as curved, cylindrical or pointy
might be indexed in an inventory, along with their relations to each other. In parts-based
theories, object recognition constructs an image into its visible parts, notes the spatial
relations among these parts, and then compares this structural description to inventories
Most people can recognize an image of a
face that’s been rotated upside down, and
stored in memory (see Figure 4.22).
they still can even when it’s been altered Like image-based theories, parts-based object recognition has major limitations.
in major ways. In particular, a distortion Most importantly, it allows for object recognition only at the level of categories and not
known as the Thatcher effect or illusion is at the level of the individual object. Parts-based theories offer an explanation for recog-
a phenomenon that often goes unnoticed.
It’s where the face is upside down but the nizing an object such as a face, for example, but are less effective at explaining how you
eyes and the mouth are changed to remain distinguish between your best friend’s face and a stranger’s face.
the right way up (see the top photo). If As you can see, there are strengths and weaknesses of image-based and parts-based
the image is then rotated into the normal
explanations of object recognition. Researchers are developing hybrid theories that
position, you’re in for a shock, as the face
suddenly looks grotesque (see the bottom attempt to exploit the strengths of each approach (Peissig and Tarr, 2007).
photo). The illusion is named after Margaret
Thatcher, former British prime minister, on Theories of face recognition
whose photograph the effect has been most
famously demonstrated. This was originally
Faces are a special category of object. In comparison to most other objects, faces are more
created by Peter Thompson, a professor at alike than different and yet we are capable of recognizing thousands of individual faces.
the University of York, England. In other words, how can we be so good at distinguishing individual faces when all faces

(a) Geons (b) Objects

1 2 5
5
3
1 2

3 4 5
5
FIGURE 4.22 An alphabet of geometric
elements Parts-based theory holds that
5 4
objects such as those shown in (b) are made
up of simpler three-dimensional components 3 3
called geons, shown in (a), much as letters 3
combine to form different words.
 SENSATION AND PERCEPTION 4 157

are so similar? Part of the answer is that we have dedicated brain regions for processing
faces. We have already discussed in Chapter 3 and above, how the fusiform gyrus is par-
ticularly active during face processing. But brain activity does not, in itself, tell us how we
process faces. Also, faces are not just objects, they are individual people and so each face
tells a story, for example who someone is, how old they are, what race they are. Somehow,
all this information is processed when recognizing faces.
To understand how face recognition works, or any complex ability for that matter,
you need a theory or model that considers the separate operations that must be involved.
For example, the Bruce and Young model of face recognition (Bruce and Young, 1986;
Burton and Bruce, 1993) takes into consideration many of the different components
that must be active during face recognition. At the most basic level, you need to work
out that a pattern is a face so you have to be sensitive to features – hopefully two eyes, a
nose and a mouth – in that quantity. If not, why not? Are they facing sideways so only
one eye is visible? Are they wearing an eye patch? Having decided that these features are
present, you then need to look at how they are arranged or configured in terms of spac-
ing. These two operations are part of the structural encoding – how the pattern is repre- STRUCTURAL ENCODING How the pattern
sented. The output from structural encoding feeds into additional processing stages that is represented.
deal with identifying emotional expressions, dynamics of any facial movements, specific
identifying features such as hairy moles and, of course, the face recognition stage. If the
face is someone you recognize, then you are able to identify them as familiar and hope-
fully remember their name. This kind of multistage model explains why people recognize
faces at all, can fail to recognize familiar faces, or, like most of us, recognize faces but not
know their name.

Perceiving depth and size

You’ve probably never appreciated the mundane benefits of knowing where you are at
any given time. If you’ve ever been in an unfamiliar environment, though, the benefits of
knowing what’s around you become readily apparent. Think of being in a house of mir-
rors: is the exit to your left or your right, or are you completely turned around? Imagine
being in a new shopping centre: was H&M the clothing store on the top floor of the west
wing, or was that a Topshop? Are those your friends over there at the food court or just
some people who look like them? Knowing what’s around you is important. Knowing
where each object is located is important too. Whether one object is above, below or to
the left or right of another is first encoded in the retinal image.
Objects in the world are arranged in three dimensions – length, width and depth –
but the retinal image contains only two dimensions, length and width. How does the
brain process a flat, two-dimensional retinal image so that we perceive the depth of an
object and how far away it is? The answer lies in a collection of depth cues that change as
you move through space. Monocular, binocular and motion-based depth cues all help
visual perception (Howard, 2002).
Monocular depth cues
If you had to wear an eye patch for a few hours each day, you might predict that you’d
have a difficult time perceiving things. After all, there must be a good reason for having
two eyes! Actually, unless you are a batsman or a pilot, you can get by quite well with just
one eye. That’s because we can perceive depth with monocular depth cues – aspects of a MONOCULAR DEPTH CUES Aspects of a
scene that yield information about depth when viewed with only one eye. These cues rely on scene that yield information about depth
when viewed with only one eye.
the relationship between distance and size. Even with one eye closed, the retinal image of
an object you’re focused on grows smaller as that object moves farther away and larger as
it moves closer. Our brains routinely use these differences in retinal image size, or relative
size, to perceive distance.
This works particularly well in a monocular depth cue called familiar size. Most
adults, for example, fall within a familiar range of heights (perhaps 1.52–1.98 m), so reti-
nal image size alone is usually a reliable cue to how far away they are. Our visual system
automatically corrects for size differences and attributes them to differences in distance.
FIGURE 4.23 demonstrates how strong this mental correction for familiar size is.
158 4
SENSATION AND PERCEPTION

FIGURE 4.23 Familiar size and relative size

When you view images of people, such as
the women in the left-hand photo, or things
you know well, the object you perceive
as smaller appears farther away. With a
little image manipulation, you can see in
the right-hand photo that the relative size
difference projected on your retinas is far
greater than you perceive. The image of the
woman in the navy blue jumper is exactly
the same size in both photos.

Monocular cues are often called pictorial depth cues because they are present even in
two-dimensional paintings, photographs and videos where the third dimension of depth
is not really there. In addition to relative size and familiar size, there are several more
monocular depth cues, such as:
• Linear perspective, which describes the phenomenon that parallel lines seem to con-
verge as they recede into the distance (FIGURE 4.24a).
• Texture gradient, which arises when you view a more or less uniformly patterned sur-
face because the size of the pattern elements, as well as the distance between them,
grows smaller as the surface recedes from the observer (FIGURE 4.24b).
• Interposition, which occurs when one object partly blocks another (FIGURE 4.24c). You
can infer that the blocking object is closer than the blocked object. However, interpo-
sition by itself cannot provide information about how far apart the two objects are.
• Relative height in the image depends on your field of vision (FIGURE 4.24d). Objects
that are closer to you are lower in your visual field, while faraway objects are higher.
© [Link]/ARPAD BENEDEK

© [Link]/SANDSUN

(a) Linear perspective (b) Texture gradient

© [Link]/LINDA MIRRO
© [Link]/CURTIS J. MORLEY

FIGURE 4.24 Pictorial depth cues Visual

artists rely on a variety of monocular cues to
make their work come to life. You can rely
on cues such as (a) linear perspective, (b)
texture gradient, (c) interposition, and (d)
relative height in an image to infer distance,
depth and position, even if you’re wearing
an eye patch. (c) Interposition (d) Relative height
 SENSATION AND PERCEPTION 4 159

Binocular depth cues

Two eyes are better than one, especially when it comes to depth perception. Binocular
depth cues exist because we have stereoscopic vision: having space between our eyes
means that each eye registers a slightly different view of the world.
Hold your right index finger up about 60 cm in front of your face, close one eye, and
look at your finger. Now alternate, opening and closing each eye rapidly. Your finger
appears to jump back and forth as you do this.
The difference in these two views provides direct and compelling information about
depth. The closer the object you’re looking at, the greater the binocular disparity – the
difference in the retinal images of the two eyes that provides information about depth. Your
brain computes the disparity between the two retinal images to perceive how far away
objects are, as shown in FIGURE 4.25. Viewed from above in the figure, the images of the
more distant square and the closer circle each fall at different points on each retina.
Binocular disparity as a cue to depth perception was first discussed by Sir Charles
Wheatstone in 1838. Wheatstone went on to invent the stereoscope, essentially a holder
for a pair of photographs or drawings taken from two horizontally displaced locations FIGURE 4.25 Binocular disparity We see
the world in three dimensions because our
(Wheatstone did not lack for original ideas – he also invented the accordion and an early eyes are a distance apart and the image of
telegraph and coined the term microphone). When viewed, one by each eye, the pairs of an object falls on the retina of each eye at
images evoked a vivid sense of depth. The View-Master toy is the modern successor to a slightly different place. In this two-object
scene, the images of the square and the
Wheatstone’s invention, and 3-D films are based on this same idea. circle fall on different points of the retina
in each eye. The disparity in the positions
Motion-based depth cues of the circle’s retinal images provides a
When you’re riding in a car, bus or train, the scene changes systematically and continu- compelling cue to depth.
ously. Nearby objects appear to zip by quickly, but faraway objects appear to move slowly
or not at all. This phenomenon is called motion parallax – a depth cue based on the move- BINOCULAR DISPARITY The difference in
ment of the head over time. The speed and direction of the images on your retina depend the retinal images of the two eyes that
provides information about depth.
on where you are looking and on how far away the objects you see are.
MOTION PARALLAX A depth cue based
The depth perception you experience from motion parallax is essentially the same as
on the movement of the head over time.
that provided by binocular disparity. Both involve mentally comparing retinal image
information from multiple viewpoints. In the case of binocular disparity, two slightly
different viewpoints are sampled simultaneously by the two eyes. In motion parallax, the
two viewpoints are sampled in succession, over time.
As you move forward through a scene, the motion cues to depth behave a little differ-
ently. As objects get closer, their image sizes on the retina increase, and their contours
move outwards on the retina, towards the side. Optic flow, the pattern of motion that
accompanies an observer’s forward movement through a scene, is a form of motion par-
allax. At any given point, the scene ahead moves outwards from the point towards which
the observer is moving. This kind of motion parallax is therefore useful for navigation
while walking, driving or landing an aeroplane.
If you have ever watched an old episode of Star Trek, you will recognize as optic flow
the visual effect on the view screen when the spaceship jumps to warp speed. Trails of
starlight out ahead expand outwards from a central point. Back on earth, you can see this
effect when you look through the windscreen as you drive through a snowstorm at night.
As your headlights illuminate the onrushing snowflakes, the flakes at the centre are far-
thest away (near the horizon) and the flakes on the periphery are closest to you.
Illusions of depth and size
We all are vulnerable to illusions, which, as you’ll remember from Chapter 1, are errors of
perception, memory or judgement in which subjective experience differs from objective
reality (Wade, 2005). These mindbugs inspired the Gestalt psychologists, whose contri-
butions continue to influence research on object perception. Recall the Müller-Lyer illu-
sion from Chapter 1 (see Figure 1.2). Even though the horizontal lines in that figure are
exactly the same length, the top horizontal line looks longer than the bottom one. That’s
because you don’t perceive the horizontal lines in isolation: your perception of them is
related to and influenced by the surrounding oblique lines.
The relation between size and distance has been used to create elaborate illusions that
depend on fooling the visual system about how far away objects are. All these illusions
160 4
SENSATION AND PERCEPTION

© PETER ENDIG/DPA/CORBIS
(a)
(b)
FIGURE 4.26 The amazing Ames room
(a) A diagram showing the actual proportions
of the Ames room reveals its secrets. The
sides of the room form a trapezoid with depend on the same principle: when you view two objects that project the same retinal
parallel sides but a back wall that’s way off image size, the object you perceive as farther away will be perceived as larger.
square. The uneven floor makes the room’s
One of the most famous illusions is the Ames room (see FIGURE 4.26), constructed by
height in the far back corner shorter than
the other. Add misleading cues such as US ophthalmologist Adelbert Ames in 1946. The room is trapezoidal in shape rather
specially designed windows and flooring than square: only two sides are parallel (FIGURE 4.26a). A person standing in one corner
and position the room’s occupants in each of an Ames room is physically twice as far away from the viewer as a person standing in
far corner and you’re ready to lure an
unsuspecting observer. (b) Looking into the
the other corner. But when viewed with one eye through the small peephole placed in
Ames room through the viewing port with one wall, the Ames room looks square because the shapes of the windows and the floor-
only one eye, the observer infers a normal ing tiles are carefully crafted to look square from the viewing port (Ittelson, 1952). The
size–distance relationship – that the man visual system perceives the far wall as perpendicular to the line of sight, so that people
and the woman are the same distance away.
But the different image sizes they project
standing at different positions along that wall appear to be at the same distance, and the
on the retina lead the viewer to conclude, viewer’s judgements of their sizes are based directly on retinal image size. As a result, a
based on the monocular cue of familiar size, person standing in the right corner appears to be much larger than a person standing in
that the man is very small and the woman is the left corner (FIGURE 4.26b).
very large.
The moon illusion is another case where incorrectly perceived distance affects the per-
ception of size (Hershenson, 1989). The full moon often appears much larger when it is
near the horizon than when it is directly overhead. In fact, the moon projects identical
retinal image sizes in both positions. What accounts for this compelling mindbug?
When the moon is near the horizon, it appears closer and hence larger because many
features – hills, trees, buildings – intervene between the viewer and
©[Link]/ERKKI MAKKONEN

the moon. Nothing intervenes when the moon is directly overhead,

so it appears smaller.
The importance of illusions
British psychologist Richard Gregory (1966) advocated that illu-
sions were more than just fun, quirky mindbugs but revealed how
the brain makes sense of the world by building models of it. These
models, or hypotheses as Gregory likened them, lead to expecta-
tions that can be violated, creating the experience that things are
not what they seem – the basis for all illusions. There are different
classes of illusions. Some illusions are purely sensory, such as
motion aftereffects where the activity of the movement-sensitive
neurons has been modified. It explains why, after staring out the
back of a moving vehicle, the scene appears to be still moving in the
The moon appears to be much larger at the
horizon than when it is high in the sky. This opposite direction when the vehicle comes to a stop (see the explanation of the waterfall
illusion happens because of visual cues at illusion below). Others are more perceptual, such as the depth and size illusions that
the horizon such as buildings and trees. depend on expectations. Illusions can be found in every modality. Irrespective of what
type of illusion one is dealing with, if it is systematic, then it indicates the way the brain
is interpreting the world by applying models. But illusions are not all in the mind. Some
 SENSATION AND PERCEPTION 4 161

FIGURE 4.27 Brightness illusions Although the

centre of the pattern in image (a) looks much
brighter than the centre of the pattern in (b), they are
identical. You can check by covering up the petals.
FROM LAENG, B. AND ENDESTAD, T. (2012) BRIGHT ILLUSIONS REDUCE THE EYE’S PUPIL.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, USA, 109, 2162–7. REPRODUCED
(a) (b) WITH PERMISSION , NATIONAL ACADEMY OF SCIENCES, USA.

illusions can change the way we behave. For example, we have a pupillary reflex that con-
stricts when we enter a bright environment to reduce the amount of light entering the
eye. Remarkably, Norwegian psychologists found brightness illusions, such as the one
illustrated in FIGURE 4.27, also cause the pupils to constrict even though the brightness is
entirely illusory (Laeng and Endestad, 2012). What should be a purely reflexive motor
behaviour determined by absolute luminance levels turns out to triggered by an illusion.
In a recent follow-up study, researchers measured pupil dilation while participants imag-
ined stimuli of varying brightness such as sunny days, night skies, cloudy skies and dark
rooms. Remarkably, even though these were entirely mental images, participants’ pupils
constricted when thinking about bright environments, indicating that the representations
we hold in memory include action-based adjustments that are normally beyond voluntary
control (Laeng and Sulutvedt, 2014).

hot science
Wishful seeing displayed the ambiguous

PSYCHOPHYSICS, 4(3), MAY 1968, PP. 189–92. WITH PERMISSION OF SPRINGER

GERALD H. FISHER, AMBIGUITY OF FORM: OLD AND NEW, PERCEPTION &
image for one second. If they
Why does the road always look longer and the hill steeper when had been told that the seal was
you are tired? On the other hand, when they are young, fit and the embarrassing outcome,
motivated to cross the finish line, runners perceive obstacles such almost all (97%) said they saw a
as distances and hills as less long and steep (Bhalla and Proffitt, donkey. If they participants had
1999). Thirsty people see glasses of water as larger and closer been hoping to see the seal to
(Balcetis and Dunning, 2010; Veltkamp et al., 2008) and even coins get the less embarrassing
seem larger and more within reach when your poor (Balcetis and outcome, they were less likely
Dunning, 2010; Dubois et al., 2010). It would seem that objects to see the donkey. This study

SCIENCE+BUSINESS MEDIA
that can help people fulfil a wish or achieve a goal seem more tested wishful seeing that was
attainable to those in need. These examples of perceptual generally out of conscious
distortions are known as wishful seeing and reveal that when we awareness, but a second
perceive the world, we apply a filter of goal-directed expectations study confirms the role of
to how we interpret it. But is wishful thinking a conscious bias to rapid processes that are out
just pay attention to things that interest us or is it more of conscious control.
unconscious than this? Due to binocular disparity, each eye produces a slightly
Emily Balcetis, a psychologist at New York University, has different retinal image that is used to generate binocular vision. To
designed some ingenious studies to investigate the role of avoid the problem of ‘double vision’, the brain suppresses the
conscious and unconscious processes operating during wishful image from one eye with the image from the other (usually more
thinking. In one study, participants were asked to report what they dominant) eye in a competitive process known as binocular rivalry.
saw on a computer screen that depicted an ambiguous line Binocular rivalry is so powerful that if two different images were
drawing that could either be seen as a seal or a donkey. presented simultaneously, individuals would only be aware of one.
They were told that if the drawing on the screen was a marine To capitalize on this quirk of the visual brain, Balcetis and her
animal, they would be forced to sing in an embarrassing karaoke colleagues (2012) asked participants to wear red-green goggles
competition in front of judges. If the computer had selected a farm so that red and green images were filtered selectively to each
animal, they would be picked to act as one of the judges. This eye. For example, the letter ‘A’ written in red would be invisible
pairing was swapped over for half the subjects. The computer then to the eye wearing the red lens and the number ‘6’ written in
162 4
SENSATION AND PERCEPTION

green ink would be invisible to the eye with the green lens. This them a forfeit. Although the presentation was extremely short,
enabled the experimenter to present two different images only 300 ms, when they wished for numbers (or letters,
separately to each eye and test which image would be depending on the reward scheme), they saw the target 64% of
suppressed by binocular rivalry. Participants were told that they the time, which was significantly above chance even though the
had to accurately report whether they saw a letter or a number processes were rapid and not under conscious control. Wishful
that was either presented in red or green, which would seeing can be considered a bias that enables individuals to
correspond either to a gain or a loss in a game to earn money. achieve their goals by spurring people on and altering their
They were further instructed not to make errors that would cost perceptions.

I know it’s around here Perceiving motion

somewhere You should now have a good sense of how we see what and where objects are, a process
A 44-year-old man was arrested for made substantially easier when the objects stay in one place. But real life, of course, is full
driving under the influence in of moving targets; objects change position over time. To sense motion, the visual system
Australia’s Northern Territory after he
asked a police officer how to get to
must encode information about both space and time. The simplest case to consider is an
the hard-to-miss Uluru (Ayers Rock, observer who does not move trying to perceive an object that does.
the huge, 348 m-high rock formation As an object moves across an observer’s stationary visual field, it first stimulates one
that appears red in sunlight), which location on the retina, and then a little later it stimulates another location on the retina.
was about 90 m in front of him, Neural circuits in the brain can detect this change in position over time and respond to
illuminated in his headlights.
specific speeds and directions of motion (Emerson et al., 1992). A region in the middle
of the temporal lobe referred to as MT is specialized for the visual perception of motion
Psychology and me (Born and Bradley, 2005; Newsome and Paré, 1988), and brain damage in this area leads
to a deficit in normal motion perception (Zihl et al., 1983).
David Crundall, Professor of Of course, in the real world, rarely are you a stationary observer. As you move around,
Psychology, Nottingham Trent your head and eyes move all the time, and motion perception is not as simple. The
University motion perception system must take into account the position and movement of your
eyes, and ultimately your head and body, in order to perceive the motions of objects
correctly and allow you to approach or avoid them. The brain accomplishes this by
monitoring your eye and head movements and ‘subtracting’ them from the motion in
the retinal image.
Motion perception, like colour perception, operates in part on opponent processes
and is subject to sensory adaptation. A motion aftereffect called the waterfall illusion is
analogous to colour aftereffects. If you stare at the downwards rush of a waterfall for
several seconds, you’ll experience an upwards motion aftereffect when you then look at
stationary objects near the waterfall such as trees or rocks. Probably one of the most
observed motion aftereffects takes place several week nights in the UK as millions of
viewers sit down to watch the spiralling opening credits to the nation’s favourite soap
opera, EastEnders. When the sequence finally comes to a stop, the viewing nation
should all experience their TV screens start to rotate in the opposite direction. What’s
going on here?
The process is similar to seeing green after staring at a patch of red. Motion-sensitive
neurons are connected to motion detector cells in the brain that encode motion in
opposite directions. A sense of motion comes from the difference in the strength of
these two opposing sensors. If one set of motion detector cells is fatigued through adap-
tation to motion in one direction, then the opposing sensor will take over. The net result
is that motion is perceived in the opposite direction. Evidence from fMRIs indicates
David Crundall is an applied cognitive that when people experience the waterfall illusion while viewing a stationary stimulus,
psychologist based at Nottingham Trent there is increased activity in region MT, which plays a key role in motion perception
University in the UK. Visit [Link]. (Tootell et al., 1995).
com/schacter to watch David talking The movement of objects in the world is not the only event that can evoke the
about his current research into traffic and
transport psychology, including the process perception of motion. Neon signs with successively flashing lights can evoke a
of learning to drive, why novice drivers are strong sense of motion, exactly the sort of illusion that inspired Max Wertheimer
more prone to collision, the effect of using who we met in Chapter 1. These illusions where the brain integrates successive
mobile phones while driving, and hazard
perception tests. He also provides advice
images or flashes of light into a single moving object are known as phi phenomenon.
for students on how they can work towards Recall, too, the Gestalt grouping rule of common fate: people perceive a series of
a career in applied psychology. flashing lights as a whole, moving object (see Figure 4.20f ). This perception of
 SENSATION AND PERCEPTION 4 163

movement as a result of alternating signals appearing in rapid succession in different

locations is called apparent motion. APPARENT MOTION The perception of
Video technology and animation depend on apparent motion. A sequence of still movement as a result of alternating
signals appearing in rapid succession in
images sample the continuous motion in the original scene. In the case of films, the sam- different locations.
pling rate (depending on which country you are in) can be 25 or 30 frames per second. A
slower sampling rate would produce a much choppier sense of motion, while a faster
sampling rate would be a waste of resources because we would not perceive the motion as
any smoother than it appears at these rates.

In summary, some regions in the occipital and temporal lobes familiar size and linear perspective; binocular cues, such as retinal
respond selectively to specific object categories, supporting the disparity; and motion-based cues, such as motion parallax, which is
modular view that specialized brain areas represent particular classes based on the movement of the head over time. We experience a
of objects. The principle of perceptual constancy holds that even as sense of motion through the differences in the strengths of output
sensory signals change, perception remains consistent. Gestalt from motion-sensitive neurons. These processes can give rise to
principles of perceptual grouping, such as simplicity, closure and illusions such as apparent motion. Illusions are important because
continuity, govern how the features and regions of things fit they reveal the way the brain operates to build useful models of
together. Depth perception depends on monocular cues, such as the world.

Audition: more than meets the ear

Vision is based on the spatial pattern of light waves on the retina. The sense of hearing,
by contrast, is all about sound waves – changes in air pressure unfolding over time. Plenty
of things produce sound waves: the collision of a tree hitting the forest floor, the impact
of two hands clapping, the vibration of vocal chords during a stirring speech, the reso-
nance of a bass guitar string during a thrash metal concert. Except for synaesthetes who
‘hear colours’, understanding most people’s auditory experience requires understanding
how we transform changes in air pressure into perceived sounds.

Sensing sound
Plucking a guitar string or striking a tuning fork produces a pure tone, a simple sound
wave that first increases air pressure and then creates a relative vacuum. This cycle repeats
hundreds or thousands of times per second as sound waves propagate outwards in all
directions from the source.
Just as there are three dimensions of light waves corresponding to three dimensions of
visual perception, so there are three physical dimensions of a sound wave. Frequency,
amplitude and complexity determine what we hear as the pitch, loudness and quality of
a sound (see TABLE 4.3).

TABLE 4.3 Properties of sound waves

Frequency
Corresponds to our
perception of pitch Low frequency – High frequency –
low-pitched sound high-pitched sound
Amplitude
Corresponds to our
perception of
loudness High amplitude – Low amplitude –
loud sound soft sound
Complexity
Corresponds to our
perception of timbre
Simple – Complex –
pure tone mix of frequencies

• The frequency of the sound wave, or its wavelength, depends on how often the peak in
air pressure passes the ear or a microphone, measured in cycles per second, or hertz
(Hz). Changes in the physical frequency of a sound wave are perceived by humans as PITCH How high or low a sound is.
changes in pitch – how high or low a sound is.
164 4
SENSATION AND PERCEPTION

• The amplitude of a sound wave refers to its height, relative to the threshold for human
LOUDNESS A sound’s intensity. hearing (which is set at zero decibels, or dB). Amplitude corresponds to loudness – a
sound’s intensity. To give you an idea of amplitude and intensity, the rustling of leaves
in a soft breeze is about 20 dB, normal conversation is measured at about 40 dB,
shouting produces 70 dB, a Slayer concert is about 130 dB, and the sound of the space
shuttle taking off a couple of kilometres away registers at 160 dB or more. That’s loud
enough to cause permanent damage to the auditory system and is well above the pain
threshold; in fact, any sounds above 85 dB can be enough to cause hearing damage,
depending on the length and type of exposure.
• Differences in the complexity of sound waves, or their mix of frequencies, correspond
TIMBRE A listener’s experience of sound to timbre – a listener’s experience of sound quality or resonance. Timbre offers us infor-
quality or resonance. mation about the nature of sound. The same note played at the same loudness pro-
duces a perceptually different experience depending on whether it was played on a
flute or a trumpet, a phenomenon due entirely to timbre. Many ‘natural’ sounds also
illustrate the complexity of wavelengths, such as the sound of bees buzzing, the tonali-
ties of speech, or the babbling of a brook. Unlike the purity of a tuning fork’s hum, the
drone of cicadas is a clamour of overlapping sound frequencies.
Of the three dimensions of sound waves, frequency provides most of the information we
need to identify sounds. Amplitude and complexity contribute texture to our auditory
perceptions, but it is frequency that carries their meaning. Sound wave frequencies blend
together to create countless sounds, just as different wavelengths of light blend to create
the richly coloured world we see.
Moreover, sound wave frequency is as important for audition as spatial perception is
for vision. Changes in frequency over time allow us to identify the location of sounds, an
ability that can be crucial to survival, and also allow us to understand speech and appre-
ciate music, skills that are valuable to our cultural survival. The focus in our discussion of
hearing, then, is on how the auditory system encodes and represents sound wave fre-
quency (Kubovy, 1981).

The human ear

How does the auditory system convert sound waves into neural signals? The process is
very different from the visual system, which is not surprising, given that light is a form of
electromagnetic radiation, whereas sound is a physical change in air pressure over time:
different forms of energy suggest different processes of transduction. The human ear is
divided into three distinct parts, as shown in FIGURE 4.28. The outer ear collects sound
waves and funnels them towards the middle ear, which transmits the vibrations to the
inner ear, embedded in the skull, where they are transduced into neural impulses.

FIGURE 4.28 Anatomy Outer ear Middle ear Inner ear

of the human ear The
pinna funnels sound Semicircular
waves into the auditory canals
canal to vibrate the Pinna Ossicles
eardrum at a rate that
Auditory
corresponds to the sound’s nerve
frequency. In the middle to brain
ear, the ossicles pick up
the eardrum vibrations,
amplify them, and pass
them along by vibrating a
membrane at the surface
of the fluid-filled cochlea waves
Sound Cochlea
in the inner ear. Here, fluid
carries the wave energy
to the auditory receptors
Auditory canal Eardrum
that transduce it into
electrochemical activity,
exciting the neurons that
form the auditory nerve,
leading to the brain.
 SENSATION AND PERCEPTION 4 165

The outer ear consists of the visible part on the outside of the head (the pinna), the audi-
tory canal and the eardrum, an airtight flap of skin that vibrates in response to sound waves
gathered by the pinna and channelled into the canal. The middle ear, a tiny, air-filled cham-
ber behind the eardrum, contains the three smallest bones in the body, the ossicles. Named
for their appearance as hammer, anvil and stirrup, the ossicles fit together into a lever that
mechanically transmits and intensifies vibrations from the eardrum to the inner ear.
The inner ear contains the spiral-shaped cochlea (Latin for ‘snail’) – a fluid-filled tube COCHLEA A fluid-filled tube that is the
that is the organ of auditory transduction. The cochlea is divided along its length by the organ of auditory transduction.
basilar membrane – a structure in the inner ear that undulates when vibrations from the BASILAR MEMBRANE A structure in the

ossicles reach the cochlear fluid (see FIGURE 4.29). Its wavelike movement stimulates thou- inner ear that undulates when vibrations
from the ossicles reach the cochlear fluid.
sands of tiny hair cells – specialized auditory receptor neurons embedded in the basilar
HAIR CELLSSpecialized auditory receptor
membrane. The hair cells then release neurotransmitter molecules, initiating a neural neurons embedded in the basilar
signal in the auditory nerve that travels to the brain. You might not want to think that membrane.
the whispered ‘I love you’ that sends chills up your spine got a kick-start from lots of little
hair cells wiggling around, but the mechanics of hearing are what they are!

Cochlear
base Hair cells

FIGURE 4.29 Auditory transduction Inside

the cochlea, shown here as though it were
uncoiling, the basilar membrane undulates
Basilar in response to wave energy in the cochlear
membrane fluid. Waves of differing frequencies ripple
varying locations along the membrane, from
Sound waves at medium frequencies low frequencies at its tip to high frequencies
cause peak bending of the basilar ‘Unrolled’ at the base, and bend the embedded hair
membrane at this point. cochlea Tip of cochlea cell receptors at those locations. The hair cell
motion generates impulses in the auditory
neurons, whose axons form the auditory
Sound wave movement nerve that emerges from the cochlea.

Perceiving pitch
From the inner ear, action potentials in the auditory nerve travel to the thalamus and
ultimately to the contralateral (‘opposite side’; see Chapter 3) hemisphere of the cerebral
cortex. This is called area A1 – a portion of the temporal lobe that contains the primary AREA A 1 A portion of the temporal
auditory cortex (see FIGURE 4.30). For most of us, the auditory areas in the left hemisphere lobe that contains the primary auditory
cortex.
analyse sounds related to language and those in the right hemisphere specialize in rhyth-
mic sounds and music.
Neurons in area A1 respond well to simple tones, and successive auditory areas in the
brain process sounds of increasing complexity (Schreiner et al., 2000). Like area V1 in
the visual cortex where adjacent areas of the visual field trigger adjacent neurons in a
topographic map, area A1 has a tonotopic organization where similar frequencies acti-
vate neurons in adjacent locations (see FIGURE 4.30, inset). A young adult with normal
hearing ideally can detect sounds between about 20 and 20,000 Hz, although the ability
to hear at the upper range decreases with age; an upper limit of about 16,000 Hz may be
more realistic. The human ear is most sensitive to frequencies around 1,000 to 3,500 Hz.
But how is the frequency of a sound wave encoded in a neural signal?
Our ears have evolved two mechanisms to encode sound wave frequency, one for high
frequencies and one for low frequencies. The place code, used mainly for high frequen- PLACE CODE The cochlea encodes
cies, is active when the cochlea encodes different frequencies at different locations along the different frequencies at different
locations along the basilar membrane.
basilar membrane. In a series of experiments carried out from the 1930s to the 1950s,
166 4
SENSATION AND PERCEPTION

Area A1

FIGURE 4.30 Primary auditory cortex Area A1 is

folded into the temporal lobe beneath the lateral Wernicke’s
fissure in each hemisphere. The left hemisphere area
auditory areas govern speech in most people. (inset)
The A1 cortex has a tonotopic organization, with lower
frequencies mapping towards the front of the brain
and higher frequencies towards the back, mirroring the Secondary
organization of the basilar membrane along the cochlea auditory cortex
Temporal
(see Figure 4.29). lobe

Nobel laureate Georg von Békésy (1899–1972) used a microscope to observe the basilar
membrane in the inner ear of cadavers that had been donated for medical research
(Békésy, 1960). Békésy found that the movement of the basilar membrane resembles a
travelling wave (see Figure 4.29). The wave’s shape depends on the frequency of the
stimulating pitch. When the frequency is low, the wide, floppy tip (apex) of the basilar
membrane moves the most, and when the frequency is high, the narrow, stiff end (base)
of the membrane moves the most.
The movement of the basilar membrane causes hair cells to bend, initiating a neural
signal in the auditory nerve. Axons fire the strongest in the hair cells along the area of the
basilar membrane that moves the most, in other words, the place of activation on the
basilar membrane contributes to the perception of sound. The place code works best for
relatively high frequencies that resonate at the basilar membrane’s base and less well for
low frequencies that resonate at the tip, because low frequencies produce a broad travel-
ling wave and therefore an imprecise frequency code.
TEMPORAL CODE The cochlea registers A complementary process handles lower frequencies. A temporal code registers low
low frequencies via the firing rate of frequencies via the firing rate of action potentials entering the auditory nerve. Action poten-
action potentials entering the auditory
nerve.
tials from the hair cells are synchronized in time with the peaks of the incoming sound
waves ( Johnson, 1980). If you imagine the rhythmic boom-boom-boom of a bass drum,
you can probably also imagine the fire-fire-fire of action potentials corresponding to the
beats. This process provides the brain with precise information about pitch that supple-
ments the information provided by the place code.
However, individual neurons can produce action potentials at a maximum rate of
only about 1,000 spikes per second, so the temporal code does not work as well as the
place code for high frequencies. (Imagine if the action potential has to fire in time with
the rat-a-tat-a-tat-a-tat of a snare drum roll.) Like trichromatic representation and oppo-
nent processes in colour processing, the place code and the temporal code work together
to cover the entire range of pitches that people can hear.

Localizing sound sources

Just as the differing positions of our eyes give us stereoscopic vision, the placement of our
ears on opposite sides of the head give us stereophonic hearing. The sound arriving at the
ear closer to the sound source is louder than the sound in the farther ear, mainly because
the listener’s head partially blocks sound energy. This loudness difference decreases as the
sound source moves from a position directly to one side (maximal difference) to straight
ahead (no difference).
 SENSATION AND PERCEPTION 4 167

Another cue to a sound’s location arises from timing: sound waves arrive a little
sooner at the near ear than at the far ear. The timing difference can be as brief as a few
microseconds, but together with the intensity difference, it is sufficient to allow us to
perceive the location of a sound. When the sound source is ambiguous, you may find
yourself turning your head from side to side to localize it. By doing this, you are changing
the relative intensity and timing of sound waves arriving in your ears and collecting bet-
ter information about the likely source of the sound. We usually localize sounds by com-
bining our efforts to hear where the source is coming from with our ability to move our
eyes to the same location. This behaviour is called visual orienting, a behavioural response VISUAL ORIENTING A behavioural

to move the eyes towards a target that might signal a potential prey or predator. Visual response to move the eyes towards a
target.
orienting, which gives animals the ability to hunt or avoid being hunted, is a basic form
of integration between the senses that we discuss next.
Multisensory integration
We began this chapter with the bizarre phenomenon of synaesthesia, but we all have
mappings between the senses – just not the weird ones that synaesthetes experience. Up
until now, we have been considering sight and sound as two independent sensory modal-
ities but much of the world is full of objects that stimulate more than one sense at the
same time. For example, if we drop a priceless Ming dynasty vase, we expect the sound of
shattering antique porcelain to occur at the same time as we see it hit the floor, followed
shortly after by the shriek of the antique dealer. The sight and the sound are synchro-
nized. It is not only sights and sounds that usually go together. For example, we have
certain expectations that if some unseen object feels sharp and jagged in the hand, then it
should also look sharp and jagged when we bring it into view. This kind of perceptual
expectancy is called multisensory integration – the perceptual representation of events MULTISENSORY INTEGRATION The
from more than one sensory modality. perceptual representation of events from
more than one sensory modality.
Multisensory integration reflects the reliable sensory properties of the world. For
example, we expect certain sights and sounds to be integrated, that is, processed simulta-
neously to produce a multisensory experience. If we see a drum being beaten with a stick,
we expect the sound of the beat to come from the same place as the drum and at the same
time. On the other hand, when synchrony between sight and sound is lost, we immedi-
ately notice, as in the case of a badly dubbed foreign film where the movements of the
actors’ mouths and voices are not synchronized (Driver, 1996). You can even create illu-
sions by applying this principle. Ventriloquists typically ‘throw’ their voice by minimiz-
ing the movements of their own mouth and accentuating the movements of the dummy,
leading to the compelling impression that the dummy is talking. In such instances, vision
is the dominant cue that produces an intersensory mindbug known as the ‘ventriloquist
effect’ (Howard and Templeton, 1966).
Multisensory integration is achieved by neurons in the brain that receive input from
more than one sensory modality (Stein and Meredith, 1993). These neurons are sensitive
to the source of the signals in terms of location and timing, as in the case of beating a
drum. When multisensory neurons receive synchronized activity from the visual and
auditory channels, they produce an enhanced response that represents the combined
activity of the different senses (Stein et al., 1989). When they are not synchronized, the
multisensory neurons are not activated. As we read later in Chapter 11, there is evidence
that the basic integration of different senses is present in babies before the first six
months. As we shall read a bit later on in the section on taste, even taste experiences seem
to be sensitive to multisensory integration.

In summary, perceiving sound depends on three physical dimensions through the thalamus to the contralateral primary auditory cortex,
of a sound wave: the frequency of the sound wave determines the area A1, in the temporal lobe. Auditory perception depends on a
pitch; the amplitude determines the loudness; and differences in the place code and a temporal code, which together cover the full range
complexity, or mix, of frequencies determine the sound quality or of pitches that people can hear. Our ability to localize sound sources
timbre. Auditory perception begins in the ear, which consists of an depends critically on the placement of our ears on opposite sides of
outer ear that funnels sound waves towards the middle ear, which in the head. Intersensory integration, where different senses are
turn sends the vibrations to the inner ear, which contains the cochlea. combined to produce a multisensory experience, is achieved by
Action potentials from the inner ear travel along an auditory pathway neurons that receive input from more than one sensory modality.
168 4
SENSATION AND PERCEPTION

The body senses: more than

Skin surface skin deep
Texture and pattern Vision and audition provide information about the
receptors
world at a distance. By responding to light and
Pain receptor sound energy in the environment, these ‘distance’
(free nerve endings) senses allow us to identify and locate the objects
Pressure receptor and people around us. In comparison, the body
senses, also called somatosenses (soma from the
Low-frequency Greek for ‘body’), are up close and personal. Haptic
vibrating receptor
perception results from our active exploration of the
environment by touching and grasping objects with
Duct of sweat gland our hands. We use sensory receptors in our muscles,
tendons and joints as well as a variety of receptors
Fat cells in our skin to get a feel for the world around us (see
High-frequency FIGURE 4.31).
vibrator receptor

FIGURE 4.31 Touch receptors Specialized

sensory neurons form distinct groups of Touch
haptic receptors that detect pressure,
temperature and vibrations against Four types of receptor located under the skin’s surface enable us to sense pressure, tex-
the skin. Touch receptors respond to ture, pattern or vibration against the skin (see Figure 4.31). The receptive fields of these
stimulation within their receptive fields,
and their long axons enter the brain via
specialized cells work together to provide a rich ‘tactile’ (from Latin ‘to touch’) experi-
the spinal or cranial nerves. Pain receptors ence when you explore an object by feeling it or attempt to grasp it. In addition, ther-
populate all body tissues that feel pain. moreceptors, nerve fibres that sense cold and warmth, respond when your skin
They are distributed around bones and temperature changes. All these sensations blend seamlessly together in perception, but
within muscles and internal organs as well
as under the skin’s surface. Both types of
detailed physiological studies have successfully isolated the parts of the touch system
pain receptors – the fibres that transmit ( Johnson, 2002).
immediate, sharp pain sensations quickly Touch begins with the transduction of skin sensations into neural signals. Like cells in
and those that signal slow, dull pain that the retina of each eye, touch receptors have receptive fields with central excitatory zones
lasts and lasts – are free nerve endings.
surrounded by doughnut-shaped inhibitory zones that, when stimulated, cause that cell’s
response to change. The representation of touch in the brain follows a topographic
scheme, much as vision and hearing do. Think back to the homunculus you read about in
HAPTIC PERCEPTION The active
exploration of the environment by
Chapter 3; you’ll recall that different locations on the body project sensory signals to
touching and grasping objects with our different locations in the somatosensory cortex in the parietal lobe.
hands. There are two important principles regarding the neural representation of the body’s
surface. First, there is contralateral organization: the left half of the body is represented
in the right half of the brain and vice versa. Second, just as more of the visual brain is
devoted to foveal vision where acuity is greatest, more of the tactile brain is devoted to
parts of the skin surface that have greater spatial resolution. Regions such as the finger-
tips and lips are very good at discriminating fine spatial detail, whereas areas such as the
lower back are quite poor at that task. These perceptual abilities are a natural conse-
quence of the fact that the fingertips and lips have a relatively dense arrangement of
touch receptors and a large topographical representation in the somatosensory cortex,
while, comparatively, the lower back, hips and calves have a relatively small representa-
tion (Penfield and Rasmussen, 1950).

Pain
Although pain is arguably the least pleasant of sensations, this aspect of touch is
among the most important for survival, as pain indicates damage or potential damage
to the body. The possibility of a life free from pain might seem appealing, but without
the ability to feel pain, we might ignore infections, broken bones or serious burns.
Congenital insensitivity to pain, a rare inherited disorder that specifically impairs
pain perception, is more of a curse than a blessing. Children who experience this dis-
order often mutilate themselves, for example biting into their tongues or gouging their
skin while scratching, and are at increased risk of dying during childhood (Nagasako
et al., 2003).
 SENSATION AND PERCEPTION 4 169

Tissue damage is transduced by pain receptors, the free nerve endings shown in

MATT CARR/GETTY IMAGES ENTERTAINMENT/GETTY IMAGES

Figure 4.31. Researchers have distinguished between fast-acting A-delta fibres,
which transmit the initial sharp pain one might feel right away from a sudden
injury, and slower C fibres, which transmit the longer lasting, duller pain that per-
sists after the initial injury. If you were running barefoot outside and stubbed your
toe against a rock, you would first feel a sudden stinging pain transmitted by A-delta
fibres that would die down quickly, only to be replaced by the throbbing but longer
lasting pain carried by C fibres. Both the A-delta and C fibres are impaired in cases
of congenital insensitivity to pain, which is one reason why the disorder can be
life-threatening.
As you’ll remember from Chapter 3, the pain withdrawal reflex is coordinated by the
spinal cord. No brainpower is required when you touch a hot stove; you retract your
hand almost instantaneously. But neural signals for pain – such as wrenching your elbow
as you brace yourself from falling – travel to two distinct areas in the brain and evoke
two distinct psychological experiences (Treede et al., 1999). One pain pathway sends
signals to the somatosensory cortex, identifying where the pain is occurring and what
sort of pain it is (sharp, burning, dull). The second pain pathway sends signals to the
motivational and emotional centres of the brain, such as the hypothalamus and amyg-
dala, and to the frontal lobe. This is the aspect of pain that is unpleasant and motivates us
to escape from or relieve the pain.
Pain typically feels as if it comes from the site of the tissue damage that caused it. In 2003, Aron Ralston was hiking in a
remote canyon in Utah when tragedy struck.
However, the sensation of pain is generated in your brain, which is why amputees can A 450-kg boulder pinned him in a 0.9-m
still feel phantom pain in their missing limb (Chapter 3). If you burn your finger, you wide space for five days, eventually leaving
will perceive the pain as originating there. When you take an opiate painkiller such as him with no choice but to amputate his own
morphine, it relieves the sensation of pain at the site of injury but is, in fact, operating on arm with a pocketknife. He then applied
a tourniquet, rappelled down the canyon,
neurotransmitter receptors in your brain. We also have pain receptors in many areas and hiked out to safety. These and similar
besides the skin – around bones and within muscles and internal organs as well. When stories illustrate that the extent of an injury
pain originates internally, in a body organ for example, we actually feel it on the surface is not perfectly correlated with the amount
of the body. This kind of referred pain occurs when sensory information from internal of pain felt. Although self-amputation
is undoubtedly excruciating, luckily in
and external areas converge on the same nerve cells in the spinal cord. One common exam- this case it was not debilitating. The
ple is a heart attack: victims often feel pain radiating from the left arm rather than from incident is documented in Ralston’s 2004
inside the chest. autobiography Between a Rock and a Hard
Pain intensity cannot always be predicted solely from the extent of the injury that Place, and is the subject of Danny Boyle’s
2010 film 127 Hours.
causes the pain (Keefe et al., 2005). For example, turf toe sounds like the mildest of
ailments; it is pain at the base of the big toe as a result of bending or pushing off
repeatedly, as a runner might do during a sporting event. This small-sounding injury
in a small area of the body can nonetheless put an athlete out of competition for a REFERRED PAIN Feeling of pain when
month with considerable pain. On the other hand, you’ve probably heard a story or sensory information from internal and
external areas converge on the same
two about someone treading bone-chilling water for hours on end, or dragging their nerve cells in the spinal cord.
shattered legs a kilometre down a country road to seek help after a tractor accident, or
performing some other incredible feat despite searing pain and extensive tissue dam-
age. Pain type and pain intensity show a less-than-perfect correlation, a fact that
intrigues researchers.
Some recent evidence indicates that subjective pain intensity may differ among ethnic
groups (Campbell and Edwards, 2012). A study that examined responses to various
kinds of experimentally induced pain, including heat pain and cold pain, found that
compared to young white adults, young black adults had a lower tolerance for several
kinds of pain and rated the same pain stimuli as more intense and unpleasant (Campbell
et al., 2005).
How do psychologists account for this puzzling variability in pain perception?
According to gate-control theory, signals arriving from pain receptors in the body can be GATE - CONTROL THEORY A theory of
stopped, or gated, by interneurons in the spinal cord via feedback from two directions pain perception based on the idea that
signals arriving from pain receptors in
(Melzack and Wall, 1965). Pain can be gated by the skin receptors, for example by rub- the body can be stopped, or gated,
bing the affected area. Rubbing your stubbed toe activates neurons that ‘close the gate’ to by interneurons in the spinal cord via
stop pain signals from travelling to the brain. Pain can also be gated from the brain by feedback from two directions.
modulating the activity of pain-transmission neurons. This neural feedback is elicited
not by the pain itself, but by activity deep within the thalamus.
170 4
SENSATION AND PERCEPTION

The neural feedback comes from a region in the midbrain called the periaqueductal
grey (PAG). Under extreme conditions, such as high stress, naturally occurring endor-
phins can activate the PAG to send inhibitory signals to neurons in the spinal cord that
then suppress pain signals to the brain, thereby modulating the experience of pain. The
PAG is also activated through the action of opiate drugs, such as morphine.
A different kind of feedback signal can increase the sensation of pain. This system is
activated by events such as infection and learned danger signals. When we are quite ill,
what might otherwise be experienced as mild discomfort can feel quite painful. This
pain facilitation signal presumably evolved to motivate people who are ill to rest and
avoid strenuous activity, allowing their energy to be devoted to healing.
Gate-control theory offers strong evidence that perception is a two-way street. The
senses feed information, such as pain sensations, to the brain, a pattern termed bottom-up
control by perceptual psychologists. The brain processes these sensory data into percep-
tual information at successive levels to support movement, object recognition, and even-
tually more complex cognitive tasks, such as memory and planning. But there is ample
evidence that the brain exerts plenty of control over what we sense as well. Visual illu-
sions and the Gestalt principles of filling in, shaping up and rounding out what isn’t
really there provide some examples. This kind of top-down control also explains the
descending pain pathway initiated in the midbrain.

Body position, movement and balance

©[Link]/TECHNOTR

It may sound odd, but one aspect of sensation and perception is

knowing where parts of your body are at any given moment. It’s not
as though your arm sneaks out of the bedroom window at night to
meet some friends. Your body needs some way to sense its position
in physical space other than moving your eyes to constantly visually
check the location of your limbs. Sensations related to position,
movement and balance depend on stimulation produced within
our bodies. Receptors in the muscles, tendons and joints signal the
position of the body in space, whereas information about balance
and head movement originates in the inner ear.
Sensory receptors provide the information we need to perceive
the position and movement of our limbs, head and body. These
Hitting a ball with a racquet provides receptors also provide feedback about whether we are performing a
feedback as to where your arms and body desired movement correctly and how resistance from held objects may be influencing the
are in space as well as how the resistance
of these objects affects your movement
movement. For example, when you swing a tennis racquet, the weight of it affects how
and balance. Successful athletes have your muscles move your arm as well as the change in sensation when the racquet hits the
particularly well-developed body senses. ball. Muscle, joint and tendon feedback about how your arms actually moved can be
used to improve performance through learning.
VESTIBULAR SYSTEM The three fluid-filled
Maintaining balance depends primarily on the vestibular system – the three fluid-
semicircular canals and adjacent organs filled semicircular canals and adjacent organs located next to the cochlea in each inner ear
located next to the cochlea in each inner (see Figure 4.28 above). The semicircular canals are arranged in three perpendicular ori-
ear. entations and studded with hair cells that detect movement of the fluid when the head
moves or accelerates. This detected motion enables us to maintain our balance, or the
position of our bodies relative to gravity. The movements of the hair cells encode these
somatic sensations (Lackner and DiZio, 2005).
Vision also helps us keep our balance. If you see that you are swaying relative to a
vertical orientation, such as the contours of a room, you move your legs and feet to
keep from falling over. Psychologists have experimented with this visual aspect of
balance by placing people in rooms that can be tilted forwards and backwards
(Bertenthal et al., 1997; Lee and Aronson, 1974). If the room tilts enough – particu-
larly when small children are tested – people will topple over as they try to compen-
sate for what their visual system is telling them. When a mismatch between the
information provided by visual cues and vestibular feedback occurs, motion sickness
can result. Remember this discrepancy the next time you try reading in the back seat
of a moving car!
 SENSATION AND PERCEPTION 4 171

In summary, touch is represented in the brain according to a and another sends signals to the emotional centres of the brain that
topographic scheme in which locations on the body project sensory result in unpleasant feelings that we wish to escape. The experience
signals to locations in the somatosensory cortex, a part of the of pain varies across individuals, which is explained by bottom-up
parietal lobe. The experience of pain depends on signals that and top-down aspects of gate-control theory. Balance and
travel along two distinct pathways. One sends signals to the acceleration depend primarily on the vestibular system but are also
somatosensory cortex to indicate the location and type of pain, influenced by vision.

The chemical senses: adding flavour

Somatosensation is all about physical changes in or on the body. Vision and audition
sense energetic states of the world – light and sound waves – and touch is activated by
physical changes in or on the body surface. The last set of senses we’ll consider share a
chemical basis to combine aspects of distance and proximity. The chemical senses of
olfaction (smell) and gustation (taste) respond to the molecular structure of substances
floating into the nasal cavity as you inhale or dissolving in saliva. Smell and taste combine
to produce the perceptual experience we call flavour.

Smell
Olfaction is the least understood sense and the only one directly connected to the fore-
brain, with pathways into the frontal lobe, amygdala and other forebrain structures
(recall from Chapter 3 that the other senses connect first to the thalamus). This mapping
indicates that smell has a close relationship with areas involved in emotional and social
behaviour. Smell seems to have evolved in animals as a signalling sense for the familiar – a
friendly creature, an edible food, or a sexually receptive mate.
Countless substances release odours into the air, and some of their odourant molecules
make their way into our noses, drifting in on the air we breathe. Situated along the top
of the nasal cavity, shown in FIGURE 4.32, is a mucous membrane called the olfactory

Olfactory nerve
to brain

Olfactory
bulb
Glomerulus

Bone

Olfactory
receptor FIGURE 4.32 Anatomy of smell Along
Olfactory neurons the roof of the nasal cavity, odourant
epithelium (ORNs) molecules dissolve in the mucous
membrane that forms the olfactory
epithelium. Odourants may then bind
ORN to olfactory receptor neurons (ORNs)
binding sites embedded in the epithelium. ORNs
respond to a range of odours and,
once activated, relay action potentials
to their associated glomeruli in the
olfactory bulb, located just beneath
the frontal lobes. The glomeruli
synapse on neurons whose axons form
the olfactory nerve, which projects
Air with odourant molecules
directly into the forebrain.
172 4
SENSATION AND PERCEPTION

OLFACTORY RECEPTOR NEURONS ( ORNS ) epithelium, which contains about 10 million olfactory receptor neurons (ORNs) –
Receptor cells that initiate the sense of receptor cells that initiate the sense of smell. Odourant molecules bind to sites on these
smell. specialized receptors, and if enough bindings occur, the ORNs send action potentials
into the olfactory nerve (Dalton, 2003).
Each olfactory neuron has receptors that bind to some odourants but not to others,
as if the receptor is a lock and the odourant is the key (see Figure 4.32). Groups of ORNs
OLFACTORY BULB A brain structure send their axons from the olfactory epithelium into the olfactory bulb – a brain struc-
located above the nasal cavity beneath ture located above the nasal cavity beneath the frontal lobes. Humans possess about 350
the frontal lobes.
different ORN types that permit us to discriminate among some 10,000 different
odourants through the unique patterns of neural activity each odourant evokes. This
setup is similar to our ability to see a vast range of colours based on only a small number
of retinal cell types or to feel a range of skin sensations based on only a handful of touch
receptor cell types.
The axons of all ORNs of a particular type converge at a site called a glomerulus
within the olfactory bulb; thus, humans have about 350 glomeruli. Different odourant
molecules produce varied patterns of activity (Rubin and Katz, 1999). A given odourant
may strongly activate some glomeruli, moderately activate others, and have little effect
on still others. The genetic basis for this olfactory coding was worked out in large part by
Linda Buck and Richard Axel (1991), who were awarded the Nobel Prize in 2004 for
their efforts.
Some dogs have as many as 100 times more ORNs than humans do, producing a cor-
respondingly sharpened ability to detect and discriminate among millions of odours.
Nevertheless, humans are sensitive to the smells of some substances in extremely small
concentrations. For example, a chemical compound that is added to natural gas to help
detect gas leaks can be sensed at a concentration of just 0.0003 parts per million. By
contrast, acetone (nail polish remover), something most people regard as pungent, can
be detected only if its concentration is 15 parts per million or greater.
The olfactory bulb sends outputs to various centres in the brain, including the parts
that are responsible for controlling basic drives, emotions and memories. This explains
why smells can have immediate, strongly positive or negative effects on us. If the slightest
whiff of an apple pie baking brings back fond memories of childhood or the unexpected
sniff of vomit mentally returns you to a particularly bad party you once attended, you’ve
got the idea. Thankfully, sensory adaptation is at work when it comes to smell, just as it is
with the other senses. Whether the associations are good or bad, after just a few minutes
the smell fades. Smell adaptation makes sense; it allows us to detect new odours that may
require us to act, but after that initial evaluation has occurred, it may be best to reduce
our sensitivity to allow us to detect other smells. Evidence from fMRIs indicates that
experience of a smell can modify odour perception by changing how specific parts of the
brain involved in olfaction respond to that smell (Li et al., 2006).
Smell may also play a role in social behaviour. Humans and other animals can detect
PHEROMONES Biochemical odourants odours from pheromones – biochemical odourants emitted by other members of their spe-
emitted by other members of their cies that can affect the animal’s behaviour or physiology. Parents can distinguish the smell
species that can affect an animal’s
behaviour or physiology.
of their own children from other people’s children. An infant can identify the smell of its
mother’s breast from the smell of other mothers. Even though the recognition of these
smells occurs outside conscious awareness, it nonetheless influences behaviour: parents
pick up their own children rather than strangers’ children, and breast-feeding becomes a
personal connection between mother and child. Pheromones also play a role in repro-
ductive behaviour in insects and several mammalian species, including mice, dogs and
primates (Brennan and Zufall, 2006). Can the same thing be said of human reproductive
behaviour?
Studies of people’s preference for the odours of individuals of the opposite sex have
produced mixed results, with no consistent tendency for people to prefer them over
other pleasant odours. Recent research, however, has provided a link between sexual
orientation and responses to odours that may constitute human pheromones. Researchers
used positron emission tomography (PET) scans to study the brain’s response to two
odours, one related to testosterone, which is produced in men’s sweat, and the other
related to oestrogen, which is found in women’s urine. The testosterone-based odour
 SENSATION AND PERCEPTION 4 173

activated the hypothalamus (a part of the brain that controls sexual behaviour; see
Chapter 3) in heterosexual women but not heterosexual men, whereas the oestrogen-
based odour activated the hypothalamus in heterosexual men but not women. Strikingly,
homosexual men responded to the two chemicals in the same way as women did: the
hypothalamus was activated by the testosterone- but not oestrogen-based odour (Savic
et al., 2005) (see FIGURE 4.33). Other common odours unrelated to sexual arousal were
processed similarly by all three groups. A follow-up study with lesbian women showed
that their responses to the testosterone- and oestrogen-based odours were largely similar
to those of heterosexual men (Berglund et al., 2006). Taken together, the two studies
suggest that some human pheromones are related to sexual orientation.

Heterosexual Homosexual Heterosexual FIGURE 4.33 Smell and social behaviour

women men men In a PET study, heterosexual women,
homosexual men and heterosexual men
were scanned as they were presented
with each of several odours. During the
presentation of a testosterone-based
AND odour (referred to in the figure as AND),
there was significant activation in the
hypothalamus for heterosexual women (left)
and homosexual men (centre) but not for
heterosexual men (right) (Savic et al., 2005).
SAVIC, I., BERGLUND, H. AND LINDSTROM, P. (2005) BRAIN RESPONSE TO
Hypothalamus PUTATIVE PHEROMONES IN HOMOSEXUAL MEN. PROCEEDINGS OF THE
NATIONAL ACADEMY OF SCIENCES, 102, 7356–61. COPYRIGHT (2005) NATIONAL
ACADEMY OF SCIENCES, USA

Other evidence also indicates that pheromones can affect human physiology. Women
who live in close proximity for extended periods, flatmates at university for example,
tend to synchronize menstrual periods. To test the hypothesis that this synchrony might
be mediated by pheromones, a group of women wore cotton pads in their armpits to
collect sweat (McClintock, 1971). The secretions were transferred to the upper lip
(under the nose) of women with whom they had no other contact. This procedure did
indeed cause the menstrual cycles of the pairs to synchronize over time, although the
mechanism remains a mystery. It does not appear to involve any conscious awareness of
the smell: the recipient women in these studies reported that they could not discrimi-
nate between the smell of the pads worn by the donor women from pads that had not
been treated. Nonetheless, the introduction of these pheromones contributed to the
regulation of the women’s bodily states.

Taste
One of the primary responsibilities of the chemical sense of taste is identifying things
that are bad for you, as in ‘poisonous and lethal’. Many poisons are bitter, and we avoid
eating things that nauseate us for good reason, so taste aversions have a clear adaptive
significance. Some aspects of taste perception are genetic, such as an aversion to extreme
bitterness, and some are learned, such as an aversion to a particular food that once caused
nausea (see the sauce Béarnaise phenomenon in Chapter 6). In either case, the direct
contact between a tongue and possible foods allows us to anticipate whether something
will be harmful or palatable.
The tongue is covered with thousands of small bumps, called papillae, which are easily
visible to the naked eye. Within each papilla are hundreds of taste buds – the organ of TASTE BUDS The organ of taste
taste transduction (see FIGURE 4.34). Most of our mouths contain 5,000–10,000 taste transduction.

buds fairly evenly distributed over the tongue, roof of the mouth and upper throat
(Bartoshuk and Beauchamp, 1994; Halpern, 2002). Each taste bud contains 50–100
taste receptor cells. Taste perception fades with age, and, on average, people lose half
their taste receptors by the time they turn 20 (Methven et al., 2012). This may help to
explain why young children seem to be ‘fussy eaters’, since their greater number of taste
buds brings with it a greater range of taste sensations. (For a striking example of extreme
taste sensitivity, see the real world box.)
174 4
SENSATION AND PERCEPTION

(a) (b) (c)

Taste pore
Microvilli
Taste receptor
Taste buds cell
Circumvallate

Papillae
foliate

Papillae
fungiform
Papillae Papilla

Nerve fibres

FIGURE 4.34 A taste bud (a) Taste buds

stud the bumps (papillae) on your tongue,
shown here, as well as the back, sides The human eye contains millions of rods and cones, the human nose contains some
and roof of the mouth. (b) Each taste bud 350 different types of olfactory receptors, but the taste system contains just five main
contains a range of receptor cells that types of taste receptors, corresponding to five primary taste sensations: salt, sour, bitter,
respond to varying chemical components
of foods called tastants. Tastant molecules
sweet and umami (savoury). The first four are quite familiar, but umami may not be. In
dissolve in saliva and stimulate the fact, perception researchers are still debating its existence. The umami receptor was dis-
microvilli that form the tips of the taste covered by Japanese scientists who attributed it to the tastes evoked by foods containing
receptor cells. (c) Each taste bud contacts a high concentration of protein, such as meats and cheeses (Yamaguchi, 1998). If you’re
the branch of a cranial nerve at its base.
a meat eater and you savour the feel of a steak topped with butter or a cheeseburger as it
sits in your mouth, you’ve got an idea of the umami sensation.
Each taste bud contains several types of taste receptor cells whose tips, called micro-
villi, react with tastant molecules in food. Salt taste receptors are most strongly activated
by sodium chloride – table salt. Sour receptor cells respond to acids, such as vinegar or
lime juice. Bitter and sweet taste receptors are more complex. Some 50–80 distinct bind-
ing sites in bitter receptors are activated by an equal number of different bitter-tasting
chemicals. Sweet receptor cells can also be activated by a wide range of substances in
addition to sugars.
Although umami receptor cells are the least well understood, researchers are honing
in on their key features (Chandrashekar et al., 2006). They respond most strongly to
glutamate, an amino acid in many protein-containing foods. Recall from Chapter 3,
glutamate acts as a neurotransmitter; in fact, it’s a major excitatory neurotransmitter. The
food additive monosodium glutamate (MSG), which is often used to flavour Asian foods,
particularly activates umami receptors. Some people develop headaches or allergic reac-
tions after eating foods containing MSG.

the real world nontasters. The remaining 25% of people are supertasters, who
report that such substances, especially dark green vegetables, are
extremely bitter, to the point of being inedible.
Supertasters There’s an evolutionary rationale for this. Aversion to bitter tastes
We all know fussy eaters. Children who don’t like to eat their is present at birth, which is not too surprising, since bitter-tasting
vegetables quickly come to mind. Even some adults shun dark substances are often poisons. However, many foods that taste
green vegetables such as Brussels sprouts, kale and broccoli bitter – those dark green veggies included – are beneficial in
throughout their lifetimes. If you enjoy these vegetables, such taste promoting health and protecting us from disease. Ironically, the very
preferences may seem a little irrational. evolutionary mechanism that may keep you from poisoning yourself
But what if different people actually experience the taste of may also keep you from ingesting some of the healthiest food
broccoli differently, not like the lie your parents told you – ‘It tastes available.
like ice cream’ – but in a qualitatively different way from other folks? There are substantial individual differences in taste preference as
About 50% of people report a mildly bitter taste in caffeine, well. For example, not everyone has taste receptors for bitter
saccharine, certain green vegetables and other substances, while sensations, based on their genetics (Bartoshuk et al., 1994). As
roughly 25% report no bitter taste. Members of the first group are another example, people from Asia, Africa and South America are
called tasters and members of the second group are called more likely to be supertasters than others. Women’s sensitivity to
bitter tastes tends to intensify during pregnancy but diminish after
 SENSATION AND PERCEPTION 4 175

menopause. Children start out as tasters or supertasters, which

© ROYALTY-FREE/CORBIS
could help explain their early tendency towards fussiness in food
preference. However, some children grow up to become nontasters.
Supertasters also experience other flavours differently from
nontasters. Supertasters get more ‘burn’ from chillies and more
creaminess from fats and thickeners in food than others do. They
also experience oral pain more intensely than nontasters (Bartoshuk,
2000). Because supertasters tend to avoid fruits and vegetables that
contain tastes they experience as extremely bitter, they may be at
increased health risk for diseases such as colon cancer. On the other
hand, because they also tend to avoid fatty, creamy foods, they
tend to be thinner and may have decreased risk of cardiovascular
disease (Bartoshuk, 2000).
The difference between the experiences of nontasters and
supertasters can be compared to the difference in experiences
among people with normal colour vision and those with genetic
colour deficiencies, where at least one of the three cone types is
missing. In each case, personal perceptual experiences differ in
ways that may be impossible for others to grasp. A colour-deficient
supertaster, in fact, has probably learned to avoid the grey broccoli.

Fussy eater or just too many taste buds? Our taste perception
declines with age: we lose about half of our taste receptors by
the time we’re 20 years old. That can make childhood either a
time of savoury delight or a sensory overload of taste.

Of course, the variety of taste experiences greatly exceeds the five basic receptors
discussed here. Any food molecules dissolved in saliva evoke specific, combined pat- Red red wine goes to
terns of activity in the five taste receptor types. Although we often think of taste as the my head
primary source for flavour, in fact, taste and smell collaborate to produce this complex Many wine experts are not as expert
as they think. When French graduate
perception. student Frédéric Brochet served two
As any wine connoisseur will attest, the full experience of a wine’s flavour cannot be identical glasses of white wine, with
appreciated without a finely trained sense of smell. Odourants from substances outside one coloured red by food colouring,
your mouth enter the nasal cavity via the nostrils, and odourants in the mouth enter experts described the ‘red’ wine as
through the back of the throat. This is why wine aficionados are taught to pull air in over tasting like normal red wine and
different from the white.
wine held in the mouth: it allows the wine’s odourant molecules to enter the nasal cavity Neuroscientist Chuck Spence of
through this ‘back door’. Oxford University attributes this wine
You can easily demonstrate the contribution of smell to flavour by tasting a few differ- colour effect to the likelihood of the
ent foods while holding your nose, preventing the olfactory system from detecting their influence of past experience and that
odours. If you have a head cold, you probably already know how this turns out. Your red signals the ripeness of fruits in
nature, triggering a shift in taste
favourite spicy curry or zesty pasta probably tastes as bland as can be. perception. Try serving white wine in
But it is not only smell that is important to taste. In the same way that the multisen- a dark glass at your next party. It is
sory integration we discussed earlier reveals expectations about the world, multiple surprisingly hard not to think it tastes
senses can contribute to flavour. For example, wine experts are notoriously easy to fool if like red wine.
you artificially change the colour of the wine (Morrot et al., 2001) (see margin box). The
idea that taste is a multisensory experience is one of the principles that guides the culi-
nary expertise of chef Heston Blumenthal. In addition to inventing weird combinations
such as snail porridge or cauliflower risotto with chocolate, Heston Blumenthal has been
applying neuroscience to enhance the dining experience at his world-famous, three
Michelin star restaurant The Fat Duck in Bray, Berkshire. For example, he uses iPods to
play a recording of waves breaking on the shoreline to heighten the flavours in his ‘Sound
of the sea’ dish of seafood and edible seaweed on a bed of sand-like tapioca. Even humble
crisps taste better if you eat them while listening to loud crunching sounds (Zampini and
Spence, 2005). In other dishes, Blumenthal manipulates colour and textures to shift the
diner’s expectations. According to Jamie Ward (2008), it is precisely this lack of multi-
sensory experience that makes dining at London’s Dans le Noir? such a disappointment.
On his visit to this restaurant, which serves guests their meals in pitch darkness so that
they can focus on flavours alone, Jamie found the food tasted much blander, while much
176 4
SENSATION AND PERCEPTION

of the dining room conversation he heard concerned arguments between a couple about
what they thought they were eating.

In summary, our experience of smell, or olfaction, is associated with which are related to reproductive behaviour and sexual responses in
odourant molecules binding to sites on specialized olfactory several species. Sensations of taste depend on taste buds, which
receptors, which converge at the glomerulus within the olfactory are distributed across the tongue, roof of the mouth and upper
bulb. The olfactory bulb in turn sends signals to parts of the brain throat, and on taste receptors that correspond to the five primary
that control drives, emotions and memories, which helps to explain taste sensations of salt, sour, bitter, sweet and umami. However,
why smells can have immediate and powerful effects on us. Smell is taste is also influenced by other modalities such as vision and
also involved in social behaviour, as illustrated by pheromones, even sound.

psychomythology
You can tell when you are being stared at draw rays coming out of the eyes when asked to explain how we
see (Cottrell and Winner, 1994), and around 60% of adults still
from behind think vision is extramission even after they have been given a
Have you ever had that strange experience that you are being lecture on perception (Gregg et al., 2001). Why is this and why do
watched from behind, turned round and discovered that someone people think they can detect the line of sight?
is staring at you? It is such a common experience that most Vision appears to exit the eyes because we move our eyes to
individuals believe that they can tell when they are being stared at fixate on the world. We adjust our eyes to align them to objects of
from behind, even though there is no way they could detect the interest, which can be under voluntary control, creating the
person doing the staring using normal perception. Edward impression that the observer is the origin and source of the visual
Titchener (1898), one of the early pioneers in our field of study, process. This intuition is supported by cultural endorsement. The
reported that his psychology students consistently believed they Italian Renaissance writer Francesco Petrarca (1304–74) described
could detect unseen gaze. Subsequent studies by Coover (1913) the look of love (innamoramento) as the transfer of particles from
found that 68–86% of psychology students reported having had the lover’s eyes into the eyes of the beloved that then work their
the feeling of being stared at. More recently, Cottrell and way towards the heart. Many cultures believe in the malevolent
colleagues (Cottrell et al., 1996) reported that over 90% of adults Evil Eye, where a curse can be cast by a malevolent glance, and
believed they could feel the unseen stares of another. today’s comic book heroes, such as Cyclops from the X-men and
Aside from paranormal explanations (see Sheldrake, 2003), Superman, have powerful rays that leave the eyes.
the origins, prevalence and persistence of this particular myth Titchner (1898) proposed probably the most compelling
may be attributable to a combination of natural phenomena explanation for why people believe that they can tell when they
working together to substantiate this notion in the minds of are being watched. It is a misinterpreted response bias. Faces are
most people. really important to us. When someone turns to look at us, we
Historically, ancient theories of vision, held most notably by instinctively turn to face them in order to address the potential
Plato and Euclid (c. 300 BCE), proposed that vision worked by interaction. So, if you are in a crowded environment such as a
energy emanating from the eyes. This extramission theory of theatre and sense that you are being stared at, you may turn
vision persisted for centuries until Alhazen (965–1040), the great round to see who is looking at you. In doing so, people behind are
Arab scientist, invented the camera obscura and proved that likely to look at you turning around. This creates the false
vision works by intromission, with light entering the eye. attribution that you turned around because you felt others staring
Nevertheless, this discovery has done little to shift the intuition at you when, in fact, the others are staring because you turn to
that vision works by something leaving the eyes. It is the natural look at them. Simple, isn’t it, and yet we can all feel those eyes
assumption of young children, as evidenced by the way they often burning in to our backs.

where do you stand? These days, marketers advocate a more subtle form of
advertising known as sensory branding (Lindstrom, 2005). The idea
is to exploit all the senses to promote a product or a brand. We’re
Perception and persuasion used to seeing advertisements that feature exciting, provocative or
In the 1950s, cinema owners experimented with a new and sexual images to sell products. In TV commercials, these images are
controversial marketing technique: subliminal advertising. They accompanied by popular music that advertisers hope will evoke an
screened films into which studios had spliced single frames overall mood favourable to the product. The notion is that the sight
containing photographs of popcorn and soda or word images such and sound of exciting things will become associated with what
as I’m thirsty. At normal projection speed, these images were too might be an otherwise drab product.
brief for the audience to perceive consciously, but cinema owners But sensory branding goes beyond sight and sound by enlisting
hoped that projecting the messages would register with the smell, taste and touch as well as vision and hearing. You probably
audience and thus increase concession sales during intermissions. recognize the distinctive aroma of a newly opened can of Play-Doh
However, scientific evidence for this kind of subliminal persuasion or a fresh box of Crayola crayons. Their scents are unmistakable, but
has been mixed at best. they’re also somewhat inadvertent. Play-Doh was first sold in 1956
 SENSATION AND PERCEPTION 4 177

and Crayola crayons appeared in 1903, long before there was any notice when the star of a film drinks a can of a well-known beverage
thought given to marketing as a total sensory experience. or drives a particular vehicle model in a car chase? Although viewers
Sensory branding is a much more intentional approach to may not notice or even be aware of the product, advertisers believe
marketing. That new-car smell you anticipate while you take a test that product placement benefits their bottom lines.
drive? Actually, it’s a manufactured fragrance sprayed into the car, Is there any harm in marketing that bombards the senses or
carefully tested to evoke positive feelings among potential buyers. even sneaks through to perception undetected? Advertising is a
Bang & Olufsen, the Danish high-end stereo manufacturer, business, and like any business it is fuelled by innovation in search
carefully designed its remote control units to have a certain of a profit. Perhaps these recent trends are simply the next clever
distinctive ‘feel’ in a user’s hand. Singapore Airlines, which has step to get potential buyers to pay attention to a product
consistently been rated ‘the world’s best airline’, has actually message. On the other hand, is there a point when ‘enough is
patented the smell of its aeroplane cabins (it’s called Stefan enough’? Do you want to live in a world where every sensory
Floridian Waters). event is trademarked, patented or test-marketed before reaching
Another form of advertising that has grown dramatically in recent your perceptual system? Does the phrase, ‘Today’s sunset was
years is product placement: companies pay to have their products brought to you by the makers of …’ cause you alarm? Where do
appear prominently in films and television productions. Do you you stand?

Chapter review

Our senses encode the information our brains composed of three layers of cells: photoreceptors, bipolar cells and
perceive retinal ganglion cells (RGCs). Photoreceptors take the form of
rods and cones; rods are specialized for low light vision, whereas
• Sensation and perception are separate events that, from the van-
cones are specialized for colour vision.
tage point of the perceiver, feel like one single process. Sensation
is simple awareness due to the stimulation of a sense organ, • The optic nerve is composed of bundled axons from the RGCs; it
whereas perception is a brain activity that organizes, identifies and leaves the eye via the blind spot at the back of each eyeball. Retinal
interprets a sensation in order to form a mental representation. ganglion cells have a receptive field that responds to light falling
anywhere in it; some responses are excitatory, whereas others are
• Transduction is the process that converts physical energy in the
inhibitory.
world into neural signals in the central nervous system. All senses
rely on transduction, although the types of energy being sensed • Cones specialized to sense red, green or blue wavelengths begin
differ, for example light waves for vision, sound waves for the process of colour vision. Combinations of these cones firing
audition. produce the spectrum of colours we can see. Cones also operate in
red-green and blue-yellow opponent combinations to contribute
• Psychophysics was a field of study during the mid- to late 1800s
to colour vision. Additive and subtractive colour mixing deter-
that sought to understand the link between properties of a physi-
mine how shades of colour can be produced.
cal stimulus and people’s psychological reactions to them.
• The optic nerve makes its way through various parts of the brain to
• Psychophysics researchers developed the idea of an absolute
terminate in area V1, the primary visual cortex, located in the
threshold, the minimal intensity needed to just barely detect a
occipital lobe. There specialized neurons respond to the sensation
stimulus, and the difference threshold, the minimal change in a
of bars and edges in different orientations. The ventral stream
stimulus that can just barely be detected. The difference threshold
leaves the occipital cortex to provide a ‘what’ visual pathway to
is also referred to as the just noticeable difference ( JND). Signal
other parts of the brain, while the dorsal stream provides a ‘where’
detection theory represents a refinement of these basic approaches
and ‘how’ pathway.
and takes into account perceived hit, miss, false alarm and correct
rejection rates. • Illusory conjunctions occur when features from separate objects
are mistakenly combined. According to feature integration theory,
• Sensory adaptation occurs when sensitivity to prolonged stimula-
attention provides the ‘glue’ necessary to bind features together.
tion tends to decline over time as an organism adapts to current
The parietal lobe is important for this process of feature binding in
conditions. This adaptive process illustrates that the perceptual
normal and synaesthetic perception.
system is more sensitive to changes in stimulation than to constant
levels of stimulation. • The modular view and the distributed representation view offer
explanations of how we perceive and recognize objects in the
world. At a minimum, humans show a great deal of perceptual
Vision: more than meets the eye constancy. Even as aspects of sensory signals change, perception
• Vision takes place when light waves are transduced by cells in the remains consistent. We are rarely misled to think that distant
eye. Light waves have the properties of length, amplitude and objects are actually tiny, that the moon increases in physical size as
purity. These physical properties are perceived as colour, bright- it rises, or that a friend who grew a moustache is a totally different
ness and saturation, respectively. person.
• Light enters the eye through the cornea and pupil, landing on the • Gestalt psychologists delineated basic perceptual principles long
retina, tissue that lines the back of each eyeball. The retina is ago, such as simplicity, closure, continuity and proximity. Gestalt
178 4
SENSATION AND PERCEPTION

psychologists also observed that we tend to perceive figures set receptor cells are located under the surface of the skin to transduce
against some kind of background. Many visual illusions capitalize pressure, texture, pattern or vibration. There are also receptor cells
on perceptual ambiguities related to these principles. for sensing temperature and pain.
• Template matching and parts-based explanations of object recog- • The somatosensory strip is organized like a homunculus; areas of
nition have strengths and weaknesses. Neither account fully cap- the body that are more sensitive occupy a greater area in the soma-
tures how humans correctly and efficiently perceive objects in tosensory strip. For example, fingertips have a greater representa-
their environment. tion than do the calves of the legs.
• Monocular, binocular and motion-based cues all enable us to per- • Pain is a useful body sense, as without it, we might quickly suc-
ceive size and depth, although we sometimes fall prey to visual cumb to the effects of unnoticed wounds. A-delta fibres and
illusions. Humans are also quite adept at perceiving motion C fibres are two types of pathways by which pain signals reach
through a variety of mechanisms. the brain.
• Gate-control theory proposes a bottom-up and a top-down way of
Audition: more than meets the ear controlling pain signals in the body. This helps to account for indi-
• Hearing takes place when sound waves are transduced by recep- vidual differences in the experience of pain.
tors in the ear. Sound waves have the properties of frequency, • Body position and movement are regulated by receptors located in
amplitude and complexity. These physical properties are perceived the muscles, joints and tendons. Balance is regulated by the semi-
as pitch, loudness and timbre. circular canals in the inner ear and to some extent by visual cues.
• There are three parts of the human ear: the outer ear, the middle
ear and the inner ear. The outer ear channels sound waves towards The chemical senses: adding 昀氀avour
the middle ear, where tiny bones (called ossicles) mechanically • Smell and taste are chemical senses; smell occurs when molecules
transmit and intensify vibrations from the eardrum to the enter the nose, and taste occurs when molecules are dissolved in
inner ear. saliva. Smell and taste combine to produce the experience of
• The inner ear contains the cochlea, which is divided along its flavour.
length by the basilar membrane. The undulation of the basilar • The olfactory epithelium, located at the top of the nasal cavity,
membrane stimulates thousands of tiny hair cells, specialized audi- contains about 10 million olfactory receptor neurons (ORNs).
tory receptor neurons embedded in the basilar membrane. The Each olfactory neuron has receptors that operate like a lock and
hair cells then release neurotransmitter molecules, initiating a key with odourant molecules. Groups of ORNs send their axons
neural signal in the auditory nerve. to a glomerulus within the olfactory bulb.
• A place code and a temporal code are involved in transducing • Pheromones are biochemical odourants that affect behaviour and
sound frequencies. A place code is used for high-frequency sounds, physiology. There is mixed evidence that pheromones affect some
whereas a temporal code is used for low-frequency sounds. aspects of human sexual behaviour.
Auditory signals travel to area A1, the primary auditory cortex in • The tongue is covered with papillae, which contain taste buds, the
the temporal lobe. organs of taste transduction. Each taste bud contains taste recep-
• The placement of the ears on the head enables us to localize sounds tor cells that respond to either salty, sweet, bitter, sour or umami
in the environment. taste sensations. Umami refers to the savouriness of foods.
• Taste and smell contribute to the perception of flavour. Odourants
The body senses: more than skin deep
from food enter the nasal cavity through the nose and the back of
• Haptic perception involves the active exploration of the environ- the mouth. Plugging your nose while you eat can make palatable
ment through touching and grasping. Four types of specialized foods taste bland or make unpalatable foods taste acceptable.

Key terms

absolute threshold (p. 135) gate-control theory (p. 169) pheromones (p. 172)
accommodation (p. 141) hair cells (p. 165) pitch (p. 163)
apparent motion (p. 163) haptic perception (p. 168) place code (p. 165)
area A1 (p. 165) illusory conjunction (p. 151) psychophysics (p. 135)
area V1 (p. 148) just noticeable difference ( JND) (p. 136) receptive field (p. 144)
basilar membrane (p. 165) loudness (p. 164) referred pain (p. 169)
binding problem (p. 151) modalities (p. 132) retina (p. 141)
binocular disparity (p. 159) modularization (p. 153) rods (p. 142)
blind spot (p. 144) monocular depth cues (p. 157) sensation (p. 133)
cochlea (p. 165) motion parallax (p. 159) sensory adaptation (p. 139)
colour-opponent system (p. 147) multisensory integration (p. 167) signal detection theory (p. 137)
cones (p. 142) olfactory bulb (p. 172) structural encoding (p. 157)
d-prime (d’) (p. 138) olfactory receptor neurons (ORNs) (p. 172) synaesthesia (p. 132)
feature integration theory (p. 151) perception (p. 133) taste buds (p. 173)
fovea (p. 143) perceptual constancy (p. 154) template (p. 155)
 SENSATION AND PERCEPTION4 179

temporal code (p. 166) trichromatic colour representation (p. 147) visual orienting (p. 167)
timbre (p. 164) vestibular system (p. 170) Weber’s law (p. 136)
topographic visual organization (p. 148) visual acuity (p. 140)
transduction (p. 134) visual form agnosia (p. 150)

Recommended reading

Enns, J. T. (2004) The Thinking Eye, The Seeing Brain. New York: neuroscience lead to the proposal of dual systems in visual
Norton. A tour through the visual system, focusing on sensations perception.
in the eye and perception in the brain. A fine summary of the key Ward. J. (2008) The Frog Who Croaked Blue: Synaesthesia and
points mentioned in this chapter and a good starting point for the Mixing of the Senses. Hove: Routledge. Written by one of
branching out to other topics in the science of vision. the world’s leading experts on synaesthesia, provides a
Goodale, M. and Milner, D. (2004) Sight Unseen. Oxford: OUP. comprehensive and yet accessible state-of-the-art survey of this
This intriguing book explores conscious and unconscious vision. phenomenon, with a deft mixture of neuroscience and first-
The authors’ arguments from studies of brain damage and person accounts.