ann. behav. med.

(2009) 37:106–116
DOI 10.1007/s12160-009-9097-4

ORIGINAL ARTICLE

Executive Functioning: Overview, Assessment, and Research

Issues for Non-Neuropsychologists
Yana Suchy, Ph.D.

Published online: 20 May 2009

# The Society of Behavioral Medicine 2009

Abstract choices and to engage in purposeful, goal-directed, and

Background Executive functioning (EF) is a multifaceted future-oriented behavior [1, 2]. EF confers an evolutionary
neuropsychological construct that can be defined as (1) advantage by freeing an organism from innate, hard-wired
forming, (2) maintaining, and (3) shifting mental sets, drives and reflexes, as well as from over-practiced, over-
corresponding to the abilities to (1) reason and generate learned, and prepotent responses. A moth, for example, will
goals and plans, (2) maintain focus and motivation to follow be drawn to a light bulb time and time again, no matter if it
through with goals and plans, and (3) flexibly alter goals and burns its wings each time. In contrast, as humans, we
plans in response to changing contingencies. possess the most highly evolved EF of all species, which
Purpose Research involving EF has become increasingly allows us the latitude of considering options and selecting a
popular in a number of disciplines including behavioral specific response to any given stimulus based on situational
medicine and personality research. contexts, previously acquired knowledge, and long-term
Methods The present article outlines some of the chal- goals.
lenges faced by EF researchers, as well as briefly reviews Because EF is a highly effortful and, from an energy
historical perspective and neuroanatomical underpinnings consumption standpoint, a costly process, it remains
of EF to provide broader context. “dormant” for much of our waking hours, coming “on
Results Clinical presentations of EF deficits, definitions line” only when the novelty and/or complexity of a given
and operationalization of the contsruct, as well as selection situation precludes an automatic, routine response. Put
of appropriate assessment methods are provided. another way, a response that is automatic or routine is, by
Conclusion The article concludes with general cautions definition, not a reflection of EF. This latter notion is
and guidelines for researchers. important and is invoked later when the concept of “skills”
will be discussed.
Keywords Executive functioning . Frontal lobes . In clinical neuropsychology, it has been understood for a
Neuropsychological assessment . Frontal-lobe syndrome . number of decades that EF represents one of a handful of
Cognition core neurocognitive domains. Outside of neuropsychology,
the EF construct has entered the research arena relatively
recently, representing a new and exciting area of interdis-
Introduction ciplinary inquiry. Researchers tackling this new direction
face a number of challenges, most notably difficulties with
Executive functioning (EF) is a multifaceted neuropsycho- defining, operationalizing, and assessing the EF construct.
logical construct consisting of a set of higher-order neuro- The present paper reviews these challenges and offers some
cognitive processes that allow higher organisms to make general guidelines.

Y. Suchy (*) Brief Overview of EF Neuroanatomy

Department of Psychology, University of Utah,
380 S. 1530 E., Rm. 502,
Salt Lake, UT, USA Traditionally, EF has been described as being associated
e-mail: [Link]@[Link] with the so-called prefrontal cortex (PFC) of the brain,
ann. behav. med. (2009) 37:106–116 107

which includes all portions of the frontal lobes that are (see Fig. 2). Thus, for example, working memory is
located anterior to motor and premotor cortices and the dependent not only on the dorsolateral PFC but also on
supplementary motor area (see Fig. 1). The prefrontal portions of the parietal lobe; response initiation is depen-
cortex can be divided into three main convexities: (a) the dent not only on the left medial and ventral prefrontal
dorsolateral PFC, often described as the substrate of cortices but also on the basal ganglia and the thalamus;
working memory [3]; (b) the superomedial PFC (which sustained attention is dependent not only on the super-
also includes the anterior cingulate gyrus), often described omedial PFC but also the integrity of many regions within
as the substrate for sustained attention, response selection, the right hemisphere and the thalamus; and so on and so
and motivation [4, 5]; and (c) the ventral (or inferior) PFC forth. In fact, virtually all EF components require the
(which can be divided into orbitofrontal and ventromedial), integrity of circuits involving portions of the PFC, the basal
often described as the substrate for inhibition, social ganglia, the thalamus, and the cerebellum, as well as
appropriateness, and sensitivity to rewards and punishments cortical areas outside of the frontal lobes [14–17]. Clearly,
[6–8]. In recent years, the most anterior portions of the the complexity of these networks is beyond the scope of
frontal lobes, known as the frontal pole (which includes this paper but suffice it to say that individual aspects of EF
anterior portions of the dorsolateral and ventral prefrontal should not be viewed as easily localized.
cortices), have also received much attention due to their role
in morality [9], empathy [10], and higher order integration of Executive Dysfunction in Clinical Populations
EF [11].
The functions of the prefrontal cortex can also be further In clinical neuropsychology, a patient with EF deficits is
subdivided by the two cerebral hemispheres, such that the typically someone who has suffered some type of a
left PFC is associated with initiation of responses, as well brain injury or is afflicted with a neurodevelopmental or
as processing of information that is verbal, concrete, or neurodegenerative illness. Typical examples would be
detail-oriented, whereas the right PFC is associated with individuals who have suffered a stroke in the frontal lobes
inhibition of responses, as well as processing of information and/or the related brain circuitry, victims of motor vehicle
that is visual-spatial, abstract or connotative, and gestalt- accidents, or patients with vascular, frontotemporal, or
oriented [12, 13]. Parkinson’s dementia. Among such individuals, EF defi-
However, it is now well understood that the PFC is not cits, especially if severe, can be extremely devastating.
the only brain area involved in EF. Because the frontal Practicing clinical neuropsychologists could likely generate
lobes are richly connected to a variety of other brain virtually infinite numbers of examples: Social inappropri-
regions, most EF processes depend on the integrity of ateness, such as persistent inappropriate sexual advances
complex networks rather than a single frontal lobe region toward strangers; lapses in judgment, such as purchasing

Fig. 1 Convexities of the

prefrontal cortex LATERAL VIEW Dorsolateral Supplementary
Motor Area
Prefrontal

Primary Motor

Fronto-polar
Secondary Motor
(premotor)
Orbitofrontal

VENTRAL VIEW

MEDIAL VIEW

Supero-medial

Ventro-medial
108 ann. behav. med. (2009) 37:106–116

Fig. 2 Reciprocal connections between various convexities of the without an outline represent subcortical or deep brain structures
prefrontal cortex and limbic and subcortical structures. Ovals outlined (amygdala, hippocampus, and basal ganglia/thalamus circuits). Curved
in black represent lateral cortical areas and cerebellum. Ovals outlined thick lines represent cortico-cortical connections, whereas thin straight
in gray represent medial cortical areas. Terms inside gray shapes lines reflect cortico-subcortical connections

100 roles of paper towels without any understanding of disorders, obsessive compulsive disorder, and schizophre-
how and why that happened; a complete lack of motiva- nia [18], as well as Axis II disorders, such as antisocial [19]
tion, such as the inability to get out of bed to the point of and borderline [20] personality disorders, have been
laying in own excrement; or being “stimulus-bound” such associated with EF impairments.
that (similar to the example of the moth and the light bulb)
a sight of food leads to eating, a sight of a pencil leads to Historical Perspective
drawing, a sight of a ball leads to kicking, a sight of a
remote control leads to button pushing, and so on, no Just as one is unaware of breathing until deprived of
matter what the consequences are. oxygen, one is unaware of the richness and complexity of
These examples illustrate the devastating functional one’s neurocognitive processes until deprived of them.
impairments seen among individuals with severe damage Thus, historically, many cognitive functions came to the
to the EF substrates. They do not, however, afford apprecia- attention of scientists only after their absence has been
tion of the individual differences in EF among non-brain- observed among brain-injured individuals. The first de-
injured people. To appreciate the range of EF abilities, scription of the absence of EF dates back to 1835 [21],
readers may consider simple EF lapses that all of us have although it was another 13 years before the most well-known
experienced at one time or another. Most of us have missed a (and often mis-cited as the first) case of EF dysfunction came
freeway exit while daydreaming; erroneously pressed the to be described. On September 13, 1848, a competent, well-
most-frequently called number on a speed dial; left our house liked, even-tempered railroad foreman named Phineas Gage
keys in illogical places, such as the freezer or the sock sustained an injury in which a tamping iron penetrated his
drawer; or responded angrily in a situation when keeping skull and severely damaged a considerable portion of his
emotions under control would have been more consistent frontal lobes. Although Phineas Gage survived the accident,
with our intentions or goals. The frequency and seriousness the injury resulted in a now well-recognized syndrome
of such lapses in everyday life can be viewed as a reflection characterized by social inappropriateness, emotional dysre-
of a person’s strengths or weaknesses in EF. gulation, and behavioral dyscontrol [22].
However, it is also important to note that very frequent Although many other cases with similar types of deficits
lapses of this type signal psychopathology or neuropsychi- were later described in the literature, it was Alexander
atric illness. In fact, many Axis I disorders, such as mood Luria, a Russian neuropsychologist, who in the 1960s and
ann. behav. med. (2009) 37:106–116 109

1970s systematically studied the relationship between someone who is not familiar with the construct, this
frontal-lobe injuries and social, emotional, and behavioral approach may well be the most comprehensive and
dyscontrol, popularizing the term “frontal-lobe syndrome” conceptually meaningful. Similarly, this approach may
[23–25]. provide a good framework for explaining the signif-
Although the term “frontal-lobe syndrome” is still occa- icance of EF research or its relevance for a person’s
sionally used today, it has fallen out of favor. Over time, it has success or quality of life. However, this approach is
become increasingly clear that the syndrome (a) does not less useful when it comes to actually operationalizing
always occur following a frontal lobe injury and (b) EF for research or clinical observations, hence, the
sometimes does occur following injuries elsewhere in the utility of the following approaches.
brain. In other words, it became increasingly clear that (2) Clinical syndromes. In clinical neuropsychology, one
equating the frontal lobes with cognitive and behavioral classification system of EF subcomponents relies on
control was inadequate. This led to the introduction of the the syndromes associated with different types of brain
term “executive functions” [26]. Although the term is now lesions [27]. In particular, lesions in the dorsolateral
widely adopted, our understanding of what it actually stands PFC and the related circuitry tend to lead to problems
for continues to be a source of controversy and debate. in the areas of reasoning, organization, planning, and
problem solving, all of which rely on, among others,
Approaches to Defining EF working memory. This constellation of symptoms has
been referred to as the “dysexecutive” syndrome.
The exact nature of the processes that fall under the Lesions in the orbitofrontal and ventromedial prefron-
umbrella of EF is still not well understood, and conse- tal cortices and the related circuitry tend to lead to
quently, there is no universally accepted definition. In fact, problems in the area of social appropriateness and
even the approaches to defining EF vary widely among making adaptive choices, relying on inhibition and
researchers. EF can be defined in a number of ways: (1) its response selection. This constellation of symptoms has
overarching evolutionary purpose (e.g., ability to make been referred to as the “disinhibited” (or sometimes
choices and to engage in purposeful, goal-directed, future- “pseudo-psychopathic”) syndrome. Lesions in the
oriented behavior), (2) the clinical syndromes observed in superomedial PFC tend to lead to problems in the
individuals who, due to brain damage, lose the abilities to area of motivation, initiation, and sustained attention,
engage in purposeful, goal directed behavior, (3) the list of referred to as the “apathetic” syndrome (see Table 1).
the complex skills that contribute to purposeful, goal-directed
behavior (e.g., reasoning and problem solving, planning,
organization), (4) the more elemental neurocognitive pro- While defining EF based on clinical syndromes is useful
cesses that form the basis for specific EF skills (e.g., working in the clinical evaluations of neurologic populations, it is
memory, sequencing, inhibition, initiation, and response less helpful in non-neurologic settings. This is because
selection), (5) atheoretical approaches, driven by available neurologic lesions are often defined by structural, not
variables or populations, (6) the neuroanatomic substrates functional, boundaries. For example, lesions due to strokes
that subserve complex skills and elemental neurocognitive follow the architecture of blood vessels, not brain functions.
processes (e.g., dorsolateral, superomedial, and ventral Additionally, with respect to brain organization, contiguity
prefrontal cortices), and (7) constructivist theories that offer does not always equal connectivity, as functional circuitry
a framework for organizing the building blocks of EF. The sometimes involves brain regions that are located far apart,
advantages and disadvantages of each approach are consid- even in separate lobes. And, unlike the contiguous brain
ered below. An overview of the associations among the areas that are affected in acquired lesions, it is these
different levels of analysis of the building blocks of EF is functional circuits that are often the target of various
shown in Table 1. neurodevelopmental anomalies, that is, errors that occur
during pre- or perinatal development. Consequently, the
(1) Overarching evolutionary purpose. As alluded to constellation of symptoms seen in neuropsychiatric popula-
earlier, the overarching purpose of EF is to allow an tions does not always resemble the neurologic syndromes
organism to make choices and to engage in purpose- described above. Thus, although understanding of these
ful, goal-directed, and future-oriented behavior. Given syndromes may be useful in formulation of hypotheses,
the complicated, multifaceted nature of EF, defining it development of research questions, and conceptualization of
simply by its overarching purpose may well be the results, description of individual symptoms (that is, deficits
best approach, as it avoids the pitfalls of incomplete or in either individual complex skills or individual elemental
poorly organized laundry lists of terms that themselves neurocognitive processes, described below) are often a more
often need further defining. When explaining EF to fruitful target for research of non-neurologic populations.
110 ann. behav. med. (2009) 37:106–116

Table 1 Definitions of the components of executive functions (EF)

Overarching purpose Cognitive skill Neurocognitive processes Corresponding

neurobehavioral syndrome

Generating short-term and Planning and reasoning Focusing attention Dysexecutive/Disorganized

long-term goals (primarily lesions in the
OR “Set Formation” Generativity & memory retrieval dorsolateral prefrontal
Working memory cortex)
Sequencing
Executing short-term and Organization Sequencing
long-term goals Working memory
OR “Set Maintenance”
Follow-through Initiation Apathetic (primarily lesions
Response selection/conflict resolution in the superomedial
Selective attention prefrontal cortex)
Self monitoring
Attentional vigilance
Social appropriateness, judgment Inhibition, Discrepancy detection Disinhibited or
Psychopathic (primarily
lesions in the ventro-
medial and orbitofrontal
cortex)
Altering short-term and Problem-solving Discrepancy detection Dysexecutive/ Perseverative
long-term goals in Cognitive flexibility (primarily lesions in the
response to changes in dorsolateral prefrontal
situations and contexts cortex)
OR “Set Shifting” Attentional shifting
Generativity & memory retrieval
Working memory

(3) Complex skills. Describing EF in terms of complex does not come on line in situations in which behavior can
skills is particularly useful in settings in which rely on learned, routine, or automatic responses.
concrete observable everyday behaviors need to be But the problem is not with the word “skills” but rather
identified. Planning, reasoning, problem solving, with the actual abilities that fall under the EF skill category.
judgment, etc., are terms that can be easily understood, Take reasoning, for example. Reasoning skills are system-
and behaviors that can be observed and reported on, atically taught and developed across all levels of formal
by research participants, patients, or caregivers. Sim- schooling, with individuals who achieve higher levels of
ilarly, using descriptions of concrete skills, researchers education typically having better developed reasoning
can make conceptually meaningful claims that can skills. Consequently, an identical situation may then require
readily be understood by other non-neuropsychology a relatively minimal involvement of EF from a person with
researchers. For example, one may say that patients well-developed reasoning, in contrast to a relatively heavy
with depression often find themselves unable to reliance on EF from a person with poorly developed skills.
generate solutions because they suffer from impair- Or, conversely, depending on the context, the same person
ment in reasoning, which falls under the umbrella of may appear to have adequate EF in a situation that has, for
EF. It is likely this ease in communication that has them, become routine, but poor EF in another, more novel
rendered complex skills arguably the most popular situation. Consider, for example, an elderly person who
approach to EF definitions. lives alone and independently takes care of her daily needs:
She pays bills, shops, and cooks; that is, she engages in
Despite its popularity, this approach is not without purposeful and goal-directed behaviors, the presumed output
problems. The major weakness of this approach is, ironically, of intact EF. However, when faced with a life-changing
inherent in its very reliance on skills—a learned ability to do situation, such as a move to a new home, this individual
something competently. In other words, skills, by definition, suddenly “develops” a severe EF deficit: Cooking, shopping,
refer to something one can acquire, learn, practice, and and bill paying are suddenly beyond her EF capacity.
improve, and as such they come on line in routine, or Having a good conceptual grasp on the construct of EF
previously practiced, contexts. Recall that EF, in contrast, allows one to understand that such individuals do not
ann. behav. med. (2009) 37:106–116 111

suddenly develop dementia. Rather, when removed from a (5) Atheoretical approaches. As if the above approaches
routine situation that requires minimal EF involvement and to defining EF were not challenging enough, things
placed into a novel situation in which EF needs to be on are further complicated by EF definitions that are
line much of the time, evidence of functional impairment opportunistic and/or atheoretical; that is, driven by
may become apparent. In other words, the latter situation is which variables, tasks, or populations are available for
simply a more valid “test” of actual EF abilities. And, thus, any given study. Such approaches generally miss the
in the final analysis, the use of specific skills for defining boat with respect to the conceptual understanding of
the construct of EF may lead to paradoxical conclusions: EF. On the one hand, readers may encounter defi-
The better practiced the skills, the less reflective of EF they nitions that oversimplify the construct of EF, using a
actually are. single term or a phrase as a synonym for the entire
Given these issues, researchers need to use caution when construct and failing to consider EF’s multifaceted
using a list of complex skills to define EF. As stated above, nature. For example, EF has been referred to by single
such definitions may facilitate communication or may be terms such as “cognitive control,” “response selection,”
necessary when ratings of everyday behaviors are used as “problem solving,” or “selective attention.” On the
the dependent variables. But in laboratory settings, discrete other hand, readers may encounter definitions of EF
neurocognitive processes likely represent a better and more that are essentially long laundry lists of terms that do
valid approach to defining and operationalizing the EF not fall neatly into any of the above-listed categories
construct. but rather mix and match some subset of the purposes,
skills, processes, substrates, and syndromes.
(4) Elemental neurocognitive processes. Unlike complex
skills, discrete elemental neurocognitive processes are Additionally, because purposeful, goal-directed behavior
less susceptible to prior learning, familiarity with a is central to human existence, the notion of EF permeates
situation, or educational achievement. They are the stuff psychology even when the construct itself is not invoked.
that cognitive psychology and cognitive neuroscience In fact, self-regulation, self-control, emotion regulation,
are made of. Under the umbrella of EF, one can find such delay of gratification, attentional control, self-monitoring,
processes as initiation, inhibition, response selection, and and response modulation, to name a few, all rely on some
working memory (see Table 1). Such processes have aspects of EF [31–36]. From among the terms listed above,
been identified by breaking down performances on attentional control deserves further mention, as there is
various cognitive tasks into their most basic compo- considerable overlap between attention and executive
nents. Historically, this has been accomplished by often processes. Specifically, automatic attention occurs sponta-
ingenious manipulations of stimulus onset asynchro- neously in response to stimuli and is present even among
nies, stimulus presentation latencies, the order of primitive species or among individuals who are consider-
stimulus presentation, or the location of stimuli in one’s ably cognitively impaired, whereas directed attention is
visual field. In recent years, such cognitive methods under EF control [37] and is sometimes included in the list
have been supplemented with functional neuroimaging, of EF skills.
allowing further validation of discrete neurocognitive
(6) Neuroanatomic substrates. In the wake of what
processes by demonstrating their reliance on predictable
appears to be an unending frustration, the definition
neuroanatomic networks [28–30].
of EF has recently come the full circle, invoking, once
Despite its considerable advantages and ever-increasing again, the all but forgotten “frontal-lobe syndrome.” In
popularity, there are some pitfalls in this approach, which particular, Donald Stuss, a prominent EF researcher,
call for caution. In particular, although various discrete has explicitly elected to focus on the functional
aspects of EF identified via these methods do correlate with neuroanatomy of the human frontal lobes first and
neurocognitive, neuroanatomic, physiologic, genetic, and the EF construct second. He writes, “We have limited
psychological variables, none of these components alone our research to patients with focal lesions of the
defines EF. Consequently, to be conceptually meaningful frontal lobes….In our opinion this is the first step in
and complete, a definition of EF needs to rely on a list of limiting the terms of reference in the study of
such elemental processes. Unfortunately, these processes executive and frontal functions” (p 291) [38]. From a
themselves are somewhat imprecisely defined, and the neuropsychological standpoint, this may, in the end,
potential interpretations of individual terms are nearly as be the most reasonable approach: After all, one cannot
numerous as the researchers who use them. For that reason, argue with what functions should be included in this
a careful and consistent use of terminology, as well construct, as the physical boundaries are undeniably
thoughtful defining of terms, should always be a goal for clear. However, the utility of this approach for person-
researchers tackling this line of study. ality and psychopathology research is questionable.
112 ann. behav. med. (2009) 37:106–116

(7) Constructivistic definitions. Given the complex nature For the purpose of organizing information into Tables 1,
of EF and the difficulties with defining it, a number of 2, and 3, a tripartite model of EF is used, which consists of
researchers have generated theoretical models that three entities: the abilities to (1) form, (2) maintain, and (3)
introduce new constructs, or latent variables, that shift mental set [41–43]. These components correspond
could explain the EF structure and function. Such with the abilities to (1) reason and problem solve, (2)
models sometimes propose a single unifying entity maintain motivation to follow through with a response in
that is hoped to capture the EF construct in its entirety, the absence of external structure, and (3) alter goals and
such as the “central executive” [39, 40]. This and other plans as needed in response to changing contingencies.
similar models provide a very useful framework for The advantage of this model is that it affords descrip-
the study of the construct of EF. However, they do not tions of different profiles of strengths and weaknesses
always lend themselves to research in which EF among the three major EF subcomponents (i.e., set
strengths and weaknesses among different populations formation, set maintenance, and set shifting), which can
are studied, as the ability of such models to truly lead to certain predictable behavioral responses. In partic-
capture all clinically and theoretically relevant aspects ular, individuals with a strength in the area of forming
of EF has not yet been demonstrated. In other words, mental set generate new ideas easily and readily come up
whether a particular model is useful needs to be with plans or solutions to problems. However, such a set of
determined based on the theoretical questions of any strengths does not necessarily translate into successful
given study and on the aspects of EF that, in theory, execution of goals or plans. For that, adequate set
uniquely characterize a given population. maintenance needs to also be present, as well as the ability

Table 2 Review of clinical measures of executive functioning and subcomponent processes

Clinical tasks Nonexecutive Set formation Set shifting Set maintenance

components
Working Generativity Cognitive Response Inhibition Initiation Attent’l
memory flexibility selection vigilance

WCST-CA Visual-spatial * *
WCST-PR Visual-spatial * *
WCST-FMS * *
Figural fluency- Visual-spatial, * * *
unique designs graphomotor,
processing speed
Figural fluency Visual-spatial * * * *
errors
Verbal fluency Language, processing * * *
words speed
Verbal fluency Language, processing * * * *
errors speed
Stroop color and Visual spatial, language, * * *
word test processing speed
Trail making Visual spatial, language, * * *
test-part B processing speed
Porteus mazes Visual spatial * *
errors
Rey Figure-Copy Visual spatial *
Towers Visual spatial * *
CPT-impulsivity *
CPT-vigilance *
Go/No-go *
commissions
Go/No-go *
omissions

“Towers” examples include Tower of Hanoi and Tower of London; “Verbal fluency” examples include Controled Oral Word Association (COWA)
and FAS; “Figural fluency” examples include Ruff Figural Fluency Test (RFFT)
WCST-CA Wisconsin Card Sorting Test-categories achieved, WCST-PR WCST perseverative responses, WCST-FMS WCST failure to maintain set,
CPT continuous performance test
ann. behav. med. (2009) 37:106–116 113

Table 3 Review of experimental measures of executive functioning and subcomponent processes

Experimental tasks Nonexecutive components Set formation Set Set maintenance

shifting
Working Generativity Cognitive Response Inhibition Initiation Attent’l
memory flexibility selection vigilance

ANT-Exec Control Visual spatial abilities * *

Reversal Learning Visual spatial abilities * * * * *
Cued Switching Verbal or visual spatial * * *
Tasks- Switching abilities, depending on the
Costs task
Uncued Switching Verbal or visual spatial * * * * *
Tasks- Costs abilities, depending on the
task
Cued Switching Verbal or visual spatial *
Tasks- Switching abilities, depending on the
errors task
Uncued Switching Verbal or visual spatial * * *
Tasks- Errors abilities, depending on the
task
N-Back Processing speed *
Iowa Gambling * * *
Task

ANT attention network test

to shift to a new plan if new situational constraints present switching abilities often provide explicit instructions to switch
themselves. Conversely, some individuals may have a [45]. It is against the backdrop of these challenges that the
weakness with respect to conceptualization and problem remainder of this section needs to be considered.
solving, but when presented with a reasonable plan
(generated by someone else), they can successfully follow Clinical Tests
through with the plan’s execution [44] (see Table 1).
From a historical standpoint, given the initial conceptualiza-
Assessment Approaches tion of EF impairments as manifestations of the “frontal-lobe
syndrome,” the earliest assessment instruments were developed
Assessment of EF has traditionally been fraught with with the goal of identifying individuals with frontal-lobe
difficulties. The first and the most obvious issue is the lack lesions. Well-known examples include the Wisconsin Card
of a definition or a theory that would provide an adequate Sorting Test (WCST) [46], the Trail Making Test-Part B
framework for assessment and test development. Another (TMT-B) [47], and measures of verbal and figural fluency [48]
central problem with EF assessment is the rich connectivity (see Table 2). These tests have been extensively validated via
(via rich neuroanatomic connections) of the EF system demonstration of their sensitivity to frontal-lobe lesions [49–
with other cognitive systems. Consequently, it is virtually 51] and continue to be popular in clinical and research settings
impossible to assess EF without incidentally assessing to this day. In fact, some of them represent among the most
other, nonexecutive processes. sensitive, if not specific, EF instruments.
The third central challenge is the fact that EF weaknesses The issue of specificity is an important one and deserves
often become apparent only in situations that lack structure. further discussion, as typical clinical measures of EF virtually
For example, some people with EF weaknesses find it difficult always assess several different components of EF as well as
to spontaneously switch from one mental set to another, other nonexecutive functions. This is not an insurmountable
finding themselves stuck when problems in everyday life problem in clinical settings, where entire batteries of tests are
arise. However, when cued by someone that switching needs used and the locus of a deficit can be deduced through
to take place, they are capable of generating ideas and solving process of elimination. In other words, if a patient performs
problems. Unfortunately, most clinical and experimental well on measures of processing speed and working memory,
testing situations unavoidably provide structure, which is but poorly on TMT-B, then it can be deduced that cognitive
often enough to help individuals with mild EF deficits to flexibility is the culprit (see Table 2). However, in many
overcome their weaknesses. As an example, tests that assess research situations, such painstaking deduction process is not
114 ann. behav. med. (2009) 37:106–116

practical, and in the final analysis, most research findings Experimental Tasks
that are based on these instruments offer minimal discrim-
ination with respect to discrete neurocognitive processes. Despite their lack of norms, experimental measures have
An additional problem with the use of clinical measures in a number of advantages over clinical instruments: They (a)
research is that some of these tests yield relatively narrow tend to be designed to maximize sensitivity and specificity
ranges of scores with low ceilings. This, again, is not to discrete aspects of EF, (b) generate normally distributed
necessarily a problem in clinical settings, where only impaired scores, and (c) are relatively resistant to previously learned
individuals need to be identified and where normal distribu- skills. Most such measures do not lend themselves to
tion among healthy persons is not necessarily relevant. clinical evaluations for reasons that are beyond the scope of
However, in research, normally distributed scores with this paper but can, if used and interpreted properly, answer
adequate range of scores are crucial. The WCST, arguably a variety of research questions. The tasks of this type can
the most misused neuropsychological measure in non- generally be divided into three main categories: (1)
neuropsychological research, is a good example of this cognitive control, (2) working memory, and (3) emotional
problem. The scores on this test are almost never normally decision making. For a list of tasks, see Table 3.
distributed, and most healthy young individuals, unless Switching tasks [45], Go/No-Go and reversal learning
neurologically impaired, produce virtually perfect scores. tasks [52, 53], computer-administered Stroop tasks [54], and,
Consequently, studies that use this test for identification of recently, the Attention Network Test [55] are the most typical
subtle EF weaknesses among individuals with depression, examples of cognitive control measures. All these tasks share
anxiety disorders, or personality disorders frequently fail to in common the need to select one of several possible
reject the null hypothesis. responses to a given stimulus (i.e., conflict resolution). This
And finally, some clinical measures are based on the often involves, among others, inhibiting a response that is
“skills” model of EF, relying heavily on practiced abilities prepotent. The prepotent status of a given response is either
such as reasoning, problem solving, and planning. While due to the response being previously over-learned (e.g., on a
the WCST is somewhat problematic in this regard, even Stroop task, word reading is prepotent to naming the color of
more problematic are the various Tower tests, and, even print) or due to being primed as part of the experimental
more so, the 20 Questions test that is part of the Delis– procedure (e.g., on a switching task, a participant generates a
Kaplan Executive Function System battery [48]. Consider particular response on one trial, but that same response is
the following situation: Subject A grew up in a family in incorrect on the subsequent trial). It is important to note that
which playing the 20 Questions game was a common and although all these types of tasks are often referred to as
popular pastime. Subject B, on the other hand, has never assessing cognitive control, each assesses somewhat different
played the game in his life. When performing the 20 sets of control processes. For example, only inhibition of a
Questions test, Subject A can readily rely on over-learned response is needed on Go/No-Go tasks, which require that
sets of strategies, rambling off a series of questions about participants respond to some stimuli, while avoid response of
the mystery object’s semantic category or the size of the any type to some other stimuli. In contrast, response
bread box. In other words, Subject A performs the test with selection is needed on Switching Tasks, which require that
minimal, if any, EF involvement. In contrast, Subject B needs participants respond to a set of stimuli in a particular way (e.
to generate novel strategies, relying heavily on EF. Clearly, g., the colors of presented designs), until cued to respond in
for each subject, the test is differentially sensitive to EF or another way (e.g., the shapes of presented designs). See
may even measure an entirely different construct. While in Table 2.
clinical settings, the appropriateness of the measure for any In contrast to cognitive control tasks, working memory
given patient can be determined via clinical interview; this is tasks only require that participants hold information in
again not practical in research situations. working memory, without any conflict resolution require-
Despite their weaknesses, there are some advantages to ment. Such tasks often utilize the so-called N-back
using clinical measures in research. The most notable paradigm, in which participants need to continually hold
among these is the fact that clinical measures virtually in mind a stimulus that was presented an N number of trials
always provide published norms. Thus, if a studied group (most typically two) back.
differs from a control group, one can use the norms to Finally, the emotional decision making tasks, such as the
determine the clinical significance of such a difference, i.e., Iowa Gambling Task (IGT), require that participants weigh,
whether the studied group is characterized by deficits (that toward a specific goal, the relative contributions of rewards
is, whether performances fall into a bona-fide impaired and punishments [56]. In the IGT, participants are
range), or just a relative weakness. Such determinations instructed to select cards that are presented face down in
typically cannot be made with the use of experimental four decks, with the goal of winning as much money as
instruments, which are discussed next. possible. Selecting cards from two of the decks (the “bad”
ann. behav. med. (2009) 37:106–116 115

decks) results in large winnings but even larger losses, Acknowledgment The author wishes to thank Drs. Paula Williams,
Timothy Smith, and Gordon J. Chelune for valuable feedback and
while selecting cards from the other two the decks (the
suggestions.
“good” decks) results in small winnings but even smaller
losses. While the IGT is the most well-known example of
these types of tasks, a number of other tasks have been References
utilized in a similar manner, such as the passive avoidance
task utilizing varied reward and punishment schedules in 1. Cummings JL, Miller BL. Conceptual and clinical aspects of the
psychopathy research [57, 58]. frontal lobes. The human frontal lobes: Functions and disorders.
Some caution in interpreting experimental tasks is in 2nd ed. New York, NY, US: Guilford Press; 2007:12–21.
order. In particular, because such tasks measure very discrete 2. Gazzaley A, D'Esposito M, Miller BL, Cummings JL. Unifying
prefrontal cortex function: Executive control, neural networks, and
neurocognitive processes, they may in some situations be top-down modulation. The human frontal lobes: Functions and
less sensitive to a generalized EF weakness than clinical disorders. 2nd ed. New York, NY, US: Guilford Press; 2007:187–
measures would be. This is because some EF deficits only 206.
become apparent as a result of the cumulative effect of a 3. Fuster JNM. The prefrontal cortex of the primate: A synopsis.
Psychobiology. 2000; 28: 125–131.
number of slight weaknesses across a number of discrete EF 4. Bush G, Frazier JA, Rauch SL, et al. Anterior cingulate cortex
subcomponents. In contrast, in other situations, assessment dysfunction in attention-deficit/hyperactivity disorder revealed by
of discrete processes may artificially inflate the impression fMRI and the Counting Stroop. Biol Psychiatry. 1999; 45: 1542–1552.
of a deficit. This happens when a given population is 5. Sewards TV, Sewards MA. Representations of motivational drives
in mesial cortex, medial thalamus, hypothalamus and midbrain.
characterized by a considerable anomaly in a single discrete Brain Res Bull. 2003; 61: 25–49.
EF process, an anomaly that only becomes apparent under 6. Angrilli A, Palomba D, Cantagallo A, Maietti A, Stegagno L.
certain very specific situations. The result, if robust and Emotional impairment after right orbitofrontal lesion in a patient
replicated, may have some very useful theoretical implica- without cognitive deficits. Neuroreport. 1999; 10: 1741–1746.
7. Mathiesen BB, Farster PLV, Svendsen HA. Affect Regulation and
tions but may also have very limited implications for how Loss of Initiative in a Case of Orbitofrontal Injury. Neuro-
one’s actual EF manifests itself in everyday life. Psychoanal. 2004; 6: 47–62.
8. Tremblay L, Schultz W. Reward-related neuronal activity
Summary during go-nogo task performance in primate orbitofrontal cortex.
J Neurophysiol. 2000; 83: 1864–1876.
9. Moll J, Eslinger PJ, de Oliveira-Souza R. Frontopolar and
The present paper provides an overview of challenges anterior temporal cortex activation in a moral judgment task.
associated with EF research, especially when conducted Arq Neuropsiquiatr. 2001; 59: 657–664.
with non-neurologic populations. A conceptual overview 10. Ruby P, Decety J. How would you feel versus how do you think
she would feel? A neuroimaging study of perspective-taking with
of different approaches to defining and assessing EF, as social emotions. J Cogn Neurosci. 2004; 16: 988–999.
well as strengths and weaknesses of such approaches, is 11. Koechlin E, Basso G, Pietrini P, Panzer S, Grafman J. The role of
provided, with the goal of affording researchers options the anterior prefrontal cortex in human cognition. Nature. 1999;
for selection of an approach that best fits their purposes 399: 148–151.
12. Lezak MD, Howieson DB. Loring DW:. Neuropsychological
and contexts. Readers are encouraged to adhere to the Assessment: Oxford University Press; 2004.
following guidelines when using the EF construct in their 13. Stuss DT, Alexander MP, Floden D, et al. Fractionation and
research: localization of distinct frontal lobe processes: Evidence from focal
lesions in humans, Principles of frontal lobe function. London:
Oxford University Press; 2002:392–407.
1. Develop a solid conceptual understanding of the 14. Aron AR. Progress in executive-function research: From tasks to
construct of EF to facilitate appropriate study design functions to regions to networks. Curr Dir Psychol Sci. 2008; 17:
and interpretation of findings. 124–129.
2. Carefully define the EF construct in their writing, 15. Braver TS, Ruge H, Cabeza R, Kingstone A. Functional neuro-
imaging of executive functions. Handbook of functional neuro-
considering the audience and the purpose of their imaging of cognition. 2nd ed. Cambridge, MA US: MIT Press;
reference to EF; in other words, different definition 2006:307–348.
approaches may be appropriate for conceptual discus- 16. Ottowitz WE, Dougherty DD, Savage CR. The neural network
sions of EF versus empirical investigations of EF. basis for abnormalities of attention and executive function in
major depressive disorder: Implications for application of the
3. Carefully define all EF terms and variables, as no medical disease model to psychiatric disorders. Harv Rev
uniform definitions exist and the literature on the whole Psychiatry. 2002; 10: 86–99.
is fraught with confusion and misinterpretation. 17. Tekin S, Cummings JL. Frontal-subcortical neuronal circuits and
4. Carefully select assessment instruments, considering clinical neuropsychiatry: An update. J Psychosom Res. 2002; 53:
647–654.
the studied population, the expected nature and severity 18. Moritz S, Birkner C, Kloss M, et al. Executive functioning in
of EF weaknesses, and the degree to which clinical obsessive-compulsive disorder, unipolar depression, and schizo-
versus theoretical questions need to be answered. phrenia. Arch Clin Neuropsychol. 2002; 17: 477–483.
116 ann. behav. med. (2009) 37:106–116

19. Morgan AB, Lilienfeld SO. A meta-analytic review of the 38. Stuss DT, Alexander MP. Executive functions and the frontal lobes:
relation between antisocial behavior and neuropsychological A conceptual view. Psychological Research. 2000; 63: 289–298.
measures of executive function. Clin Psychol Rev. 2000; 20: 39. Baddeley A, Della Sala S, Roberts AC, Robbins TW, Weiskrantz
113–156. L. Working memory and executive control. The prefrontal cortex:
20. LeGris J, van Reekum R. The Neuropsychological Correlates of Executive and cognitive functions. New York, NY US: Oxford
Borderline Personality Disorder and Suicidal Behaviour. Can J University Press; 1998:9–21.
Psychiatry. 2006; 51: 131–142. 40. Baddeley A. The central executive: A concept and some
21. Lyketsos CG, Rosenblatt A, Rabins P. Forgotten frontal lobe misconceptions. J Int Neuropsychol Soc. 1998; 4: 523–526.
syndrome or 'Executive Dysfunction Syndrome'. Psychosomatics. 41. Osmon DC. Complexities in the evaluation of executive functions.,
2004; 45: 247–255. Forensic neuropsychology: Fundamentals and practice. Lisse,
22. Macmillan M. An odd kind of fame: Stories of Phineas Gage. Netherlands: Swets & Zeitlinger Publishers; 1999:185–226.
Cambridge, MA US: The MIT Press; 2000. 42. Osmon DC, Suchy Y. Fractionating frontal lobe functions: Factors
23. Luria AR. The frontal lobes and the regulation of behavior. of the Milwaukee Card Sorting Test. Arch Clin Neuropsychol.
Psychophysiology of the frontal lobes. Oxford England: Academic 1996; 11: 541–552.
Press; 1973. 43. Suchy Y, Kosson DS. Forming, switching, and maintaining mental
24. Luria AR, Homskaya ED, Blinkov SM. Impaired selectivity of sets among psychopathetic offenders during verbal and nonverbal
mental processes in association with a lesion of the frontal lobe. tasks: Another look at the left-hemisphere activation hypothesis.
Neuropsychologia. 1967; 5: 105–117. J Int Neuropsychol Soc. 2006; 12: 538–548.
25. Luria AR, Majovski LV. A frontal lobe syndrome due to a 44. Suchy Y, Blint A, Osmon DS. Behavioral Dyscontrol Scale:
cerebrovascular accident affecting the corpus callosum. Clin Criterion and predictive validity in an inpatient rehabilitation unit
Neuropsychol. 1979; 1: 17–19. population. Clin Neuropsychol. 1997; 11: 258–265.
26. Lezak MD. The problem of assessing executive functions. Int J 45. Jersild AT. Mental set and shift. Archives of Psychology. 1927,
Psychol. 1982; 17: 281–297. Whole number 89.
27. Duffy JD, Campbell JJ III, Salloway SP, Malloy PF. Regional 46. Heaton RK, Chelune GJ, Talley JL, Kay GG, Curtiss G.
prefrontal syndromes: A theoretical and clinical overview. The Wisconsin Card Sorting Test Manual: Revised and Expanded.
frontal lobes and neuropsychiatric illness. Arlington, VA, US: Lutz: Psychological Assessment Resources, Inc.; 1993.
American Psychiatric Publishing, Inc.; 2001:113–123. 47. Reitan RM. Trail Making Test: Manual for administration,
28. Fink GR, Halligan PW, Marshall JC, et al. Neural mechanisms scoring, and interpretation. Indianapolis: Indiana University
involved in the processing of global and local aspects of Medical Center; 1958.
hierarchically organized visual stimuli., Brain (Vol. 120). Oxford: 48. PsyCor. Delis-Kaplan Exucutive Function System. San Antonio:
Oxford Univ Press, http:\\[Link]; 1997: 1779-1791. The Psychological Corporation; 2001.
29. Nielson KA, Langenecker SA, Garavan H. Differences in the 49. Demakis GJ. A meta-analytic review of the sensitivity of the
functional neuroanatomy of inhibitory control across the adult life Wisconsin Card Sorting Test to frontal and lateralized frontal
span. Psychology & Aging (Vol. 17). Washington, DC: American brain damage. Neuropsychology (Vol. 17). Washington, DC:
Psychological Assn, http:\\[Link]; 2002, 56-71. American Psychological Assn, http:\\[Link], 2003, 255-264.
30. Sylvester C-Y, Wagner TD, Lacey SC et al. Switching attention 50. Demakis GJ, Harrison DW. Relationships between verbal and
and resolving interference: fMRI measures of executive functions. nonverbal fluency measures: implications for assessment of
Neuropsychologia (Vol. 41). Amsterdam: Elsevier Science, http:\\ executive functioning. Psychol Rep. 1997; 81: 443–448.
[Link]; 2003, 357-370. 51. Reitan RM, Wolfson D. Category Test and Trail Making Test as mea-
31. Carlson SM, Wang TS. Inhibitory control and emotion regulation sures of frontal lobe functions. Clin Neuropsychol. 1995; 9: 50–56.
in preschool children. Cogn Dev. 2007; 22: 489–510. 52. Buss AH. Some determinants of rigidity in discrimination-reversal
32. Knolle-Veentjer S, Huth V, Ferstl R, Aldenhoff JB, Hinze-Selch learning. J Exp Psychol. 1952; 44: 222–227.
D. Delay of gratification and executive performance in individuals 53. Drewe EA. Go-no go learning after frontal lobe lesions in
with schizophrenia: Putative role for eating behavior and body humans. Cortex. 1975; 11: 8–16.
weight regulation. J Psychiatr Res. 2008; 42: 98–105. 54. Quinn SOB, Quinn EP. The Stroop Interference Effect in Young
33. Magar ECE, Phillips LH, Hosie JA. Self-regulation and risk- Children: A Developmentally Appropriate Approach. Ind Diff
taking. Personality Individ Differ. 2008; 45: 153–159. Res. 2005; 3: 183–187.
34. Osaka N. The frontal lobe correlates of metacognition: Neural 55. Fan J, McCandliss BD, Sommer T, Raz A, Posner MI. Testing the
interpretation based on the cognitive neuroscience of working efficiency and independence of attentional networks. J Cogn
memory. Jpn Psychol Rev. 2007; 50: 216–226. Neurosci. 2002; 14: 340–347.
35. Suchy Y, Kosson DS. State-dependent executive deficits among 56. Bechara A, Damasio H, Tranel D, Damasio AR. The Iowa
psychopathic offenders. J Int Neuropsychol Soc. 2005; 11: 311– Gambling Task and the somatic marker hypothesis: Some
321. questions and answers. Trends Cogn Sci. 2005; 9: 159–162.
36. Vohs KD, Baumeister RF, Schmeichel BJ, et al. Making choices 57. Newman JP, Patterson CM, Howland EW, Nichols SL. Passive
impairs subsequent self-control: A limited-resource account of avoidance in psychopaths: The effects of reward. Pers Individ
decision making, self-regulation, and active initiative. J Pers Soc Differ. 1990; 11: 1101–1114.
Psychol. 2008; 94: 883–898. 58. Newman JP, Schmitt WA. Passive avoidance in psychopathic
37. Posner MI, Dehaene S. Attentional networks. Trends Neurosci. offenders: A replication and extension. J Abnorm Psychol. 1998;
1994; 17: 75–79. 107: 527–532.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.