Psychology Internal Assessment
“Impact of introducing a delayed task on the recency effect in free
recall.”
Higher Level
Personal code: ksr244
Session: May 2024
Word Count: 2016
TABLE OF CONTENTS
INTRODUCTION......................................................................................................................... 3
EXPLORATION............................................................................................................................ 6
Design........................................................................................................................................ 6
Sampling & characteristics of participants................................................................................ 6
Controlled variables................................................................................................................... 6
Materials required...................................................................................................................... 7
Procedure................................................................................................................................... 7
ANALYSIS......................................................................................................................................9
EVALUATION............................................................................................................................. 11
BIBLIOGRAPHY........................................................................................................................ 15
APPENDICES.............................................................................................................................. 16
Appendix 1: WORD LIST....................................................................................................... 16
Appendix 2: BRIEFING NOTES............................................................................................ 17
Appendix 3: DEBRIEFING NOTES....................................................................................... 18
Appendix 4: CONSENT FORM..............................................................................................19
Appendix 5: RAW DATA TABLE...........................................................................................20
Appendix 6: MEAN CALCULATIONS..................................................................................21
Appendix 7: STD DEVIATION CALCULATIONS............................................................... 23
Appendix 8: MANN-WHITNEY U-TEST CALCULATIONS.............................................. 27
Appendix 9: CRITICAL VALUES TABLE............................................................................ 29
INTRODUCTION
In cognitive psychology, the Atkinson-Shiffrin model was first spoken about by Richard Atkinson
and Richard Shiffrin. This model is named after them but it is commonly called the Multi-Store
Model of Memory1, in 1968. The model significantly argues that human memory is inclusive of 3
distinct memory stores. Starting with the ‘sensory store’ which instead of processing the
information carried by the stimulus, detects and retains it for later use in short-term memory. It
comprises the Iconic memory (visual) and the echoic memory (auditory). The short-term store,
the working memory of the system, receives and holds input from the sensory store as well as the
long-term store upon recall. The amount of information that may be processed at once is
significantly constrained by short-term memory's limited capacity. Information decays rapidly if
it is not rehearsed well, and then the well-attended information transfers to the long-term
memory store. The long-term store is more or less permanent. It is possible to “copy”
information from this store and move it to the short-term store, where it can be processed and
altered and therefore is known as the process of retrieval. Atkinson and Shiffrin's multi-store
model (MSM) of memory was experimented with by researchers- Glanzer and Cunitz (1966)
which was also supported by the theory of the Serial Position Effect. This psychological
phenomenon is known as the “Serial Position Effect,” in which an individual better recalls the
words at the beginning and at the end of a list as compared to the words in the middle.2 This
theory was first introduced by Hermann Ebbinghaus. It is a type of cognitive bias that
1
Multi-Store Memory Model: Atkinson and Shiffrin. (2022). Retrieved 13 June 2023, from
[Link]
2
Serial Position Effect. (n.d.). Retrieved from [Link]
encompasses both the Primacy effect (first few words are better recalled) and the Recency effect
(last few words are better recalled)3.
Diagram 1: Multi Store Model of Memory (MSM)4
The study of Glanzer and Cunitz (1966)5 was based on the Multi-Store Model of Memory and the
Serial position effect. The short and long term memory was investigated as two independent
memory stores. The study has three conditions: Condition 1: Immediate recall after memorizing
the words (Immediate free recall) Condition 2: the subjects were presented with a filler activity
of counting backward for 30 seconds and then recalling (delayed free recall). Condition 3: A
similar delayed recall condition but with 10 seconds of filler activity (delayed recall). The
researchers found that while the primacy effect remained the same, the recency effect was
significantly removed in both the delayed free recall conditions.
3
Serial Position Effect (Example + Definition) - Practical Psychology. (2019). Retrieved 13 June 2023, from
[Link]
4
Memory - Psychology A Level. (2023). Retrieved 13 June 2023, from
[Link]
5
Dixon, T. (2018). Key Study: The Primacy and Recency Effects (Glanzer and Cunitz, 1966) | IB Psychology.
Retrieved 13 June 2023, from
[Link]
fects/
We aimed to investigate whether a filler task would impair free recall and influence the recency
effect.
The Multi-Store model of Memory holds much significance even today, especially in the learning
and education sector. Strategically, it is useful for students as teachers use this as a tool in the
form of repetition or rehearsal to allow the transfer of concepts from the student’s STM to LTM.
The Serial position effect is relevant in marketing schemes when designing advertisements and
placing key elements at the beginning or end to create a greater impact.
Null hypothesis (H0): The number of words that are correctly recalled will not be significantly
affected by the delayed recall.
Research hypothesis (H1): There will be no recency effect as the filler activity will significantly
lower the number of words recalled.
Independent variable: Filler activity
Dependent variable: Number of words correctly recalled
EXPLORATION
Design
The independent measures design was used for the experiment, through which demand
characteristics were restricted and the order effect, such as the practice and fatigue effect, was
avoided by using a separate group of participants for each condition. Instead, if a repeated
measures design is used where the participants are exposed to both conditions, it would cause
them to guess the aim of the experiment.
Sampling & characteristics of participants
A total of 30 students were recruited from the 11th-grade physics HL class through a
convenience sampling technique which was an easier, quicker, and fairer method. The sample
comprised 19 males and 11 females, who were then equally divided into the two conditions-
Delayed free recall and Immediate free recall- using the random team generator software. The
participants were from the 16-18 age group, with a mean age of 16.9 years. All participants were
non-psychology students, had bilingual proficiency in the English Language, and had good
eyesight. This criterion was set up to eliminate any kind of disparity between the participants’
way of assimilation of the English words which would affect their rate of recall.
Controlled variables
Certain controlled variables in the experiment included the quantity of words and the type of
words (two-syllable) on the word list so that both conditions were treated equally without any
extraneous variables affecting their results. Moreover, the time given to recall the words (30
seconds) and the type of filler task (drawing geometrical figures) were also controlled during
the experiment. The same briefing & debriefing notes and the consent form was given to all
subjects.
Materials required
A random team generator software6 was used for dividing the sample into two conditions to
remove any kind of bias and to ensure that the participants were fairly distributed. To select
words for our wordlist, we used the site random word [Link].7 The words were
generated using certain fillers, such as the number of letters in each word (less than 10) and the
category of words (a mix of adjectives and nouns). The wordlist was presented using an Ariel
font of size 13. Additionally, 30 pencils and blank sheets of paper for all subjects were needed.
A stopwatch was an essential item to time the participants while reading the words and then for
the recall activity.
Procedure
All participants were aptly briefed8 and were then asked to sign the consent form.9 Condition A
(immediate free recall) was presented with a list of 15 words10 for 30 seconds. They were then
immediately asked to note down all the words they could remember on the blank sheet. While
6
Random Team Generator: Split list of names in random groups. (2023). Retrieved 13 June 2023, from
[Link]
7
word?, W., word?, W., word?, W., & generator?, I. (2023). Random Word Generator. Retrieved 13 June 2023, from
[Link]
8
Refer to Appendix 2
9
Refer to Appendix 4
10
Refer to Appendix 1
participants of condition B (delayed free recall), were given a filler activity right after the
wordlist was shown to them. They were asked to draw any geometrical figures, either 2D or 3D,
that they knew of, on the sheet of paper (for 30 seconds). After this, they recalled and wrote
down as many words as they could. All the answer sheets were then collected followed by
debriefing.11 They were all thanked for their contribution to the experiment.
11
Refer to Appendix 3
ANALYSIS
According to Table 1, the results are per our original hypothesis, i.e. more words are correctly
recalled by the participants in the condition with immediate free recall than by the participants in
the condition with delayed free recall that performed a distraction task. The mean number of
words12 correctly recalled by the delayed recall condition was 7.73, which is lower than the
immediate recall condition which had a mean of 11.13. Therefore, showcases the negative effect
that a filler activity task has on the ability to recall. The standard deviation scores13 are lower
for the immediate recall condition with a value of 2.02 than for the delayed recall condition with
a value of 2.35. Additionally, there were no outliers found.
Table 1: Mean and standard deviation values for delayed and immediate free recall conditions-
Conditions DELAYED FREE RECALL IMMEDIATE FREE
CONDITION RECALL CONDITION
Mean number of words 7.73 11.13
correctly recalled
Standard deviation 2.35 2.02
12
Refer to Appendix 6
13
Refer to Appendix 7
Graph 1: Mean values of 15 words for delayed free recall and immediate free recall conditions-
The Mann-Whitney U test was completed to further validate the independent measures design
used in the experiment, which was done by testing the statistical significance of our findings.
Using a one-tailed hypothesis, the investigation’s ratio data was changed to ordinal data, and a
final U value of 32.5 was obtained.14 The value is lower than the critical regions of 72 and 56 at
0.05 and 0.01 levels respectively.15 Thus, the research hypothesis- stating that the rate of recall of
the last few words on the list would be significantly reduced by the filler activity, will be
accepted while rejecting the null hypothesis.
14
Refer to Appendix 8
15
Refer to Appendix 9
EVALUATION
Comparison with Background Theory
The findings corroborate with the Multi-Store Model of Memory.16 Participants recalled more
words in the immediate free recall condition as the words were active in their STM. However, in
the delayed free recall condition, the filler activity delay of 30 seconds displaced the information
from their STM and thereby didn’t transfer to the LTM. The significance of these results explains
how the delay task influenced the recency effect. Therefore, the findings of this study are well
explained by the MSM which concludes that the filler activity had impaired the maintenance
rehearsal of the the participants in the delay recall condition.
Comparison with the original study
According to the original study’s findings, the recency effect was eliminated in both of the
delayed-free recall conditions due to the filler task.17 Our statistical findings match those of the
original study that had a higher average of recalled words in the immediate free recall condition
(11.3 in the replicated study) and lower by the delayed free recall condition (7.73 in the
replicated study).
16
Dixon, T. (2021). The multi-store model of memory (Atkinson and Shiffrin, 1968) | IB Psychology. Retrieved 13
June 2023, from
[Link]
1968/
17
(2023). Retrieved 13 June 2023, from
[Link]
df
Experimental design
Using an independent measures design was beneficial for the investigation as it eradicated any
demand characteristics such as practice or fatigue effect which would impact our results. On the
contrary, this experimental design doesn’t control individual variability. To rule out this problem,
a matched pair design could be employed in the study. Matching participants based on certain
variables like age and gender can be a great way to control confounding variables and to obtain a
more reliable comparison between both conditions.
Sampling technique
It was ensured that all subjects had English language proficiency so the understanding of the
word list wouldn’t be hampered. Their English levels were similar thus increasing the credibility
of the results. However, we failed to take into consideration other important characteristic
requirements that might influence the rate of recall. For instance, if a participant had a learning
disability, it might take them longer to assimilate information (words) or pay enough attention to
the word list than the other participants. This should have been avoided by mentioning another
sampling criterion which would rule out anybody that would have engaged with the LRC
(Learning Resource Centre), as they are more likely to be diagnosed with such learning
disabilities.
Procedure
Our study’s procedure was strong as a legible font, size, and color were used to present the word
lists to both the conditions from the same presentation. This ensured no unfair advantage was
given to any participants which may influence their recall rates. An equal amount of time was
given to both the conditions which further validates the results. Although most variables were
controlled in the procedure, we left blank sheets of paper and pens on the tables where they were
evident to the participants throughout the presentation of the word list. This may have given off
some hints about the actual aim of the investigation and might have caused the subject’s attention
to deviate for the short period when the word lists were presented. To improve on this, the paper
and pens should have been kept inside the tables until instructed to use. Furthermore, this would
also save the time used to distribute them.
Materials
The words utilized were not related to physics keeping in mind our target population because that
would have made it easier for the participants to memorize and recall which would have
manipulated our results leading to the unreliability of the findings. This wordlist used by the
replicated study was different from the one used by Glanzer and Cunitz (1966). We used words
that are more commonly used by people, such as “sweet” or “sour”, which reduced the difficulty
level of remembering the words and may have impacted the results. It would have been better if
the word list comprised equally difficult (“homogeneous”) and easy (“love”) words.
Therefore, our findings support the hypothesis that the introduction of a filler task would reduce
the number of words accurately remembered by the delayed free recall condition.
BIBLIOGRAPHY
● Atkinson and Shiffrin Model of Memory (Multi-Store Model). (2019). Retrieved 13 June
2023, from [Link]
● Coolican, H. (2019). Research methods and statistics in psychology (7th ed.). Routledge.
● Dixon, T. (2018). Key Study: The Primacy and Recency Effects (Glanzer and Cunitz, 1966) |
IB Psychology. Retrieved 13 June 2023, from
[Link]
primacy-and-recency-effects/
● Dixon, T. (2021). The multi-store model of memory (Atkinson and Shiffrin, 1968) | IB
Psychology. Retrieved 13 June 2023, from
[Link]
-atkinson-and-shiffrin-1968/
● Glanzer and Cunitz (1966) Study Full [Link]. (2023). Retrieved 13 June 2023, from
[Link]
2licHN5Y2h8Z3g6NjcxZTIzNGZhZjljNzljMg
● Memory - Psychology A Level. (2023). Retrieved 13 June 2023, from
[Link]
● Serial Position Effect (Example + Definition) - Practical Psychology. (2019). Retrieved 13
June 2023, from [Link]
● (2023). Retrieved 13 June 2023, from
[Link]
erial_position_effect.pdf
● Heiman, G. W. (2001). Understanding research methods and statistics: An integrated
introduction for psychology. Houghton Mifflin.
APPENDICES
Appendix 1: WORD LIST
Serial no. Words
1 Happy
2 Double
3 Cherry
4 Sunday
5 Sweet
6 Glass
7 Heart
8 Sour
9 Brain
10 Women
11 White
12 Summer
13 Power
14 Breakfast
15 Ring
Appendix 2: BRIEFING NOTES
Greetings participants,
In today’s investigation, a random word list will be presented before all of you for a short
controlled period. Following that, all of you will be asked to draw any kind of geometrical
figures, 2D or 3D on the sheet of paper that would later be provided, for exactly 30 seconds. All
of you will then be tested through a recall activity, as we calculate the number of words that are
correctly recalled, irrespective of the order they were presented in.
All participants are thereby allowed to withdraw from the experiment at any point in time. If
anyone wishes to know their results after the experiment, they are free to contact us at this email
address. We want to ensure that everybody is comfortable and ready to begin with the
experiment. You may ask any questions or queries before we proceed.
Thank you.
Appendix 3: DEBRIEFING NOTES
We would like to thank you for agreeing to be a participant in our study which is now over. In
this study, we wanted to investigate whether a delayed recall would affect the number of words
you recalled due to the theory of primacy and recency effect. Simply put, we wanted to study the
effect of the interference task, which was to draw as many geometric shapes as possible on a
piece of paper, on the number of words you would be able to remember. The interference task
was solely just to keep your mind occupied to avoid you from running the words through your
head during the thirty seconds before you were asked to note down as many of them as you could
remember.
Kindly contact us at the email addresses written on the smart board, in case anyone has any
queries or would like to gain an insight into their results.
Appendix 4: CONSENT FORM
Name: ___________________
Age: ______
Sex: Male Female Other
● The protection of participants will be ensured, both physically and mentally.
● Personal information of participants would be kept confidential and their identities would
not be revealed.
● All participants have full rights to withdraw from the study.
● Participants are free to contact the researchers to know their results after the study.
● Participants would be briefed and debriefed before and after the study respectively.
I thus accept to participate in the study as per the guidelines mentioned above and provide my
consent.
Signature: ______________ Date: ______________
Appendix 5: RAW DATA TABLE
Number of words correctly Number of words correctly
Participants recalled (Delayed free recall recalled by (Immediate free
condition18) recall condition19)
1 9 10
2 7 12
3 9 12
4 4 10
5 6 9
6 6 10
7 5 14
8 9 7
9 10 13
10 12 12
11 9 8
12 7 13
13 11 14
14 4 12
15 8 11
TOTAL 116 167
18
(N.d.). Retrieved from
[Link]
df
19
(N.d.). Retrieved from
[Link]
df
Appendix 6: MEAN CALCULATIONS
Amount of words correctly Amount of words correctly
Participants recalled by (Delayed free recall recalled by (Immediate free
condition20) recall condition21)
1 9 10
2 7 12
3 9 12
4 4 10
5 6 9
6 6 10
7 5 14
8 9 7
9 10 13
10 12 12
11 9 8
12 7 13
13 11 14
14 4 12
15 8 11
TOTAL 116 167
MEAN 7.73 11.13
20
(N.d.). Retrieved from
[Link]
df
21
(N.d.). Retrieved from
[Link]
df
Mean=total sum/total number of participants22
Mean for delayed free recall= 116/15
Mean for delayed free recall= 7.73
Mean=total sum/total number of participants
Mean for immediate free recall= 167/15
Mean for immediate free recall= 11.13
22
Mean - Formula, Meaning, Definition | How to Find Mean?. (2023). Retrieved 13 June 2023, from
[Link]
Appendix 7: STD DEVIATION CALCULATIONS
Table 2: Standard deviation for Delayed free recall condition-
Participants Words recalled by Delayed (X1-M1)= D1 (D1)²
free recall condition
1 9 9-7.73= 1.27 1.61
2 7 7-7.73= -0.73 0.53
3 9 9-7.73= 1.27 1.61
4 4 4-7.73= -3.73 13.91
5 6 6-7.73= -1.73 2.99
6 6 6-7.73= -1.73 2.99
7 5 5-7.73= -2.73 7.45
8 9 9-7.73= 1.27 1.61
9 10 10-7.73= 2.27 5.15
10 12 12-7.73= 4.27 18.23
11 9 9-7.73= 1.27 1.61
12 7 7-7.73= -.073 0.53
13 11 11-7.73= 3.27 10.69
14 4 4-7.73= -3.73 13.91
15 8 8-7.73= 0.27 0.07
TOTAL 116 82.89
23
X1 is each data point
M1 is the mean value
N is the participant number
σ is the value of the standard deviation for Delayed free recall condition24
√82.89/15= 2.35 (Delayed free recall condition)
23
Standard Deviation - Formula | How to Calculate Standard Deviation?. (2023). Retrieved 13 June 2023, from
[Link]
24
Heiman, G. W. (2001). Understanding research methods and statistics: An integrated introduction for psychology.
Houghton Mifflin.
Table 3: Standard deviation for Immediate free recall condition-
Participants Words recalled by- (X1-M1)= D1 (D1)²
Immediate free recall
condition
1 10 10-11.13= -1.13 1.27
2 12 12-11.13= 0.87 0.75
3 12 12-11.13= 0.87 0.75
4 10 10-11.13= -1.13 1.27
5 9 9-11.13= -2.13 4.53
6 10 10-11.13= -1.13 1.27
7 14 14-11.13= 2.87 8.23
8 7 7-11.13= -4.13 17.05
9 13 13-11.13= 1.87 3.49
10 12 12-11.13= 0.87 0.75
11 8 8-11.13= -3.13 9.79
12 13 13-11.13= 1.87 3.49
13 14 14-11.13= 2.87 8.23
14 12 12-11.13= 0.87 0.75
15 11 11-11.13= 0.13 0.01
TOTAL 167 61.63
X1 is each data point
M1 is the mean value
N is the participant number
σ is the value of the standard deviation for Immediate free recall condition25
√61.63/15= 2.02 (Immediate free recall condition)
25
Heiman, G. W. (2001). Understanding research methods and statistics: An integrated introduction for psychology.
Houghton Mifflin.
Appendix 8: MANN-WHITNEY U-TEST
CALCULATIONS
Participants Words recalled by- Delayed Rank (a) Words recalled by- Rank (b)
recall group Immediate recall group
1 9 13 10 17.5
2 7 7 12 24
3 9 13 12 24
4 4 1.5 10 17.5
5 6 4.5 9 13
6 6 4.5 10 17.5
7 5 3 14 29.5
8 9 13 7 7
9 10 17.5 13 27.5
10 12 24 12 24
11 9 13 8 9.5
12 7 7 13 27.5
13 11 20.5 14 29.5
14 4 1.5 12 24
15 8 9.5 11 20.5
TOTAL 116 152.5 167 312.5
U represents the Mann-Whitney value.
N1 is the calculated number of participants in the delayed-free recall condition.
N2 is the calculated number of participants in the immediate free recall condition.
R(a) is the number of ranks in the delayed-free recall condition.
R(b) is the number of ranks in the immediate free recall condition26
DELAYED FREE RECALL CONDITION:27
U(a) = N1 ×N2 + ([N1{N1+1}]/2) - R(a)
U(a) = 15 ×15 + ([15{15+1}]/2) - 152.5
U(a) = 192.5
IMMEDIATE FREE RECALL CONDITION:
U(b) = N1 ×N2 + ([N1{N1+1}]/2) - R(b)
U(b) = 15 ×15 + ([15{15+1}]/2) - 312.5
U(b) = 32.5
The smaller value of 32.5 would be taken. Therefore the final value is U=32.5
26
Coolican, H. (2019). Research methods and statistics in psychology (7th ed.). Routledge.
27
(N.d.). Retrieved from
[Link]
df
Appendix 9: CRITICAL VALUES TABLE28
28
Mann Whitney U Test (Wilcoxon Rank Sum Test). (2023). Retrieved 13 June 2023, from
[Link]