Replication of Stroop Effect Experiment

This assignment involves conducting a replication study of the Stroop effect, a well-known experiment in cognitive psychology. The study aims to investigate how conflicting stimuli (in this case, an incongruent wordlist) affect response time. Participants are presented with either a congruent or incongruent wordlist and their reaction times are measured. The assignment includes consent forms for participants, details on the experimental design, and references to relevant studies in cognitive psychology.

Psychology Internal Assessment (HL)
An experiment to investigate if conflicting stimuli would affect
the reading response time for international
high school students
Candidate name : Raja Aiman Ariff
Candidate number : glq227
Subject : Psychology –HL
Date of Submission : 1st November 2017
Word count : 1727 words

Abstract
This replication study is on ‘The Effect of Interfering Color Stimuli Upon
Reading Names of Colors Serially’, which had been carried out by the original study
of Stroop (1975)1. The aim of experiment is to investigate if the conflicting stimuli
would affect the response time of international high school students in Malaysia.
The participants were selected using opportunity sampling. The participants were
exposed different conditions from the two wordlist given. A black wordlist
associated with a black ink (congruent) and color wordlist with a different coloured
ink (incongruent); conflicting stimuli. The independent variable for this study is the
congruent wordlist (black). For the dependent variable is the response time of the
participants in seconds. This study is a repeated measure design and applies a
counter-balancing technique to negate the order effects presented in participants.
Resulting with a mean of 9.53 seconds for Condition A (congruent) and 10.97
seconds for Condition B (incongruent). As for the standard deviation, we obtained
1.63 seconds for Condition A and 1.99 seconds for Condition B. The experiment is
then validated for its significance using the Wilcoxon signed-match rank test and it
was found that the difference between the mean response-time was 0.0434, below
than (P<0.05) to be significant. This further supports our research hypothesis when
Condition A has longer response time in contrast to Condition B, which has a shorter
response time. Therefore, the results from our study led to rejecting the null
hypothesis and supporting Stroop’s (1935) original study.
1 Stroop, J, "Studies of Interference in Serial Verbal Reactions", Journal of Experimental Psychology:
General, Vol. 121, no. 1, 2017, pp. 15-23.

Contents Page
Introduction.............................................................................................................................. 4
B: Method : Design.................................................................................................................... 5
C: Method : Participants.......................................................................................................... 6
D: Method : Procedure............................................................................................................. 6
E-Results: Descriptive.............................................................................................................. 7
F-Results: Inferential............................................................................................................... 8
G: Discussion............................................................................................................................. 8
Appendices.............................................................................................................................. 10
Bibliography........................................................................................................................... 17

Introduction
Perception2 is a cognitive progress in the brain that looks at conscious mental
activities in performing the act or the process of knowing. Cognition comes
alongside with the information-processing theory, which consists of two
subsidiaries, that is serial processing (the theory of scanning an individual article or
unit one at a time) and automatic processing (processing multiple items at a time).
These processes can be interfered by disturbances within a stimuli-causing stress
that is induced by receiving from too much information and the reduction of
attention. This disruption process is known as the 'Stroop Effect'.
John Ridley Stroop, an American physiologist that had conducted the Stroop Effect
(1935)3 study that was profound enough to be relevant until the 21st century. The
study focuses on the interference in serial verbal reactions and one of his
experiments in the study looks at ‘The Effect of Interfering Color Stimuli Upon
Reading Names of Colors Serially’4. His aim was to see whether conflicting stimuli
affects the time of response when participants were given a reading task. The
procedure was set up with repeated measure design tasks consisting of a ‘word test’
(congruent) and a ‘color test’ (incongruent). There were 70 participants who were
college undergraduates (14 males and 56 females). The following conditions were
administered to the study: Firstly, the participants faced the congruent test
(Condition A) which they had to read a list of words that were printed in black ink.
Secondly, the participants faced the incongruent test (Condition B) which the words
mismatched the following coloured ink. As for the results, the participants that faced
the incongruent test took a longer a time to name the colour, which was 2.3 seconds
longer than the Condition A. The reason behind why Condition B participant
2 A Law, Pearson International Baccalaureate Diploma: International Editions, in , 1st ed., Hove, Pearson
Education Limited, 2017, p. 66.
3 Stroop, J, "Studies of Interference in Serial Verbal Reactions", Journal of Experimental Psychology:
General, Vol. 121, no. 1, 2017, pp. 15-23.
4 Ibid.

responded slower was due to the conflicting stimuli (incongruent words and colour)
that had affected the cognition process and the verbal response5.
Robert Conrad (1964)6 operated an experiment that focuses on to recode sensory
information into language, and more specifically on verbal (spoken) language. The
subjects were presented with strings of visual letters from the Brown-Peterson task.
Prior to recalling the memory set, they were asked to perform an interference task
for a short period before the recalling process begins. Throughout the study, Conrad
assessed on whether the letter recalled was the one confused more often with the
visual or the acoustic form of that letter; which was done in the preparatory phase.
The subjects recoded the letters verbally from the given letters they receive visually
although acoustic confusions were intruded. This suggested that STM only
contained verbal representations. Theorists concluded that autonomous mechanism
associates with STM when storing verbal representations of recently perceived
stimuli.
Aim: To investigate if the conflicting stimuli would affect the response time of
international high school students.
Experimental hypothesis: Speed of reading will be shorter for congruent condition
compared to incongruent condition.
Null hypothesis: There will be no significant difference in the delay in the time of
responses (seconds or milliseconds). Any significant differences that occurred are
all by chance.
5 Ibid.
6 L Barsalou, Cognitive Psychology: An Overview for Cognitive Scientists, in , 1st ed., Hove, Psychology
Press, 2014, pp. 99-100, <https://books.google.com.my/books?
id=3kbrAgAAQBAJ&printsec=copyright&source=gbs_pub_info_r#v=onepage&q=conrad&f=false>
[accessed 11 October 2017].

B: Method : Design
The method applied upon the study comprised of both conditions, which will be
placed into comparison. The suitable experimental method for this study is through
repeated-measures design (RMD). This method allows us to assess the differences
and the effectiveness of both conditions conducted in the study7. This is to take into
account for the independent differences in reading time. When conducting the
experiment, we need to further look at other factors that puts into consideration for
example; repeated exposure, which is where a participants gets used to their
objective (reading task) and performs better recitation of the words. Therefore,
counter-balancing was applied.
The participants were chosen through opportunity sampling so the participants
carried out in this study were subjects who are available at that time. Before the
experiment begins, we clearly read aloud the standardised briefing instructions to
each of them. Then they were are asked to fill out the consent form consisting
essential details like age and gender so that they can partake in this experiment. The
confidentiality of their results must be kept hidden and are able to look at them
anytime when the experiment has ended. The following materials we used were a
stopwatch, to record the response-time for both of the conditions in the two
wordlists. Misreading the incongruent list receives an additional 2 seconds to their
overall results.
Independent variable (I.V): Changed of the colour words of the incongruent
wordlist or the congruent wordlist.
7 Saul, M, "Experimental Design", Simply Psychology, 2017.

Dependent variable (D.V): Measuring the time in seconds for participants to state
the colour words.
C: Method : Participants
The target population in the study looks at students from an international college.
The participants were ranged from the ages of 16-19 within both of the genders.
We’ve gathered results from 11 participants consisting 6 females and 5 males. The
participants must have the basic knowledge of English, to ensure that the timing of
responses are not affected when verbally read out.
D: Method : Procedure
For the oppurtunity selection process, we come up to any student in a library and
asked if they wanted to participate in our experiment. After agreeing to participate,
they are walked and seated in a quiet room; this is to avoid any environmental
distractions. The materials we used to note down all the participants timing-
response is typed into a Google Spreadsheet (Network Cloud Platform) using a
computer.
Procedure:
1)Prepare all the recording materials and the recommended materials such as the
incongruent8 and the congruent9 wordlists.
2)Read out the introductory standardised briefing10 on the procedure and the
ethical concerns of the experiment.
8 Appendix IV
9 Appendix V
10 Appendix V

3)Then the consent form11 was given to the participants on the agreement of
participating to this study.
4)The lists is given out at random so it is assosciated through a labelled coin toss for
each given list that is handed out to the participants.
5) Prepare a stopwatch and a recording sheet.
6) Record the timing of the responses and jotted it down in seconds.
7) After reading out the first list, let them rest for 10 seconds.
8) Proceed with the other list and record the timing of the response.
9) Lastly, read out the debrief12 on the aims and the general hypothesis of the
experiment which is looking at if conflicting stimuli had an effect on their responses.
The total recorded time of each list was measured in seconds. Lastly, the participant
were then given a debrief on the aims and the general hypothesis of the experiment
that relates to see if conflicting stimuli had an effect on their responses.
11 Appendix VI
12 Appendix VII

E-Results: Descriptive
Graph 1: Mean of data of both conditions
Looking at the bar chart, we observe that the results supports our hypothesis where
Condition B shows higher delays than the Condition A. This suggest that conflicting
stimuli affects the response time of the participant.
Conditions Condition A Condition B
Mean 9.526363636 10.96909091
Standard Deviation 1.991825567 1.632894808
Table 2: Standard deviation of both conditions
In the dispersion of the data, we can see that average and the standard deviation of
the results from both conditions. The ratio level of measurement is under time
which is seconds. The mean for congruent wordlist ≈9.53 seconds and for the
incongruent wordlist ≈10.97 seconds. The average supports the research hypothesis
B (Incongruent Wordlist)A (Black Wordlist)

showing participants in Condition B to have a higher response time than Condition
A.
The standard deviation shows the dispersion of the collected results for each
condition. The standard deviation for congruent wordlist is ≈±1.992 and for the
incongruent wordlist is ≈±1.633. It shows that Condition A has more variance
between each results that were collected.
F-Results: Inferential
Observing to see if there is a significance in both of the conditions, we used the
Wilcoxon Signed Ranked Test. I chose to acquire this statistical test for my data as it
looks for significant difference. Since it is designated for repeated measures
experiments, we can say that it can be converted into an ordinal level from an
interval level of data. Gaining the data for a repeated measure design using this
statistical test would be more suitable for the relevant data recorded in this study.
In the Wilcoxon Signed Rank test table13, there are 11 pairs of both conditions from
the 11 participants that we gathered. Since this is a P(2-tail) test, we focus on
looking if the p-value is lower than 0.5 (P<0.05) because any value lower than that
point makes the data significant. We’ve obtained 0.0434 as our p-value, resulting
our data to be significant. Hence, suggesting to disregard the null hypothesis and
accepting the research hypothesis.
G: Discussion
13 Appendix II

Initially, our overall results showed similarities with Stroop’s (1935)14 and supports
our experimental hypothesis of displaying delays in all of their responses with the
conflicting stimuli.
In the descriptive and inferential results, there was some clear differences present
in-between of having the mean 9.53 and 10.97 seconds for the black and
incongruent words. Then as for standard deviation, we retrieved 1.99 and 1.63
seconds. Results that were obtained from the Wilcoxon signed-matched ranks test
creates the data to be substantial in a sense of having it to reach significant at a level
of p>0.05 encouraging to reject the null hypothesis and retain the experimental
hypothesis. Hence, supporting Stroop’s (1935)15 that the response time for the
incongruent wordlist will have an increase compared to the congruent wordlist.
The information processing theory along with serial processing, and automatic
processing can interpret why participants experience a conflicting stimuli during
the experiment. This indicates that participants were unable to automatically
process the words for the incongruent wordlist due to the exposure of a conflicting
stimulus that was presented. This therefore had a higher mean in the response time
than the black wordlist.
In conclusion, the mean response time for Condition B was greater than Condition A.
Through the use of the Wilcoxon Test Statistics, the results suggest the difference in
mean response times between the conditions was statistically significant, therefore
encouraging to disregard the null hypothesis and sustain with the experimental
hypothesis. This study supports the hypothesis and the results from the original
study of Stroop (1935)16. This study could support ‘Stroop effect’ much further if we
14 John Stroop, "Studies of Interference in Serial Verbal Reactions", Journal of Experimental Psychology:
General, Vol. 121, no. 1, 2017, pp. 15-23.
15 Ibid.
16 John Stroop, "Studies of Interference in Serial Verbal Reactions", Journal of Experimental Psychology:
General, Vol. 121, no. 1, 2017, pp. 15-23.

conducted different test such as changing the font size or the objects when carrying
out further research.

Appendices
Appendix I
Condition A Condition B (color)
7.9 10.43
9.56 13.36
7.43 9.91
11.78 11.62
7.9 8.56
8.8 9.4
10.15 10.69
10.2 8.45
8.35 15
12.46 12.03
10.26 11.21
Table 1: Raw data of each participants reading out both of the conditions; in seconds
What we’ve collected from the raw data, we can see that there are some anomalies.
The anomalies were due to participants slow reading, and faster results in Condition A
(Blacklist).

Appendix II
Image 1: Wilcoxon Matched Pairs Sign Rank Test applied on the results taken from the
experiment

Appendix III

Appendix IV
Blue
Red
Green
Orange
Purple
Red
Brown
Blue
Green
Pink
Orange
Purple
Red
Brown
Blue
Green
Orange
Brown
Pink
Red
Red
Brown
Blue
Pink
Green

Appendix V
Red
Purple
Brown
Blue
Pink
Purple
Green
Brown
Red
Blue
Purple
Green
Blue
Brown
Red
Green
Blue
Pink
Brown
Red
Purple
Green
Blue
Brown
Pink

Appendix VI
[Stroop Effect] Consent to take part in research
I............................................. voluntarily agree to participate in this research study.
I understand that even if I agree to participate now, I can withdraw at any time or refuse to
answer any question without any consequences of any kind.
I understand that I can withdraw permission to use data from my interview within two weeks
after the interview, in which case the material will be deleted.
I understand that participation involves...[outline briefly in simple terms what participation in
your research will involve].
I understand that I will not benefit directly from participating in this research.
I understand that all information I provide for this study will be treated confidentially.
I understand that in any report on the results of this research my identity will remain
anonymous. This will be done by changing my name and disguising any details of my
interview which may reveal my identity or the identity of people I speak about.
I understand that signed consent forms will be retained in [specify location, security
arrangements and who has access to data] until [specific relevant period – for students this
will be until the exam board confirms the results of their dissertation].
I understand that under freedom of information legislation I am entitled to access the
information I have provided at any time while it is in storage as specified above.
I understand that the researchers will debrief me regarding the experiment after it is over
Signature of participant

Appendix VII
Thank you for partaking in our replication experiment! In simple terms, the
experiment aimed to see if conflicting stimuli, in our case the incongruent wordlist
would have an effect on response time. The replication study is a partial
reproduction of Stroop (1935).
Condition A represented the congruent wordlist in the form a black wordlist, and
condition B represented the incongruent wordlist. A time difference between the
two conditions should be present due to the idea of automatic processing being
disrupted when it is presented with conflicting stimuli, as more concentration is
required. The replication study is a partial reproduction of Stroop (1935).
If you would like to withdraw your results please feel free to do so, there will be
no penalties.
Bibliography
1) Barsalou, L, Cognitive Psychology: An Overview for Cognitive Scientists. in , 1st ed.,
Hove, Psychology Press, 2014, pp. 99-100, <https://books.google.com.my/books?
id=3kbrAgAAQBAJ&printsec=copyright&source=gbs_pub_info_r#v=onepage&q=conrad
&f=false> [accessed 11 October 2017].
2) Dyer, F. ‘The Stroop phenomenon and its use in the study of perceptual, cognitive,
and response processes’. Memory and Cognition, vol. 1, no.2, 1973, pp 106-120.
3) Law, A, Pearson International Baccalaureate Diploma: International Editions. in , 1st
ed., Hove, Pearson Education Limited, 2017, p. 66.
4) Stroop, J, "Studies of Interference in Serial Verbal Reactions", Journal of Experimental
Psychology: General, Vol. 121, no. 1, 2017, pp. 15-23.
5) Sternberg, S, "MEMORY SCANNING: MENTAL PROCESSES REVEALED BY REACTION-
TIME EXPERIMENTS", American Scientist, no. 4, 1969, pp. 421-457.

6) Saul, M, "Experimental Design", Simply Psychology, 2017.