Investigating the Stroop Effect

Title: Investigating the Stroop Effect: How the Congruency of Stimuli Affects

Response Time

Course: 175.102 Psychology as a Natural Science

Assignment: Lab Report Part 1A

Name: Anika O’Connell-Temple

Student Number: 20023303

Student Email Address: [email protected]

Degree Programme: Bachelor of Arts

Lecturer: Dr. Corinne Bareham-Waldrock

Submission Date: 11:55PM, 7th May 2021

Word count: 2124

Abstract

Research was conducted in order to support Stroop’s theory that the brain’s response

time is hindered when it is forced to deal with conflicting stimuli. The experiment

aimed to demonstrate that words presented in the congruent condition would be faster

to categorise and fewer mistakes would be made within the congruent condition when

compared to the incongruent condition. A total of 354 participants partook in the

experiment that employed a repeated measures design to test congruent and

incongruent conditions. Volunteers were shown three lists of words, including a test

list which consisted of X’s instead of colour names, and asked to identify the colour

of the ink used to write the words whilst ignoring the words themselves, which

consisted of colour names that either matched the ink colour or didn’t. The variables

being measured were the time participants took to respond to congruent and

incongruent lists, as well as the accuracy of their responses. The results showed that

the list of congruent words was both faster to respond to and provided more accurate

responses, while the list of incongruent words presented more variability within

participants responses. Thus, since both hypotheses were proved to be correct, the

research conducted also corroborates with Stroop’s original theory about how the

congruency of stimuli affects response time.

Investigating the Stroop Effect: How the Congruency of Stimuli Affects

Response Time

The understanding that words are faster to read than objects are to name is

fairly common in experimental psychology. A study executed by Cattell (1886)

highlighted this, it exemplified that the cognitive correlation between a word and the

idea of that word is automatic due to the frequency of its use. Stroop (1935) later

elaborated on this theory with his colour-word test, a neuropsychological exam that

tests the ability to prevent cognitive interference, which happens when the processing

of a stimulus affects the simultaneous processing of another attribute of said stimulus.

His test consisted of naming the colour of the ink in which the name of another colour

is written. The extra time taken to respond to this type of stimulus where the ink

colour does not match the written word (known as incongruent), compared to the one

used in naming the colour of the control stimuli (for example, groups of asterisks or

coloured shapes), is what is called the Stroop effect.

The traditional and most accepted explanation since Stroop's own formulation

to date has been the speed of processing theory. This theory states that because

reading is faster than naming the colour, the incorrect response will be faster than the

correct one and will interfere with it, resulting in a delay when trying to name the

colour of the ink (Fraisse, 1969). Another possible explanation is known as

automaticity, which implies that recognising and naming colours is not an automated

task whereas reading is, therefore resulting in more attention needed to name a colour

(Macleod and Dunbar, 1988). 

An additional explanation known as the perceptual-encoding account was

presented by Hock and Egeth (1970). Its premise was that incompatible information

from a colour word as opposed to a neutral control delays perceptual encoding of ink-
 Investigating the Stroop Effect

colour information. However, this theory was objected to by Dyer (1973) who argued

that their conclusions depended on accepting a null hypothesis and the theory has

since been pushed aside.

In its traditional form, the stimuli used in the task were always names of

colours written in a different colour ink, and its response time was compared with that

of a control group (Stroop, for example, used coloured squares). One of the most

important modifications of this original version was made by Klein (1964), whose

fundamental contribution varied the type of words used to manipulate the interference

effect. In his most well-known experiment he used four colours (red, green, yellow

and blue) with which he coloured the following types of stimuli: meaningless

syllables, uncommon words, common words, incongruent combinations of words

whose meaning related to colours (for example, the word lemon in red ink), different

colour names than the four ink colours used, and the names of the four ink colours

used written in another colour. The results showed that, compared with control stimuli

(groups of coloured asterisks), those six types of stimuli not only caused interference,

but said interference gradually increased from the first to the last condition. The

results showed that any stimulus composed of letters (even if it lacks lexical meaning)

causes interference. Additionally, said interference is related both to the frequency

that the words are used and to the degree to which the words suggest a certain colour.

Since the original experiment conducted by Stroop, many variations have been

made to test different phenomena. Macleod (1991) provides a comprehensive

overview of variations, including sorting and matching versions, picture versions, as

well as auditory and emotional analogs. This experiment, however, was a replication

of that of Stroop’s. The aim of the experiment was to investigate how manipulating

the independent variable, which in this case was condition, affects the dependent
 Investigating the Stroop Effect

variable, also known as response time. The major difference between existing studies

that have investigated the Stroop effect and this one is that we ran it using an online

tool called Qualtrics.

H1: Words presented in the congruent condition will be faster to categorise

than words presented in the incongruent condition.

H2: Fewer mistakes will be made within the congruent condition in

comparison to the incongruent condition.

Method

Participants

A total of 354 participants enrolled in the Massey University course

participated in the experiment. 14 of the participants were excluded because they were

under the age of 17 or did not disclose their age, leaving a total of 340 valid

responses. The mean age of participants was 31.96 (SD = 10.26) and this ranged from

17 to 74. Out of the 340 participants: 57 were male, 278 were female, 2 were

undisclosed, and 3 identified as non-binary. The study did not require ethical

approval, but the Massey University Code of Ethics was followed in the design and

running of the study.

Design

The study consisted of a within-subject experiment that used a repeated

measures design. The two-level independent variable was condition, which could

have either been congruent (ink colour matched the written colour name) or

incongruent (ink colour did not match the written colour name). The dependent
 Investigating the Stroop Effect

variable was response time (how long it took each participant to read the list of words

and identify the colour in which the words were written).

Software

The experiment was designed using a tool called Qualtrics which provides

advanced conditional logic tools that allow for complex experimental designs and

user-tailored survey paths. Given that this is a web-based experiment, the computer

equipment used could not be controlled.

Procedure

Before the experiment commenced, participants were asked to read through and

agree to a consent form that briefly covered what they were required to do and

explained what the data would be used for. The participants were then given an option

to disclose their age and gender. Next, instructions and an example were provided

which described the activity and how to answer.

The first part of the experiment consisted of a practice block, participants were

provided with 10 exercises where they were asked to identify whether sets of X’s

were written in blue, red, or green using multiple choice selection. The second block

of exercises required participants to identify the ink colour in which the words BLUE,

RED, and GREEN were written. For this set, the condition was congruent meaning

the ink colour matched the name of the colour written (For example, the word RED

written in red ink). The third block of exercises also required participants to identify

the ink colour in which the words BLUE, RED, and GREEN were written, but for this

set the condition was incongruent, meaning the ink colour did not match the name of

the colour written (For example, the word GREEN written in blue ink). Finally,

participants were presented with a simple debriefing message and given their results
 Investigating the Stroop Effect

which consisted of how long they took to complete the congruent and incongruent sets

of exercises, as well as how many correct answers they had out of 10 for each block.

The manipulation that was used was either to present words in a matching (congruent)

colour, or words in a non-matching (incongruent) colour. The total time that it took to

respond to each list of words was measured.

Results

Within the 340 valid submissions, it was found that the average response time

for the congruent list was 17.23 seconds, with a standard deviation of 6.44 seconds

representing the variability. Amongst the same submissions, it was found that the

average response time for the incongruent list was 20.05 seconds, with a standard

deviation of 9.73 seconds (See Table 1). Most participants responding to the

congruent list of words were able to respond within 10 to 15 seconds, whereas most

participants responding to the incongruent list of words took 15 to 20 seconds (See

Figure 1). Participants were, on average, able to answer with 99.1% accuracy and a

variability of 4.6% when responding to the congruent list of words, whilst participants

responding to the incongruent list of words were able to answer with 93.9% accuracy

and a variability of 21.4% (See Table 1). These results show that the list of congruent

words was both faster to respond to and provided more accurate responses, while the

list of incongruent words presented more variability within participants responses.

Table 1

Response Time and Accuracy Based on the Condition of Stroop Word Lists

Response Time Accuracy

Mean (SD) Mean (SD)

Condition

Congruent List 17.23 (6.44) 9.91 (0.46)

Incongruent List 20.05 (9.73) 9.39 (2.14)

Note: Response time is measured in seconds whilst accuracy is a mark out of

ten.

Figure 1

Number of Responses Given Within Varying Time Frames Based on Condition

180
160
140
Number of Responses

Congruent List Incongruent List

100
80

60
40
20

0
0-5

10-15

20-25

35-40

45-50

>60
5-10

15-20

25-30

30-35

40-45

50-55

55-60

Response Time (Seconds)

Discussion

The experiment provided results that supported both of the given hypotheses.

The data collected regarding response time supports the theory that words presented

in the congruent condition are faster to categorise than words presented in the

incongruent condition given that the response time for the former is 14.06% faster on

average than that of the latter. Additionally, the data collected regarding accuracy

suggests that fewer mistakes are made when responding to the list of congruent words

compared to the list of incongruent words given that answers are 5.2% more accurate

on average when responding to the former compared to the latter.

These findings mirror those of Stroop’s original study, who also found that the

list of incongruent words resulted in a higher average when measuring response time.

The results can also be compared to work done by Sheibe, Shaver and Carrier (1967)

who found that if the colour of the ink used matched the name of the colour written,

then the time taken by participants to name the colour decreased in comparison to

when those factors did not match.

There are various limitations regarding the participants used in the experiment.

Factors such as gender, age, and whether or not English is a person’s first language

could have possible effects on the given results, and thus they cannot be used to

generalise society as a whole. For instance, if a participant’s first language is anything

other than English or they are not perfectly fluent, it could be deduced that they would

be at more of a disadvantage than someone who is a native English speaker when

taking a Stroop test where the words are presented in English. Factors such as this,

amongst others, are inevitable and because of that it may be more accurate to test

groups individually based on these factors to see how results may vary. Future

research could be conducted in a variety of languages to determine how people

perform in their native language compared to a secondary language. Additionally,

researchers could take other factors, such as age, into account to determine what effect

they have on participant’s performance.

A strength this experiment may have presented compared to others in the field

is the number of participants. Collecting data based off a larger group of people, 340

in this case, presents the opportunity for more diversity amongst participants which

increases the validity when attempting to produce results that represent society as a

whole. Another possible strength this experiment had was its accessibility due to it

being available through a web-based tool instead of a lab run experiment. This could

have affected factors such as a participant’s levels of anxiety or comfort while taking

the test. Future research could also play off these components to test how the

conditions of a participant’s environments affect their performance.

Overall, the results of the experiment met the expectations presented in the

hypotheses, regarding both the response time and the accuracy of responses. It can be

concluded that when the condition of the word list is congruent, participants presented

both a faster response time and more accurate answers compared to when the

condition of the word list is incongruent. These findings reinforce Stroop’s belief that

the brain’s response time is hindered when it is forced to deal with conflicting stimuli.
 Investigating the Stroop Effect

References

Cattell, J. M. (1886). The time it takes to see and name objects. Mind, 1886, 11, 63-

65.

Dyer, E N. (1973). The Stroop phenomenon and its use in the study of perceptual,

cognitive, and response processes. Memory and Cognition, 1, 106-120.

Fraisse, E. (1969). Why is naming longer than reading? Acta Psychologica, 30, 96-

103. 

Hock, H. S., & Egeth, H. (1970). Verbal interference with encoding in a perceptual

classification task. Journal of Experimental Psychology, 83, 299-303. 

Klein, G. S. (1964). Semantic power measured through the interference of words with

color-naming. American Journal of Psychology, 77, 576-588. 

MacLeod, C. M., & Dunbar, K. (1988). Training and Stroop-like interference:

Evidence for a continuum o fautomaticity. Journal of Experimental

Psychology: Learning, Memory, and Cognition, 14, 126-135.

Posner, M. I. (1978). Chronometric explorations of mind. Hillsdale, NJ: Erlbaum.

Scheibe, K., Shaver, E R., & Carrier, S. C. (1967). Color association values and

response interference on variants of the Stroop test. Acta Psychologica, 26,

286-295.

Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of

Experimental Psychology, 18, 643-662.