Replication of the Stroop Effect

Abdus Azad


Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

This experiment is a replication of the Stroop effect, in which you are supposed to identify colors, where colors are in different texts. Also the name of the word when in different colors. There was one experimenter and the participants were 20 (15 females, 5 males) Hunter College Psychology 250 students. The age ranged from 19-32 with a mean age of 23.5. The Experimenter tested the participant’s in two effects and three conditions. In one effect they were to identify the word, in three conditions congruent (matching the color), incongruent (different color), and control (neutral black text). In another effect the students were tasked with identifying the color in three conditions, congruent (same color matched), incongruent (different text), and control (neutral black text). We hypothesized there will be an increase in the mean time when participants are tested in the incongruent conditions due to interference. Using a one-way repeated measures ANOVA the result of this study found that there is a significant difference in incongruent condition when participants were tasked to name the color F (1.554, 23.32) = 7.434, p= 0.0055. Using a second one-way ANOVA for naming words, the results showed no significance F (1.845, 27.68) = 0.1289, p= 0.8642.

A Replication of The Stroop Effect

When we focus our attention to do one task, this is called selective attention. This means to exclude any other stimuli which may cause distraction. While on the other hand divided attention is when we have the ability to divide our attention to a few tasks, sometimes while doing these tasks, it may become an automatic process which makes dividing your attention between these two tasks much easier. Automatic processing, is when you cannot control your thought process, it can be thought as implicit thinking, in other words thinking that may occur unconsciously.

Automatic processing may not always be helpful, you may have conflict with automatic and controlled processing. The Stroop Effect tested how these processes worked. James McKeen Cattell (1886) had originally conducted research and found that objects and colors when compared to their corresponding words, took longer to speak aloud. He knew there was an interference with automatic and controlled processing. Even though he had already made this connection, Ridley Stroop (1935) is more recognized for work with color and words. Stroop in his experiment had used six colors and did three different experiments. The experiment’s consisted of participants reading lists of colors, on one list color was written in black, and on the other list was color written in different colored inks. They were supposed to identify the color of the words and also timed. The other experiments that he did was similar involving shapes. The way Cattell and Stroop’s research differs is that Cattell had concluded there was an interference between automatic and controlled processing, while Stroop’s research was more about developing an understanding as to why this interference is occurring. In Stroop’s article “Studies of Interference in Serial Verbal Reactions” he had concluded that your mind can automatically decide the semantic meaning of a word, hence when the color matched the word the time it took to figure out the color was much quicker vs when the colors did not match the word.

Further replications of the Stroop test were performed, J.D. Dunbar and C.M. McLeod’s (1990) replication like Stroop also consisted of a congruent condition (words with the same color as the text) and incongruent conditions (colors don’t match the word text). They also added a control condition, the words were in a neutral color. Flowers, Warner, and Polansky (1979) did a variation of Dunbar and MacLeod’s Stroop test. They used rows numbers, the number was the same in each row and they asked the participants to determine how many numbers were in the rows. The findings by both Danbar, McLeod (1990) and Flowers, Warner, Polansky (1979) showed there was a noticeable gap in time and errors made in the incongruent condition compared to the control and congruent conditions. With their replications, we can conclude that participants in the incongruent group take more time identifying the color and made more errors

Haely (1994) spoke of further research on automatic processing. An experiment was conducted on how we process words that we use frequently, words like “of”, “the”, etc. It showed that participants found it harder to focus on the individual letters of the words. She gave the participants some English text to read and asked them to circle every letter “t” that they saw. The findings showed that participants frequently missed letters that were in more common words, words like “the”, “then”, etc. When it came to less used words they were able to more easily identify it. These findings showed how we automatically process words. Words that are commonly used daily, when reading them, we see a whole entity, instead of the individual components. This way our brain automatically reads words further supports Stroop’s findings.

In our experiment we will attempt to do a modified replication of Stroop’s experiment to see how our results correspond to his. We will have two groups one will be to Name The Color, and the other to Name The Word. Our null hypothesis is that there will be no significance mean difference in time recorded for participants in congruent and incongruent groups from the control group. The alternative hypothesis for our experiment is that the time recorded for participants in the incongruent groups will be significantly more. From the many studies done on Stroop Interference, mostly all suggesting that the incongruent group will take more time to identify the color or word due to Stroop Interference, this is where our alternative hypothesis has come from.



The participants of this study were all Hunter College students. There were 5 males and 15 females, ages from 19-32 (M=23.50, SD=3.80). 10 of the participants spoke English as a second language. The hours of sleep the participants had varied from 4 -7 hours (M=5.36, SD=1.02). Only one participant had reported vision problems. Ethnicity and socioeconomic status were not taken into consideration. All of the students who participated were from the Psychology 250, class that meets on Mondays – Thursdays from 8:00 am to 11:20 AM. The participants will all be compensated in the form of a letter grade from the professor.


The testing was taken on a consisting of three conditions. Control (color of word written in black text), congruent (word was written in same color), and incongruent (color of word was different from text), the time it took to answer was recorded by the computer. The results were displayed on the screen and were written on paper with a pen or pencil. The students transferred their results into a shared excel data table which was uploaded on Blackboard. The participants completed the test in a computer lab using computers.


There was informed consent and the students were briefed on the experiment with no deception. The students were all presented with the same instructions. We conducted a within subjects design, to select the order of the participants a counterbalancing method was used to avoid possible sequence effects. Each participant was assigned a number from 1-6. Depending on your assigned number you would start with a different condition. 8 participants started with the Control, 6 participants started with the congruent, and the remaining 6 started with the incongruent condition. 3 participants were chosen at a time and went took the experiment on different computers in three different rooms. Once in the room the participant closed the door, and started with the assigned condition. There was two effects, in one the participant was required to Name The Word, and the other the participant was supposed to Name The Color. Each effect had three conditions. The control condition was black colored text displaying a color. The congruent condition had both had a text corresponding to the color. The incongruent condition had a text of a color displayed in a different color. Words would display and the user was required to click the answer as quickly as possible, after one condition was completed they moved on to the next. After a participant was complete another participant shortly followed to the empty computer following the sequence previously mentioned. We chose to exclude any extreme values from our data any person with values ± 2.0 SD from the mean


When the participants were required to Name The Word, in the control condition the mean score and standard deviation was (M=1494.89, SD= 314.19), in the incongruent condition the mean score and standard deviation was (M=1545.93, SD=283.30, and in the congruent condition the mean score and standard deviation was (M=1483.74, SD= 264.03) see Figure 1. In the second effect when the participants were required to Name The Color, in the control condition the mean score and standard deviation was (M=1661.22, SD=248.22), in the incongruent condition the mean score and standard deviation-n was (M=17.63.41, SD=416.93), and in the congruent condition the mean score and standard deviation was (M=1400.78, SD=454.59) see Figure 2 . A one way repeated measures ANOVA was conducted twice once to test Name The Color and another to test Name The Word. It was found there was significance difference in mean time for Name The Color group F (1.554, 23.32) = 7.434, p= 0.0055, the post hoc analysis shows incongruent had a significant mean time slower at (p <0.005). For the Name The Word group F found no significance F (1.845, 27.68) = 0.1289, p= 0.8642. Our hypothesis was partially correct, our results show that when naming color there is a significant delay in the incongruent condition. While when naming the word there is no significance.


The aim of this study was to do a modified replication of the Stroop effect. We hypothesized that participants would have an increase in interference with more complex tasks. In other words, the mean reaction time will be significantly higher between the incongruent conditions.

The results proved our hypothesis partially correct, which was that participants would have an increase in mean time in the incongruent condition. In other words, the mean reaction time will be significantly higher in Naming The Color and Naming The Word incongruent conditions. Our result found significance when participants saw words with different colors and they were required to name the color. There was no significance when they were required to name the word. The findings were somewhat inconsistent with our hypothesis because other previous studies also showed that when the word conflicts with the ink color, people are slower to respond and they are faster if the word agrees with the ink color, (Cohen, Dunbar, McClelland, 1990). So, previous studies agree with our results.

Our findings were consistent with the first and second experiment, which was done by Stroop (1935). Our results agree with Stroop’s study because in both studies tested participants in different conditions and the conditions were counterbalanced to avoid order affects. Stroop had found a delay in his second experiment when the color was supposed to be named with different words, and no significant delay in the first experiment. Our replication yielded similar results.

In both studies, participants had a task of naming words of colors which were written in a different color (incongruent) and then also naming neutral words (control). Since previous research had already explained that when a word complements its ink color, it leads to the processing of naming the word and the color together and if the color and the word are different it takes a longer time of reaction or response, (Cohen, Dunbar, McClelland, 1990), we can conclude that both studies would have similar findings.

Our result was also consistent with the second study done by Algom, Eidels, and Townsend et al. (2009) which found that when a participant is given the task of naming color words that are printed in color, they report the ink color faster if that color word is the name of the color rather than the name of a different color.

By doing this study we were able to see that when faced with two tasks at the same time our brain responds to the immediate visible one. So, when we see the word blue written in the color green, we automatically are triggered to name the word, which is blue because that is processed in our brain first. However, if we are told to say the name of the color that the word is written in, like blue written in green, it takes a longer time because now the brain has to overcome the first step in automatically just recognizing the word, we have to voluntarily pay attention to the word and it’s incongruent color to name just the color of the word, which takes a longer time, as explained by Cattell (1886), Posner and Snyder (1975), Shiffrin and Schneider (1977) in their studies.

Some limitations of this study was that the sample size was too small (N=20) so it was not representative of the larger population. There is a biased sample, the age group of the participants is fairly narrow. So the results obtained would not be acceptable to generalize the whole population. With a larger sample, the results may have varied. Also, the experiment was done too early in the morning when people are usually tired, sleepy, moody or hungry. So, their moods can have a great effect in how much time they take when testing in the three different conditions. And since, condition 3 took more attention and concentration, certain moods or hunger could have affected their concentration.

Future studies should examine look at Stroop test in regards to color and color-object naming instead of just color-word naming to see if our brain works the same way for both tasks. The sample size should be much larger and there should be more variability. Also, gender should be divided equally because in our study, there were 9 females and only 3 males. Also, future studies should do the Stroop task with other different age groups and compare how one age group’s mean reaction time is different from that of a younger or older age group.


Cattell, J. M. (1886). The time it takes to see and name objects. Mind, 11, 63-65.85). Hillsdale, NJ: Erlbaum.

Cohen, J., Dunbar, K., and McClelland, J. (1990). On the Control of Automatic Processes: A Parallel Distributed Processing Account of the Stroop Effect.

Psychological Review, 97 (3) 332-361. Retrieved from

Eidels, A., Townsend, J., & Algom, D. (2009). Comparing perception of stroop stimuli infocused versus divided attention paradigms: Evidence for dramatic processing differences Cognition, doi:10.1016/j.cognition.2009.08.008.

Flowers, J.H., Warner, J.L., & Polansky, M.L. (1979). Response and encoding factors in “ignoring” irrelevant information.Memory & Cognition, 7,86-94

Healy, A. (1994). Letter detection: A window to unitization and other cognitive processes in reading text. Psychonomic Bulletin & Review, 3, 333-334

Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human information processing: II. Perceptual learning, automatic attending, and a general theory. Psychological Review, 84, 127-190.

Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 12, 643-662.

Figure1. -. The control condition the mean score and standard deviation was (M=1494.89, SD= 314.19), in the incongruent condition the mean score and standard deviation was (M=1545.93, SD=283.30, and in the congruent condition the mean score and standard deviation was (M=1483.74, SD= 264.03). The bars represent SD from the mean

Figure 2.- The control condition the mean score and standard deviation was (M=1661.22, SD=248.22), in the incongruent condition the mean score and standard deviation-n was (M=17.63.41, SD=416.93), and in the congruent condition the mean score and standard deviation was (M=1400.78, SD=454.59). The “**” represents a p<.005. The bars represent SD from the mean

You Might Also Like

I'm Alejandro!

Would you like to get a custom essay? How about receiving a customized one?

Check it out