Research of The Features and Mechanisms of Stroop Effect

Introduction

The studies evaluated in this paper all correspond to the Stroop effect, either by directly studying the effect or by determining the mechanism through which it occurs. The Stroop Effect is experienced when a written word differs from the color it is written in, and can be understood as a delay in reaching a correct response when presented with mismatching word and color combinations.

Research

J. Ridely Stroop was the man who first introduced the Stroop effect, coined after his own name, to modern psychology. In his 1935 study “Studies of Interference in Serial Verbal Reactions” Stroop studied interference’s effect on reaction times when asked to recite the words printed. He based his hypothesis off of prior works on interference and inhibition stating, that incongruence in the word-color pairing would result in an increased reaction time or an incorrect response. He conducted two versions of his experiment one with reading the word as the task and a second with naming the color of the print.

For the first he utilized 70 (14 male, 56 female) undergraduate volunteers. This experiment consisted of four conditions, two lists with two forms, and participants were asked to read the lists as fast as they could without mistakes as the experimenter followed along on a list printed in black ink. Half read the lists in the order b1, d2, d1, b2 and the other half read them in the reverse order to account for practice or fatigue effect.

The second part of the experiment utilized 100 volunteer undergrads with the task being to name the color not the word. Stroop found that word-color incongruence did not reliably increase the time when reading the words, but when naming colors the presence of words naming other colors increased the reaction times. He then concluded, “The associations that have been formed between word stimuli and the reading response are evidently more effective than those that have been formed between color stimuli and naming response.” (Stroop, 1935)

Though there was no formal statement of whether Stroop’s hypothesis was supported based on the findings it is clear that his hypothesis was partially supported. There is not much room for improvement, Stroop accounted for practice and fatigue effects within his methods and had a decent representation of the population. If I had to improve in anyway it would be to include more males in the study.

Zajano and Gorman in their 1986 study, “Stroop Interference as a Function of Percentage of Congruent Items”, sought to validate the Stroop effect, while investigating the potential of additional contributors to response competition. They hypothesized: “if the interference effect can be accounted for wholly in terms of such response competition, the mixed list used in the present study should result in a direct linear function of response times in relation to percentage of congruent items”. (Zajano & Gorman, 1986)

Their experiment consisted of a one-factor within-subjects design, with 33 (31 female, 2 male) undergraduate students who volunteered to undergo the study. The subjects were then administered 11 lists of varying color-word congruence levels, in a randomized order, followed by an additional 0% congruence list. The results showed that response competition caused the difference in reaction times, and with a 95% CI all points were significantly affected by this response competition.

The authors concluded that due to the “curvilinear” nature of the data that the inhibition is consistent with that of selective attention caused through response inhibition. Their hypothesis was supported by their results. The only issue is this experiment has a lack of subjects; there are only 2 males with 31 females this is not a good representation of the population and therefore is unable to be generalized in a confident manner.

MacKinnon, Geiselman, and Woodward studied how effort affected Stroop interference in their 1985 study “Effects of Effort on Stroop Interference”. Their study utilized 64 (32 male, 32 female) subjects selected from a UCLA introductory psychology class. Each subject was given 4 lists: a practice with color neutral words, a control with plus (+) signs instead of words, and 2 Stroop lists. The subjects were then asked to name the color of the print while being timed by the experimenter.

The order was counter balanced across the subjects. The experiment featured two conditions: low incentive, the subjects were prompted that the trial was just practice, and high incentive, the subjects were prompted that the trial was a competitive game. The authors hypothesized “that competition and the existence of a desired reward would lead to greater [subject] effort than a no-competition procedure. As a result of this task-specific effort manipulation, Stroop interference should be reduced if the interference can be brought under attentional control.” (MacKinnon, Geiselman, & Woodward, 1985)

The results were that low incentive had no significant effect, and high incentive had the effect of lowering Stroop interference supporting the hypothesis. The authors concluded that Stroop interference could be reduced if effort is directed to do so. I don’t see any room for improvement here; the subjects were not told the true intentions of the study to avoid good/bad subject bias and there was a good example of the population.

Augustinova and Ferrand studied the automaticity of reading printed words in their 2014 study “Automaticity of Word Reading: Evidence from the semantic Stroop paradigm.” Their study did not contain any subjects because it was only a comprehensive evaluation of other works rather than an independent study. Because this was a compilation of several other previous studies there was no procedure or formal hypothesis to report. However, where this study lacked formal procedure it reconciled this deficit with comprehensive data.

The following finding can represent the sum of this data: no matter how much effort is employed to inhibit it, the reading of a word is automatic. Augustinova and Ferrand concluded that the Stroop effect could be caused or if nothing else amplified by the automaticity of reading words when asked to name the color of the print if the word represents a different word. This study has no solid experimentation but yet makes a claim of causality; I would have like to see some original work done rather than rehashing several prior studies together.

The final study is “Time Course of Inhibition in Color-response and Word-response versions of the Stroop task.” completed Sugg and MsDonald in 1944. This experiment used 56 subjects, who were undergrads, required to participate in for a class at New Mexico State University. This study seeks to validate or refute the hypotheses of the previous works which this one is based, however it is never stated what these hypotheses are.

The subjects were faced with a screen, upon pressing the start button the first stimulus appeared, and then after a set time the second stimulus and so forth. Each stimulus consisted of a word inside of a rectangle, and each trial changed the word and the color of the rectangle. Each subject completed a “Trial Block” of 60 trials, 20 for each condition, congruent, incongruent, and neutral. Their study found that there was a main effect of task (naming color or naming word) and congruence. The authors concluded that significant inhibition was only found in short SOA (time between stimuli appearing) conditions and decreased SOA increased inhibition. A formal hypothesis would improve this study immensely; however clearly stating the previous studies’ hypotheses which it was testing would be a good start.

Critique

Through evaluating the aforementioned studies I learned a significant amount about the Stroop effect. The Stroop effect occurs when there is incongruence between the word meaning and the color of the print. However as expressed by Stroop’s 1935 experiment this effect is only significant when asked to report the color rather than the word meaning. Stroop further explains this mechanism by concluding that the associations between word stimuli and reading response is stronger than color stimuli and naming response. This is explained and validated by the automaticity of reading written words as expressed in Augustinova’s and Ferrand’s 2014 study, where they concluded that it is inevitable to read a word when it is present on paper.

Meanwhile Zajano and Gorman further supported Stroop by concluding that response competition was the only cause of the inhibition. These findings make sense and correlated with an experiment I had run myself. I believe Augustinova’s and Ferrand’s study could have been based more in directly observed experimentation rather than just restating and evaluating older studies, however due to the scope of data they present to support their conclusion I am satisfied with their findings. Overall the collective findings of these experiments make sense, because humans are incapable of being able to multitask at an efficient rate. And due to the automaticity of word reading when asked to discern the color of the font, there will be an inherent response competition.

I only have a few concerns with the studies and experiments that I found. The majority lying in the lack of original work or procedures, and lack of clearly stated hypotheses. While the methods and science were good, some of these studies cannot be called experiments because they do not seek to answer a single clearly stated scientific question. Aside from this fault that seems to frequent my chosen studies I do not have many concerns with the quality of these studies.