The Stroop Effect
The Stroop Effect and its Influence on Accuracy and Reaction Time
One of the earliest notions of attention and its importance can be traced back before the scientific study of psychology as a whole. Vives (1538), in his book, De Anima et Vita, found that people retain stimuli more accurately if a person mentally attends to and concentrates on the stimuli at hand and thus, strengthens the memory system all together. These findings may seem obvious today; however, Vives’ (1538) work was centuries ahead of his time. Since Vives’ (1538) early work, attention has been sorted into various categories such as selective attention, sustained attention, and divided attention (Perry and Hodges, 1999). Furthermore, Vives would help set the stage for others, including James (1890), Norman and Bobrow (1975), Shiffrin and Schneider (1977), and Stroop (1935) who would study and theorize exactly what it means to “pay attention.”
James (1890) theorized that selective attention is the way in which a person consciously focuses on visual, tactile, or auditory information while ignoring other irrelevant stimuli that may be impending on the senses. En lieu of visual selective attention, James (1890) postulated a spotlight model or a model in which the visual field is narrowed to a spotlight around an area of focus, allowing information to be extracted. Spotlight attention, today, is still considered an attentional mechanism that operates in parallel, automatically scanning the visual field for recognizable features (Bengson, Lopez-Calderon, & Mangun, 2012). Additionally, this visual spotlight not only possesses a particular selective focus, but an area of fringe in which low-resolution or foggy information may be collected, and a margin area, or an area in which information is outside the visual field and completely inaccessible for further processing. Furthermore, James (1890) insisted that everyone knows what attention is, as it is an experience of everyday life that shortens our reaction time and allows people to conceive, perceive, distinguish, and remember.
While James’ spotlight model came long before the cognitive revolution in the 1950s, Norman and Bobrow (1975) furthered attention theories after the revolution. They postulated that attention, including selective attention, might not be a spotlight-like contraption, but a pool of resources that can be allocated as the mind sees fit. Different tasks compete for and demand different amounts attention from this resource pool, subsequently dividing the resources between one or more tasks (Lesgold, 1989). With this division, come limits. For one, the pool of attention is resource limited, meaning that there is an inability to fully concentrate and accurately perform more than one task at the same time. For example, if a person is texting while driving, the resource pool of attention is split between driving and typing coherent words and sentences. Neither of the tasks are performed particularly well as the bulk of the attentional resources tend to shift between the two exercises, ultimately neglecting the other task (Norman & Bobrow, 1975). Furthermore, the resource pool is also data-limited, or a limitation in which insufficient sensory input is available to perform a task, or the skills needed to carry out a particular task are unavailable.
Taking Norman and Bobrow’s (1975) resource pool model, Shiffrin and Schneider (1977) aimed to explain the automatic processes behind attention, or automaticity. They argued that when a person first learns a task, a great deal of attentional resources and controlled processing go into doing so. Though after much practice and repetition, a skill is developed and less attention is needed, making the process automatic. This is commonly seen in reading. For novice children, reading is difficult. They must learn the names of the letters, what sounds that they make, and the grammatical rules for their particular language. This results in dedicating a large portion of the attention resource pool to reading, making it hard for them to concentrate on anything else during the task. After repetition and practice over months and sometimes years, however, reading becomes a fluid skill to the point when a person is presented with a word, it is almost impossible not to read it (Schwanenflugel, Meisinger, Wisenbaker, Kuhn, Strauss, & Morris, 2006).
Such automaticity has been used to explain various psychological tasks such as the Stroop effect (Stroop, 1935). The Stroop effect, rightfully named after J.R. Stroop, is a demonstration in which the psychological interference of reading specific words hampers the naming of the color of the words. For example, if the word is “blue” printed in red, when asked to name the color of the print, the reader will take longer to say “red” as compared to naming the colors of neutral symbols such as x’s, or in Stroop’s (1935) second experiment, squares.
Though automaticity is the most common theory, there are other theories that aim to explain the Stroop effect include parallel distributed processing and processing speed (Cohen, Servan-Schreibe, & McClelland 1992). Exploring these theories, the current study aims to replicate much of Stroop’s (1935) famous experiment. Such an experiment has been used time and time again to measure a person’s capacity for selective attention as well as their mental processing speed (Lamers, 2010). While Stroop’s (1935) study is very useful in its own right, the current experiment gives way to the possibility that, because participants in the former study named colors out loud, people may name colors faster or slower silently and to themselves rather than vocally. Furthermore, Stroop’s (1935) study only had two conditions, a control condition in which participants were asked to name the color of squares, and an incongruent condition in which the name of the word did not match its color. This alludes to the possibility that perhaps color words with congruent color names may or may not cause interference. The present study takes this into consideration in extending Stroop’s (1935) study, postulating that the reaction time in naming colors with incongruent words, for example, the word “blue” in red print, may be significantly longer than naming a color with its congruent word or naming the colors of neutral stimuli such as a row of x’s.
Methods
Participants
The current study sampled 52 college students, ranging from 19 to 33 years of age. Of the 52 participants, 39 were women with a mean age of 22.55. The remaining 13 men had a slightly smaller mean age of 21.08. Both men and women were selected from several sections of an upper level psychology course, Memory and Cognition. It should be noted that the participation of the students was a class requirement and that one participant did not report her age.
Apparatus
The participants were initially asked to run the Memory and Cognition Laboratory (Daily, 2012). This program enabled the current study to replicate and extend experiments done by researchers such as Stroop (1935). To ensure that the timing and exposure of each of the trials were consistent, a system of timing functions known as ExacTicks was used (Ryle Design, 2000).
Procedure
Besides the fact that Stroop (1935) did not have access to a computer or software and that his participants named colors out loud, the present study was extremely similar with regard to the procedure and collection of data. Within the Memory and Cognition Laboratory Software, each participant encountered 36 total trials. The 36 trials were split between three conditions, subsequently creating 12 control trials, 12 congruent trials, and 12 incongruent trials. Within these conditions the words red, green, blue, and yellow would each appear three times printed in red, green, blue, or yellow font. In the congruent condition, the word matched the color of the print. For example, if the word was “red”, the word would be printed in red print. Furthermore, the length of strings of x’s in this condition matched the length of the color words. Lastly, the incongruent condition was the condition in which the print color did not match the color named by the word. In other words, the word “red” printed in blue ink.
After the participants pressed the “begin” button, they were further prompted to press the “ready” button to ensure that the participant was prepared before each trial. After the “ready” button was pressed, the screen clears and a gray fixation cross appeared in the middle of the window. This cross would remain on the screen for .5 seconds, followed by a stimulus. Said stimulus was then displayed until the participant pressed one of the four response options. The participant was then to match the color of the stimulus to one of the response options. After they responded, the stimulus was then replaced by feedback reading “Correct Response” or “Incorrect Response” depending on whether the participant’s answer was correct or not. This feedback would appear on the screen for 1.0 second and would then be replaced by the “ready” prompt once again.
Results
It was expected that reaction time of the incongruent condition would be greater than the control or congruent condition. For all analysis in this paper, α was set to .05. The mean percentage of response time in each condition can be seen in Figure 1. As expected, stimulus type had a significant effect on participants’ response times, F (2, 102) = 24.87, p < .05, MSE = 11449.427, η2 = .33. Post hoc analysis using Tukey’s HSD revealed that participants took significantly longer to respond to the incongruent stimuli than they did to respond to either the congruent or control stimuli, but that participants’ response times did not differ in response to the congruent or control stimuli.
Discussion
As predicted, participants did indeed take longer to react under the incongruent condition than they did in the congruent or controlled conditions. Though Stroop’s (1935) study did not have a condition in which print color matched color words, or congruent condition, our findings were consistent with Stroop’s (1935) in that the neutral stimuli posited a quick response in color naming, while it took longer for participants to name the colors of incongruent words. Therefore, a strong case for interference is presented. This interference is likely explained by the cognitive fight between automaticity (Shiffrin & Schneider, 1977), and selective attention (Sabri, Melara, & Algon, 2001).
With regard to these findings, theories of automaticity and selective attention in the Stroop effect postulate that naming colors, unlike reading, is not an automatic process, but a controlled process. It is not something that is done frequently throughout every day, unlike reading; thus, color recognition and naming requires more selective attentional resources (Stirling, 1979). Furthermore, when a person is prompted to name the color, they automatically read the word first and must make a conscious effort to override this impulse, suggesting that automaticity, in this case, is stronger than selective attention (MacLeod & Dunbar, 1988).
Other researchers such as Cohen, Servan-Schreibe, and McClelland (1992) argue that selective attention’s struggle with automaticity may not be the only explanation of the Stroop
effect and that processing speed plays also plays a role. Cohen et al. (1992) theorize that the brain recognizes colors faster than it does reading. Because of this, when presented with the option to read or recognize colors, reading is inherently faster. While people know they should name the color and not read the word, reading inherently comes first and involuntarily, ultimately causing processing confusion and slow reaction time.
Though Cohen et al. (1992) acknowledged that processing speed was the simplest explanation of the Stroop effect, Cohen, Dunbar, and McClelland (1990) also entertain a parallel distributed processing model in which certain tasks and skills, such as reading and color naming, create different pathways within the brain. These pathways are then strengthened by repetition and practice. In the case of Stroop Interference, Cohen et al. (1992) suggest that reading possesses a stronger pathway than does color naming. When performing the Stroop Task, both of these pathways are activated; however, because the reading pathway is stronger or more hallowed out, the reading pathway has a more profound effect and slows down response time.
Future directions regarding the Stroop effect and the current study include the investigation as to why or how much the distance between the response keys increases or hinders Stroop interference (Chen & Proctor, 2012). In the present study, response options were rather close together suggesting more of an affect than if they were farther apart. Furthermore, limitations include the lack of generalizability within the participant sample considering that the sample was not particularly diverse; however, there is no reason to believe that the influences Stroop interference are different for other races and groups of people.
References
Bengson, J., Lopez-Calderon, J., & Mangun, G. (2012). The spotlight of attention illuminates
failed feature-based expectancies. Psychophysiology, 49, 1101-1108.
Cohen, J.D., Dunbar, K., & McClelland, J.L. (1990). On the control of automatic Processes: A
parallel distributed processing account of the Stroop effect. Psychological Review, 97,
332-361.
Cohen, J.D., Servan-Schreiber, D., & McClelland, J. L. (1992). A parallel distributed processing
approach to automaticity. The American Journal of Psychology, 105, 239-269.
Daily, L.Z. (2012). Memory and Cognition Laboratory (Version 4) [Computer software].
Shepherdstown, WV: Larry Z. Daily.
Hiatt, K., Schmitt, W., & Newman, J. (2004). Stroop tasks reveal abnormal selective attention
among psychopathic offenders. Neuropsychology, 18, 50-59.
James, W. (1890). The principles of psychology (pp. 402-458). New York: Henry Holt and
Company.
Lamers, M.J. (2010). "Selective attention and response set in the Stroop task". Memory and
Cognition 38, 893–904.
Lesgold, A. (1989). Foundations for a psychology of education (pp. 60-62). Hillsdale, N.J.: L.
Erlbaum Associates.
MacLeod, C. M., & Dunbar, K. (1988). Training and Stroop-like interference: Evidence for a
continuum of automaticity. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 14, 126–135.
Norman, D.A., & Bobrow, D.J. (1975). On data-limited and resource-limited
processes. Cognitive Psychology, 7, 44-64.
Perry, R. J., & Hodges, J. R. (1999). Attention and executive deficits in alzheimer's disease A
critical review. Brain, 122, 383-404.
Proctor, R., & Chen, J. (2012). Dissociating influences of key and hand separation on the Stroop
color-identification effect. Acta Psychologica,141, 39-47.
Ryle Design (2000). ExacTicks (Version 1.1)[computer software]. Mount Pleasant, MI: Ryle
Design.
Sabri, M., Melara, R., & Algom, D. (2001). A confluence of contexts: Asymmetric versus global
failures of selective attention to Stroop dimensions. Journal of Experimental Psychology:
Human Perception and Performance, 27, 515-537.
Schwanenflugel, P., Meisinger, E., Wisenbaker, J., Kuhn, M., Strauss, G., & Morris, R. (2006).
Becoming a fluent and automatic reader in the early elementary school
years. Reading Research Quarterly, 41, 496-522.
Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human information
processing: 11. Perceptual learning, automatic attending, and a general theory.
Psychological Review, 84, 127-190.
Stirling, N (1979). "Stroop interference: An input and an output phenomenon". Quarterly
Journal of Experimental Psychology, 31, 121–132.
Stroop, J. (1935). Studies of interference in serial verbal reactions. Journal of Experimental
Psychology, 18, 643-662.
Vives, J. (1538). De anima et vita. Torino: Bottega d'Erasmo.
