Summary Of The Stroop Experiment

Cognitive Inhibition and Mental Processing Speeds for Congruent, Incongruent and other Conditions: The Stroop Experiment
In the study of association and interference, The Stroop Experiment is perhaps the most definitive study. First devised in 1935, the experiment demonstrated the relationship between interference and reaction time when performing specific tasks (Stroop, 1935). This experiment was birthed by Münsterberg’s theory which questioned whether alterations in common tasks were inhibiting for individuals (Stroop, 1935). Müsnsterberg concluded automatic association functions can occur irrespective of existing contrary associations (Stroop, 1935).
This topic of cognitive inhibition and response is of particular interest as it has

use in the field of mental decline; increasingly present in Australia’s aging population (Cubit, 2011). In the past, the Stroop Experiment has been regarded as a measure for recognising or diagnosing frontal lobe damage as well as diagnosing specific cognitive impairments such as dementia, ADHD and schizophrenia (Heflin et. al., 2011). Furthermore, psychological conditions in which certain subjects or word topics are prominent have been explored to reveal that the presence of these words inhibit the ability of an individual to quickly and accurately perform the task (e.g Anorexia Nervosa and the word ‘weight’) (Redgrave, et. al., 2008). Performance on the Stroop Test can be indicative of coping styles as well as individual neurological patterns and tendencies (Chang, Yusoff & Begum, 2015).
Individual performance in the Stroop Experiement is dependent on biological processes of cognitive control, the level of which increases when the stimulus is competing or incompatible (Carter & van Veen, 2007). When such incompatibility is detected, the anterior cingulate cortex is thought to be responsible (Carter & van Veen, 2007). This encourages the dorsolateral prefrontal cortex to become stimulated and correct the incompatibility or ‘conflict’ (Carter & van Veen, 2007). Such processes are engaged throughout the duration of the Stroop Experiment, as stimulus is purposefully conflicting.
This study aims only to replicate the findings of such works and assess the reproducibility of results across a removed sample.

It is predicted that results will be congruent with theory formulated from initial studies, as these findings have remained consistent for decades and are a logical extensions of supported evidence. The consensus among previous experiments, it that words presented which are congruent with their colour, will have the fastest reaction time, words which are contrast to their colour will have the slowest reaction time, and words which are non-colour will have an intermediate reaction time (Ergen et. al., 2014).
The independent variable for this experiment was the type of word presented and consisted of three variants; congruent, incongruent, non-colour. The dependent variable was the reaction time for each response, recorded in milliseconds throughout this experiment.
Three related hypotheses were formulated to aid in quantifying the relationship between said variables. These are; when the word is a colour name and is presented in the colour it names, colour naming will be at its fastest, when the word is a colour name and is presented in a colour different to the one it names, colour naming will be at its slowest and when the word is not a colour name, colour naming will be at an intermediate speed.

Method
Participants
Participants were students enrolled in an undergraduate psychology degree at the Queensland University of Technology, who were present during their allocated tutorial over a one week period. This equated to 357 students. No pre-test procedures were conducted to determine quality of eyesight or participant intelligence. Similarly no demographic factors were recorded or restricted participation in the experiment.
Design
The independent variable was the word type presented and included three variants; congruent, incongruent and non-colour. Each participant was presented with an equal number of each variant, in a randomised order as was controlled by the software program. The dependant variable was the reaction time recorded in milliseconds by the software program, which presented the results at the conclusion of the experiment for each individual.
Apparatus and Procedure
The apparatus used to present the stimuli was 17 inch computer monitor provided by QUT which concomitantly recorded the participants’ responses. The software program was installed on each computer immediately prior to the commencement of the experiment.
The program presented a three to six letter word which was either congruent (e.g. word red in red writing), incongruent (e.g. word red in blue writing) or non-colour (e.g. bed in red writing) in one of four colours (red, yellow, blue, green). There was 96 trials completed for each participant which included 48 non-colour words and 48 colour words. Participants were able to control the pace at which the stimulus was presented through the use of the ‘Next’ button. The participant’s task was to select the colour of the text in which the word was presented, not the colour which the word named, as quickly as possible. This was repeated until all trials were completed. A warning was given if no response had been given within two seconds of if the given answer was incorrect. Following the completion of the test, the software program produced an averaged total for each participant, for each word type. The participant was then required to write this figure down, and communicate it to the supervisor of the group.
Results
The data presented was the average reaction time for the 357 participants in selecting the correct colour, for each variation of the word type. Incorrect responses, and those which exceeded 2000 ms were not included in this calculation. These scores were tabulated for the entire cohort sample and the average was calculated and presented below (Figure 1).

Figure 1.

Stroop Test Average Reaction Time (ms) for the Sample in Congruent, Incongruent and Non-colour words
Figure 1 presents the average score in milliseconds and shows that findings support the given hypothesis. For the congruent condition, the average speed of response was the fastest (M = 716.80, SD = 71.52) when compared to other data sets. As hypothesised, colour incongruent words had the slowest reaction time (M = 784.56, SD = 108.50), however the variability of this variable was notably larger than the other two variables. The time for non-colour words was intermediate (M = 736.13, SD = 78.18), however was closer to the speed for congruent word types indicating a marginally faster reaction time.
These initial results appear to be in line with the hypothesis given. Results of the dependent group t-tests varied substantially between data sets. The difference between Colour Congruent and Colour Incongruent reaction times was significant, t(356) = -17.36, p < .001. Between Colour Congruent and Non-colour words, there is also a significant difference, t(356) = -9.11, p < .001. There was also a significant difference between the Colour Incongruent and Non-colour words, t(356) = 14.61, p < .001. Each of these comparisons conforms to the definition of statistically significant and hence can, with credibility, support the research

hypothesis.
Discussion
Data collected and analysed from the completion of this Stroop Experiment, has been supportive of all three hypotheses. Results revealed that the fastest reaction time was for the congruent word variable while the slowest was the incongruent word variable. The non-colour words showed an intermediate reaction time. This is supportive of the overarching theory which states that reaction time is inhibited when the stimulus is changed or not congruent.
A lack of demographic data reduced the potential of the results found in that an abundance of a certain characteristic would not have been detected and hence could have varied results. Due to a limited knowledge of the algorithm used throughout the software, it is possible that a pattern of presented words emerged which could improve the reaction times, as participants would be able to anticipate or predict the variant before it is presented. This aimed to be recovered through the randomisation of the words. Furthermore, it is plausible that it would take longer to process the six letter words than it would the three letter words. This was corrected by using a unanimous word length for non-colour words, and a similar orthography to the colour words i.e. red vs. bed.
In light of these limitations, it remains likely that the most logical conclusion for the variations in reaction time is due to the changes in word type, as external factors were been moderated to an extent and numerous previous studies have supported this same conclusion.
The standard deviation results for the Colour Incongruent variable was notably high in comparison to the other variables. The study’s conclusion is likely to remain accurate however, regardless of possible anomalies, in that other studies of a similar nature similarly found that this variable was associated with the slowest reaction time (Carter & van Veen, 2007).
Seeing as this experiment was designed to replicate previous findings, it can be concluded that the theory behind the Stroop Experiment is supported further from these findings. Although there is extensive research into this experiment and the field of association and interference, there continues to be an opportunity to clarify certain specific elements of the field. With the global ageing population, this experiment could be modified and utilised in the assessing of mental decline, and aid in the improvement of public health.