Apparatus
Standardized instructions were read to participants prior to the experiment, ensuring the instructions were understood. In addition to word lists (see Appendix A), which were required to carry out the experiment and provide participants with stimuli. The sets of cards (reading and writing conditions) each had 80 words (8 rows of 10), letter strings or X’s printed on one side (font style and size Arial black 14). Cards used in the control conditions (testing reading and naming accuracy) either had coloured words in black ink or rows of X’s printed in coloured ink. Whereas the experimental conditions (six) included coloured words (red, green, blue and orange) printed in the same ink, opposing colour words, colour-related words of high frequency, non-colour or neutral related words of high frequency, low frequency neutral words and letter strings. It is important to note the words were printed in a ‘pseudo-random order’, whereby a line had no more than two same words, reducing the likelihood of participants remembering the stimuli. Lastly, a stopwatch was required to measure the time taken to read the set of 80 words on the card.
Design
A mixed design was used, consisting of one within-subjects factor (control conditions comparing reading and naming) and one between-subjects factor (colour-relatedness). In the experimental condition (independent groups), separate groups of participants were tested in each condition, resulting in each individual taking part in a single trial. Whereas for the control conditions, a direct comparison between reading and colour naming was made, so each participant took part in both. There was no random assignment of participants to each experimental condition. Yet the order that words were printed had been in a randomized order, reducing any possible differences. The independent variables were the type of task (reading or naming the colour) and condition (Classic Stroop, other colour, colour related, high frequency neutral, low frequency neutral and letter strings). The dependent variable was the time taken to respond to the stimuli, comparing the experimental and control conditions.
Procedure
Prior to the study, all participants gave their informed consent to taking part in the experiment. The control conditions consisted of two stages: measuring reading and colour naming response time and then ranking time taken to name the colour, so to match the experimental groups and allocate participants to each condition (ensuring each takes part in one experimental condition). The participants were placed into groups of three and conducted the task in a separate cubicle, whereby each participant took turns to read the stimuli aloud while another recorded time. Before the stopwatch was pressed, the cards were hidden faced-down until the task commenced. Throughout the trial, the other participants were told not to look at the test cards, so to reduce participant biases.
In the control conditions, the participants were separately tested in the first condition (reading) and after, the second (colour naming). This consisted of participants reading out loud the words written on the card as soon as it was turned over, aiming to complete the task as quickly and as accurately as possible. When completed, the participant reported they were ‘finished’ and the stopwatch was paused. Whereas in the second condition (colour naming), the participants were instructed to name aloud the ink colour of the ‘X’s. Participants were also encouraged to correct any errors throughout the task. After both control conditions, participants each individually wrote their time (in seconds) for condition B in large text and the class organized the order of times by forming a single line across the room. Each participant was then assigned to one of the six experimental conditions, so to organize colour-naming times into different conditions (reduces possible confounding variables). For example, the six fastest naming participants had been allocated to different conditions, ensuring variety in each and the experimental condition was carried out.
In the experimental condition, each participant was asked to name aloud the ink colour of each word or letter string when the card was turned over. It was clearly indicated that participants should only name the ink colour and ignore the word written. As before, students were encouraged to correct any errors and indicate when they had finished. This procedure was repeated for each participant. It is important to note that participants were told not to purposefully distort their vision (such as by squinting), or cover up or point to parts of words, since this could reduce the internal validity of the experiment.
Results
The results (raw data in Appendix B) were collected in order to assess whether the Stroop effect had an influenced on response time. It was found that participants responded quicker when reading (M=30.8 1 (d.p.)) than colour naming ink words (M=47.1 (1.d.p.)). Although, a higher degree of variation was observed in colour naming time (SD=8.0 (1 d.p.)) than for reading time (SD=4.4 (1 d.p.)),
Therefore, the hypothesis that there is a difference between identifying the colour and reading word time can be accepted as less time was taken to read the colour of the stimuli. The mean score for each condition was calculated by observing the difference between the times taken to name the experimental condition minus the time of the control condition (colour-naming increment). A related samples t-test (one-tailed) between reading and colour naming time was run. Reading time (M=30.8, SD=4.4) was found to be significantly faster than colour naming time (M=47.1, SD=8.0), t(37)= 16.2, p< .001, one-tailed.
As for the experimental condition, participants took the longest to respond to other colour words (mean of 15.3 (1 d.p.)), while the low frequency neutral words had the shortest response time (mean of 1.6 (1 d.p.)).
Reading time Colour naming time
Mean 30.8 (1 d.p.) 47.1 (1 d.p.)
SD 4.4 (1 d.p.) 8.0 (1 d.p.)
Table 1: A representation of the mean and standard deviation time (seconds) for the control conditions (comparing reading and colour naming words)
Save your time!
We can take care of your essay
- Proper editing and formatting
- Free revision, title page, and bibliography
- Flexible prices and money-back guarantee
Figure 1: A graphical display (bar chart) to demonstrate the mean response time (seconds) for the control conditions (comparing reading and colour naming words).
Classic Stroop Other colour Colour-related High-frequency neutral Low frequency neutral Letter strings
Mean 13.4 15.3 12.0 5.2 1.6 2.0
SD 8.6 16.0 13.4 1.7 2.6 3.9
Table 2: A representation of the means and standard deviation time (seconds) for each of the six experimental conditions
Figure 2: A graphical display (bar chart) to demonstrate the mean time (seconds) taken to respond for each of the six experimental conditions
Discussion
A comparison was made between reading and colour naming time to investigate whether the Stroop effect was present when identifying colour. It was found that reading coloured ink words took less time (seconds) than naming the written colour. In addition to the experimental conditions which observed a slower response rate to words that implied or were related to a certain colour. Therefore, the results display support for the hypothesis that states there is a ‘gradient interference effect’ when naming colours, as the less colour-related the stimuli were, the faster the response. This could be explained through a sense of competition between the semantic processes in which both required attention, thus increasing time taken to name the colour. Although these findings provide support for previous research into the Stroop effect, the validity of the experiment has been questioned.
The results can be interpreted as the result of interference of colour stimuli on the written names of the colours (when two different words are presented in conjunction). The graded interference effect (between colour-naming and word reading) is understood as a result of words stored in the semantic network of memory, and thus, a specific word may ‘activate’ another such as ‘sky’ may hint at ‘blue’. Though meaningless words may have similar letters to a colour, which could similarly elicit a response, such as ‘blir’ provoking ‘blue’. Therefore, the order in which stimuli are presented may be as equally important as the word itself.
Previous literature into the Stroop effect also found significant results, which increases support for the hypothesis that reading words is simpler than naming the colour. Evidence and existing theoretical perspectives have been reviewed (MacLeod, 1991) and perhaps there is a lack of statistical significance, whereby the difference between the conditions may be too small. Nevertheless, research has considered variables that could influence the Stroop effect, such as the content of the stimuli (appearance of words). A significant study conducted by Henderson (1973) consisted of participants who were told to find a specific letter (upper case) in a jumble of letters and found that if the stimuli had more of the same letter, but lowercase, then the task was reported as more challenging. Similarly, naming global letters may be just as relevant, such as if letters appeared in the same shape as the letter, then participants responded more effortlessly at a quicker rate (Navon, 1977). Though it may alternatively be the quantity of stimuli, such as the more the amount of same numbers presented, the greater the difficulty to respond (Morton, 1969).
Although this study found similar results to Klein’s experiment (1964), with no significant outliers, the differences in experiments should be noted. For example, Klein used 90 participants (15 within each condition), whereas this study consisted of 38 undergraduate students from a single university. Therefore, despite finding support for the existence of the Stroop effect, there are limitations that could reduce support for the results. The evident low population validity therefore makes the findings difficult to generalize to the wider population. Additionally, the population of students could imply previous effects of practice or training at the task, which may have been encountered before. As a result, this could influence the response time to the colour stimuli when conflicted with the written form. Though on the other hand, possible factors such as sex differences in naming colours were not noted, so perhaps the known cognitive and structural brain variances between the genders could have been a confounding variable.
Furthermore, it could be argued that the stimuli conditions did not accurately represent what was intended, such as the ‘neutral’ condition, whereby words such as heart may imply a certain colour (red). This reduces the internal validity of the study, as there was not clear measurement of non colour-related words. Alternatively, there is a lack of consideration for individual differences, whereby certain non-related colour words may be associated with previous experiences, possibly associated with colour. Another factor that may have influenced the participant’s response is word length. In this case, words that had three or six letters were not separated, which could influence the reading time and thus, be the cause of result differences. Similarly, the environmental setting in which the experiment was conducted reflects low ecological validity, since the task and surrounding is quite atypical for everyday life. Though possible noise was overlooked, which could distract the participant during performance and reduce the reliability of the findings. Despite these limitations, the design was appropriate to the aim, especially since participants took part in separate trials (experimental condition), preventing influence of prior experience. Additionally, as in Klein’s experiment (1964), participants were matched in reading speed of colours and sorted into groups, which reduces possible outliers from individual differences in reading times.
In terms of future research, factors such as possible interferences in attention should be explored, as the results may have been influenced by how motivated the participants were to respond. Perhaps brain-imaging technology would enable researchers to accurately measure the psychological processes behind responding to words associated with colour or whether brain areas related to attention were activated. For a holistic perspective, the Stroop effect should be investigated in relation to spatial tasks, such as the location of a word (top, side or bottom) rather than focusing on the stimuli itself. Alternatively, the verbal qualities of the task may be significant, especially in exploring how the pitch in which words are responded in could contribute to the Stroop effect (Hamers & Lambert, 1972). It was found that if ‘High’ pitched words were said in a high tone, the task was facilitated, rather than reading words in the opposite tone. Therefore, further research should take into account any extraneous or confounding variables that could influence colour naming, rather than by the graded interference effect. Overall, while taking into consideration the strengths and limitations of this research, it can be concluded that the Stoop effect is valid as the evidence presented establishes that reading was faster than naming the colour.
Stuck on your essay?
