Directional stroop test

Abstract

This study assessed differences in gender performance between males and females for reaction time and error rate on the Directional Stroop Test. To date no study has explored the relationship between gender performance and the Directional Stroop Task, although prior studies have shown that females perform better than males on the Classical Stroop Test. 50 university students (25 males and 25 females) were exposed to 3 stimuli types, in the first they were required to locate the position of the word on the screen by pressing a button to indicate their response, in the second stimuli type participants were required to locate the position of the arrow shown and in the third they were required to select the direction the arrow was pointing in. Findings showed that although females respond faster to the stimuli presented to them and males make fewer errors, overall gender does not have a significant effect on reaction time and error rate.

The Effect of Gender, Arrows and Words on Reaction Time and Error Rate

The Stroop Effect is widely thought to cause interference because the information of one stimuli interferes when a person is trying to process another stimuli. An example of the Classical Stroop Effect is if the word red was presented to a person in the ink colour blue; the person would automatically try to read the word but the task requires for them to read aloud the ink colour the word has been presented in, this causes interference and results in a slower response. Interference is thought to occur because the process involved in naming words is thought to be automatic, compared to the process involved in naming colours or pictures which is thought to require a voluntary effort (as cited in Macleod, 1991).

The famous Classical Stroop effect is named after J. Ridley Stroop who discovered this effect in the 1930s. In Stroop’s first experiment, he observed that his participants were slightly slower when reading words which were printed in an incongruent ink colour, that is to say that the word did not match the ink colour it was printed in, for example the word blue printed in green ink, compared to reading words which were printed in black ink; this is a demonstration of the effects interference can have on reaction time (as cited in Wuhr, 2007).

There have been countless replications in addition to many variations of the Classical Stroop study. In 1977 Wheeler conducted a study in which 36 participants, all of which were college students, were required to first read words aloud and then name the ink colour aloud, they then had to repeat this but indicate their response by pressing buttons. Wheeler found that it took participants twice as long to name ink colours than read the word aloud, however there was no time difference when participants pushed buttons to indicate their responses; interference may occur in the Classical Stroop Test because participants respond verbally (as cited in Wheeler, 1977).

There are two theories which attempt to explain the cause of interference which occurs in the Stroop Effect. The first is Speed of Processing Theory which suggests that words are read faster than colours are named and that because of the speed that we read it makes it difficult for us to name the colour the word is printed in, this difficulty causes interference which effectively results in a slower reaction response or incorrect answer. The second theory is Selective Attention Theory which proposes that naming colours requires more attention than reading words, this may be because people are so fluent at reading and therefore this is more of an automatic process, whereas naming the ink colour of a word is not an activity one comes across everyday and therefore this requires more of our attention (as cited in Stroop, 1935).

The Selective Attention Theory is consistent with the distinction between two basic types of processing information. Schneider and Shriffin suggested that information can be processed either automatically or in a controlled way. Automatic processing is thought to be involuntary, effortless, fast and requires minimal attention, on the other hand controlled processing is thought to be voluntary, relatively slow and requires more effort and attention (as cited in Schenider and Shriffin, 1977). The Stroop Task has achieved broad recognition as a way to differentiate between controlled and automatic processes. As cited in Cothran and Larsen (2008) processing tasks may become automatic from exposure to repeated experiences, these repeated experiences of particular stimuli strengthens the memory representation a person has which eventually leads to less effort and attention being required. Some psychologists have questioned whether interference can be overcome or reduced by practice; in a study by Shor, Hatch, Hudson, Landrigan and Shaffer (1972) researchers observed that interference was still present even after participants had training and practice of between 30-50 days (as cited in Shor, Hatch, Hudson, Landrigan & Shaffer, 1972). However, in another study researchers found that older and younger participants displayed a decrease in interference for the colour-word Stroop test after humdreds of practice trials, although older participants consistently displayed greater interference effect throughout the practice compared to younger participants (as cited in Davidson, Zack & Williams 2003).

The type of Stroop variation which has been developed for this study is the Directional Stroop, an example of this is if the word “right” was presented on the left hand side of a screen and the participants were asked to locate the position of the word whilst ignoring its content. One of the first studies investigating the Directional Stroop Effect was conducted by Shor (1970,1971) who in the first condition inserted the words left, right, up and down into arrows, which were pointing in the directions named but were not congruent to the name in the arrow, for example, an arrow may be pointing right but it will have the word “left” embedded into it. Shor observed that interference was present when participants identified the words with a manual response, but not when they identified the words with a vocal response. In the second condition of the study Shor only used arrows and observed that interference was present when participants identified the arrows using both manual and vocal responses, but this interference was small in relation to the Classical Stroop. This study also indicates the importance of the type of stimuli being used in studies, arrows have produced interference for both manual and vocal responses compared to arrows which have directional words embedded into them (as cited in Baldo, Shimamura & Prinzmetal 1998).

Furthermore, in a study by Clarke and Brownell (1975) participants were asked to decide whether arrows were pointing either up or down whilst positioned inside a rectangle; in a study by White (1969) participants were asked to name the position of the words North, East, South and West inside a rectangle; both studies found similar results to Shor, with small but reliable interference.

There have been many studies which have included variations within the Directional Stroop Test (Greenwald 1977; Dyer 1972; Pieters 198; Roefols, Turrenout & Coles 2006). In studies done by Seymour (1973/1974) and Palef and Olsen (1975) participants were asked to name the locations of words in relation to a fixation point, both studies noted that interference was present (as cited in MacLeod, 1991).

Many studies have also looked at the link between gender and Stroop performance. Even before the Stroop task, a psychologist named Ligon (1932) claimed that girls were able to name colours faster than boys, although he did not note any difference in their word-reading speed ability (as cited in MacLeod, 1991). This claim may be supported by some of the findings that researchers have discovered in their studies; one study found that females performed better than males on 2 colour card tests (Golden 1974), another study found that females gave a significantly higher performance than males after both sexes completed the colour-word test 5 times (as cited in Sarmany, 1977).

According to Baroun (2006) females have been found to have an advantage over males for colour recognition and have been reported to be quicker at reacting to Stroop tasks; another psychologist suggested that this may be due to the general response speed females have (as cited in Jensen, 1965). Other psychologists such as Shen (2005) suggested that males take a longer than females on the Stroop task because males have a different cognitive strategy to females (as cited in Shen, 2005).

There has always been controversy over whether there are differences in gender performance on the Stroop test and many studies have attempted to test whether differences in gender performance exist or not. In a study by Baroun and Alansari (2004) 210 university students took part in the Stroop test, 96 were male and 114 were female, although Baroun and Alansari found that Kuwaiti students (140 of the participants) had greater interference than British students (70 of the participants), no gender differences were observed for Stroop interference (as cited in Baroun & Alansari, 2004). However in 2006, a study by the same researchers found differences in gender performance in Kuwaiti males and females on the Stroop test, researchers found that Kuwaiti females were faster on the colour card Stroop than Kuwaiti males (as cited in Baroun & Alansari, 2006).

The dependant variables for this study are reaction time and error rate and it is important to research what previous investigators have found in relation to reaction time and error rate, as this provides valuable contributions as to what should be expected. Mekarski, Cutmore and Suboski (1996) tested 6 males and 8 females in a Stroop test in which reaction time and error rate was measured. Researchers found that males were consistently slower than females but error rates did not differ significantly (as cited in Mekarski, Cutmore & Suboski, 1996). Boyden and Gilpin (1798) established that errors made on 2 tests were correlated positively for males but not for females, they indicated that this may be due to males being linked distractability and being impulsive. Many psychologists have suggested that although females may be faster in naming colours for the Stroop Test this does not relate to the measuring of interference (as cited in Macleod, 1991).

As mentioned previously the type of stimuli used in studies is important. In one study 36 undergraduate students were required to respond to one stimulus whilst ignoring another, so if presented with an arrow participants were required to ignore the word stimuli, and if presented with a word then they were required to ignore the arrow stimuli. Researchers found greater interference for participants when they responded manually to the word stimuli whilst ignoring the arrow stimuli as well as when they responded vocally to the arrow stimuli whilst ignoring the word stimuli. Although interference was present for both types of stimuli the magnitude of interference seems to be dependent on response modalities; vocal or manual (as cited in Baldo, Shimamura & Prinzmetal 1998).

Many of the studies investigating the effects of gender performance that have been discussed have tested reaction time but not error rate. This study has been carried out because there is a lack of studies investigating the difference in gender performance for the Directional Stroop, although there are handfuls which have done so for the Classical Stroop. This study aims to explore the Directional Stroop Effect using words and pictorial targets in the form of arrows and its effect on reaction time and error rate between males and females. This study will examine whether interference occurs in participants for both words and pictorial targets and whether gender has any effect on reaction time and error rate. The hypothesis for this study is that there will be a significant difference in performance between males and females, which will be measured by reaction time and error rate. A further hypothesis being tested is that stimuli type will effect reaction time and error rate.

Method

Participants

The participants used for this study were 50 University of Greenwich students (25 males and 25 females) who were obtained through stratified random sampling method; a sampling method which allows individuals to be equally selected for participation therefore reducing bias, but also selecting individuals who belong to a certain sub-group, this study only required 25 males and 25 females. This study was conducted in the U.K.

Materials

The materials used for this study include a borrowed laptop from the University of Greenwich with the software application Superlab. Superlab is an effective application for presenting visual stimuli on the screen, and was used in this study to present word and arrow stimuli in addition to recording the reaction time and error rates of participants to explore the performance level between males and females.

Design

This is a quantitative study and a repeated measures design was used to reduce individual differences and test the effect of 3 factors at the same time. The independent variables are gender and stimuli; gender has 2 levels (male and female) and stimuli has 3 levels (word-location, arrow-location, arrow-direction). The dependent variables for this study are reaction time (in milliseconds) and error rate. This study has used a correlation analysis which describes how one variable relates to another, in this study we are looking at whether gender has an effect on the reaction time and error rate of participants performance and whether stimuli type can affect reaction time and error rate.

Procedure

Participants were first given an information sheet detailing the procedure of what they would have to do and any ethical risks involved in the study and were then asked to sign the consent form declaring that they had read and understood the details of the study. To maintain confidentiality of records participants were asked to invent a personal code, known only to them and which would be used to identify their data should they choose to remove it. Each participant’s personal code was entered into Superlab before they started their study to ensure that data recorded concerning their reaction time and error rate would be saved under their personal code. Participants were informed that there would first be a practice session to ensure they understood what to do; the practice session consisted of 2 slides, after the practice session participants began with the first condition of the study which was responding to where the location of a word on the screen was by pressing a button. The words used in this study were up, down, left and right; participants were presented with a RB-530 response pad (see appendix a) which they used to indicate their responses. The second condition of the study was responding to where the location of an arrow was on the screen by press the correct button on the response pad and the third condition was participants responding to the direction in which the arrow on the screen was pointing; each of the conditions consisted of 10 slides, participants reaction time was measured from when the stimuli appears on the screen to when participants press any button. After the experiment participants were provided with a debrief sheet containing details of what the study was investigating and contact details of the investigator.

Results

The raw data (Appendix B) was analysed using PASW Statistics 17. A mixed design was performed on reaction time and error rate separately. The between subject factor was gender (male/female) and the within subject factor was stimuli (word-location/arrow-location/arrow-direction).

Discussion

The results show that females have performed faster and made fewer errors for the word-location stimuli compared to males, similarly males have performed faster and made fewer errors for the arrow-direction stimuli. However, arrow-location is the stimuli in which both males and females have scored high, one of each dependant variable; males made the least errors but females gave the fastest responses. Results also show that stimuli type affects participants reaction time significantly, but does not have a significant effect on error rate, additionally results showed a weak stimuli X gender interaction on reaction time in particular and a significant stimuli X gender interaction on error rate, however gender alone does not seem to have a significant effect on reaction time or error rate.

The results from this study indicate that overall women respond faster to the Directional Stroop and that males and females do not differ significantly in terms of error rate. The findings show that arrow-direction is the stimuli in which participants take the longest to respond and also tend to make the most errors, this may be explained by the Selective Attention Theory, which may suggest that naming the direction of an incongruent arrow may require more attention as the participants have to focus on the point of the arrow, compared to naming the location which does not require as much focus on the stimuli, but more on its location on the screen.

Many psychologists and studies (Ligon 1932; Golden 1974; Sarmany 1977; von Kluge 1992; Baroun 2006; Baroun & Alansari 2006) have previously implied that females respond faster than males, additionally Jensen (1965) proposed that this may be because of the general response speed females have. The results from this study are similar to the findings in the study performed by Mekarski, Cutmore and Suboski (1996) who found that males responded slower but there were no significant differences between males and female for error rate. The results for the present study show that overall females responded the fastest and males made the fewest errors.

50 University of Greenwich students were used as participants for this study and 50 participants can be assumed to be a small and unrepresentative sample of the general population. For future studies a larger sample could be collected and participants could be selected for their age as well as their gender, age would be interesting variable to examine and investigate how people from different age groups differ in performance.

This study took place in the U.K. and therefore lacks in ecological validity, the participants from this study were all obtained from the U.K. and were all volunteers therefore it is difficult to draw up appropriate conclusions about cause and effect relationships because the findings from this study may not be generalisable to people from other countries. Baroun and Alansari (2004) found that Kuwaiti students displayed greater interference than British students, this is a good example of how the present study may not be generalisable to other people. However, the study Baroun and Alansari conducted in 2006 did find similar results to the present study; females were reported to respond quicker to the stimulus presented to them than males, therefore there may be certain variables which have been found world-wide and others which still need to be investigated. The experiment also lacks ecological validity because it was not a realistic situation, people would not usually press buttons on a response pad indicating their answers to what they see on a laptop screen.

In this study all the participants started with the word-location stimuli and then moved to the arrow-location stimuli and then finally finished with the arrow-direction stimuli, this may have led to the order of stimuli being confounded. Participants have been reported to respond slowest and make the most errors to the arrow-direction stimulus, but participants may have been affected by an extraneous variable which may have effectively slowed down their reaction time and therefore affected their results. Extraneous variables are situations or factors which can have an impact on participants performance and results. In future replications of this study the order of presenting stimuli could be counterbalanced which is a method used to avoid the confounding of variables. Participants could be divided into different groups and each group could be shown the stimuli in a different order.

Before the actual experiment started, participants were given the opportunity to experience a practice session, however as this practice session was only 2 slides long and only used the word-location stimuli participants may not have understood or had the chance to properly understand what they were asked to do. For future replications of this study it can be suggested that participants experience at least 2 slides of each stimuli type which they will be presented to them in the real experiment so that they can actually practice properly and have have a real opportunity to understand what there are being instructed to do.

Another limitation of this study is that slides showing each stimuli one by one flows straight to the next one as soon as the participant presses a button on their response pad. Participants are informed at the beginning of the study that they must try to respond as quickly as they can, and in haste some participants may have missed the slide which describes the next set of instructions for the next stimuli type. Due to this error participants may respond slowly because they are suddenly exposed to an arrow instead of a word which they have been exposed to in the previous slides, in addition to this if participants miss the slide informing them to switch from selecting the location of the arrow to the direction the arrow is pointing in then they may make errors, which may as a result be due to misunderstanding of instructions. It may be more effective if after every stimuli type is presented there is either a loud beep sound indicating the start of a new stimuli type or for each stimuli type to be opened in a separate window as to avoid any confusion. Future replications should also include congruent stimuli to begin with so that reaction time and possibly error rate can be compared to when participants are shown uncongruent stimuli.

Unfortunately the eye-sight of participants were never tested before the study therefore there is no guarantee of whether any of the participants were long sighted and may have had difficulty viewing what was on the screen. Although the words and pictorial targets used were of a large size, future studies should ensure a check-up of participants eye-sight before the study to ensure that this variable does not affect the outcome of the results.

As discovered in the study conducted by Baldo, Shimamura and Prinzmetal (1998) response modalities can have a great impact on the amount of interference observed. Future studies into gender and the Directional Stroop could use both manual and vocal responses from participants, this would help expand knowledge on how type of response could affect interference. In addition to this the present study presented words and pictorial targets separately, future replications could include a condition which presents words and pictorial targets combined as well as separate.

This study used a quantitative design method which is statistically reliable as well as replicable. However the results provided are also likely to be limited as they provide a less descriptive and detailed account of the actual human behaviour which takes place during interference for the Stroop test. It is believed that for this study a quantitative approach was the right design method, there are no studies which could be found, which have investigated gender and its effect on reaction time and error rate for the Directional Stroop therefore any future psychologists who wish to investigate these variables and effects further will be able to easily replicate the methods used for this study.

A future replication of the Directional Stroop Effect could use a matched pairs design in which group one are tested as if they are taking the hazard perception practice section of a driving test, participants would need to locate the uncongruent directional words and pictorial targets, which could be in the form of an arrow or other road signs. The second group of the matched pairs design study would also need to locate the same uncongruent directional words and pictorial targets but would also have a distractor task such as trying to text someone from a mobile phone at the same time they are locating the uncongruent stimuli. This type of study should be able to provide some insight into the type of interference drivers experience and which type of distractors, such as making a call, sending a text message to someone, or even applying make-up, cause the most accidents or interference.

In conclusion, the results support the hypothesis that stimuli type affects reaction time, although stimuli type does not seem to have any affect on the error rate of participants. In consistency with some previous studies (Merkarski, Cutmore & Suboski 1996; Baroun & Alansari 2004) the results from the present study has found that females respond faster than males, which may be explained by Jensen (1965) who suggests that this is possibly related to the general response speed of females. Results also show that males make fewer errors than females, however there is no overall significant difference between gender performance for the Directional Stroop Effect.

References

• Baldo, J.V., Shimamura, A.P. and Prinzmetal, W. (1998) Mapping symbols to response modalities: Interference effects on Stroop-like task. Perception and Psychophysics, Vol. 60 (3), 427-437.
• Baroun, K. and Alansari, B. (2004) Gender and culture performance differences on the Stroop colour and word test: A comparative study. Social Behaviour and Personality, Vol. 32 (30), pp.235-246.
• Baroun, K. and Alansari, B. (2006) Gender differences in performance on the Stroop Test. Social Behaviour and Personality, Vol. 34 (3), 309-317.
• Cothran, D. and Larsen, R. (2008) Comparison of inhibition in two timed reaction tasks: The Colour of Emotion Stroop Tasks. Journal of Psychology: Indisciplinary and Applied, 142 (4), 343-385. Retrieved from PsycINFO database.
• Davidson, D., Zack, R. and Williams, C. (2003). Stroop interference, practice, and aging. Aging Neuropsychology and Cognition, Vol. 10 (2), 85-98.
• Golden, C.J. (1974) Sex differences in performance on the Stroop colour and word test. Perceptual and Motor Skills, Vol. 39 (3), 1067-1070. Retrieved from PsycINFO database.
• Jensen, A.R. (1965). Scoring the Stroop Test. Acta Psychologica, Vol. 24, (5), 398-408.
• MacLeod, C.M. (1991) Half a century of research on the Stroop Effect: An Integrative Review. Vol. 19 (2), 163-203. Psychology Bulletin.
• Mekarski, J.E., Cutmore, T.R. And Suboski, W. (1996) Gender Differences during processing of the Stroop Task. Perceptual and Motor Skills, Vol. 83 (2), 563-568.
• Sarmany, I. (1997) Different performance in Stroop’s Interference Test from the aspect of personality and sex. Studia Psychologica, 19, (1) 60-67. Retrieved from PsycINFO database.
• Schneider, W. and Shriffin, R.M. (1977) Controlled and Automatic Human Information Processing: Detection, Search and Attention. Psychological Review, 84, pp. 1-66.
• Shen, X. (2005) Sex differences in perceptual processing performance on the colour-kanji Stroop Task of visual stimuli. International Journal of Neuroscience, Vol. 115 (12), 1631-1641.
• Shor, R.E. Hatch, R.P., Hudson, L.J., Landrigan, D.T. and Shaffer, H.J. (1972) Effect of practice on Stroop-like spatial directions task. Journal of Experimental Psychology, Vol. 94 (2), 168-172.
• Stroop, J.R. (1935) Studies on interference in serial verbal reaction. Journal of Experimental Psychology, 18, 643-662.
• von Kluge, S. (1992) Trading accuracy for Speed: Gender Differences on a Stroop task undder mild performance anxiety. Perceptual and Motor Skills, Vol. 75 (2), 651-657.
• Wheeler, D.D. (1977) Locus of interference on the Stroop Test. Perceptual and Motor Skills, Vol. 45 (1), 263-266.
• Wuhr, P. (2007) A Stroop Effect for Spatial Orientation. The Journal of General Psychology, 134, (3), 285-294.