The Counting Stroop And The Interference Effect

An experiment based on the work of Bush, Whalen, Rosen, Jenike, McInerney and Rauch (1998), was performed to determine whether there would be a difference in the reaction times of the participants when they counted number words (e.g., “two written three times”), and animal words (e.g., “bird written twice”). In the present experiment participants were tested under two different conditions: Number Condition (e.g., “33333”), and Control Condition, which contained letters or symbols (e.g., “@@@). The results of the experiment showed that participants reaction times while counting letters or symbols was faster than when counting numbers. Therefore, there is a significant difference in reaction times under the two conditions.

Introduction

Extensive research has been done over the years about the interference effect in cognition. Cattell (1886) reported that objects (and colors) took longer to name aloud than the corresponding words took to read aloud. Saying “red” to a patch of color was slower than saying “red” to the word red. His explanation was that in the case of words and letters the association between the idea and the name has taken place so often that it has become automatic ( cited by MacLeod 1991, p.163)

One of the most well known experiments in this aspect is Stroops’ (1935). In this experiment, Stroop first compared reading a list of words printed in black with reading the same list of words printed in incongruent colours. Stroop found that there was little difference in reading time for the two lists. He then compared the naming of colours for a list of solid colour squares with the naming of colours for a list of words printed in incongruent colours. Subjects averaged 74% longer to name ink colours of incongruent words. This led Stroop to conclude that there is less interference with word reading than with colour naming.

Earlier studies by Muller and Shumann (1894) argued that more time was needed to read a series of nonsense syllables when the stimulus syllables were associated with other syllables in the meantime (cited by Stroop, 1935 p. 643). A quote by Kline suggests the above method, “If a is already connected with b, then it is difficult to connect it with k, b gets in the way” (Kline, 1921, p. 270). Nonsense syllables were also used by Shepard and Fogelsonger (1913) in a series of experiments in association and inhibition. Only three subjects were used in the experiments and the changes introduced to produce the inhibition were so great in many cases as to present new situations. The results showed an increase in time of response due to the increase in the complexity of the situation.

In card sorting Bergström (1893 and 1894), Brown (1914), Bair (1902), and Culler (1912) found that changing the arrangement which the cards were being sorted produced interference effects. Bair and Culler also found that interference with opposing habits decreases with practice. Culler (1912), carried out two experiments to suggest this theory. In one experiment the subjects were shown a series of numbers and they had to press a particular key on the typewriter with a particular finger. Then the keys were changed, and the numbers had to be written with different fingers. In the other experiment the subjects were trained to react with the right hand to ‘red’ and with the left hand to ‘blue’ and then the stimuli were interchanged. Both experiments found that interference decreased with practice.

More recently the Stroop interference effect has been aid of many neuropsychological studies that have tried to detect neurological problems in the context of attention and it has been widely used in the neurological, neurosurgical and neuropsychiatric spheres (MacLeod, 1991). Several paradigms have shown that people with anxiety disorders have selective processing of threat cues (Dobson, 2010). An experiment by Mathews and MacLeod (1985) used a variant of the Stroop effect called the emotional Stroop colour-naming task and found that anxious subjects who were worried about physical harm took longer than controls to colour-name threatening words (“disease”, “coffin”) as opposed to neutral (“holiday”, “welcome”).  Whalen, Bush, McNally, Wilhelm, McInerney, Jenike and Rauch (1998) used the emotional counting Stroop and they proposed that it will be a useful fMRI probe of anterior cingulate (brain region involved in the integration of emotional and cognitive information) affective division function in anxiety disorders. Several studies have consistently shown that anxious subjects are more attentive for or have more difficulty discarding threat related stimuli than normal controls (Bar-Haim, Lamy, Pergamin, Marian, Bakermans-Kranenburg and Marinus (2007).

The present experiment was based in another variant of Stroop effect which is called the Counting Stroop. Bush et al. (1998) used the newly developed ”Counting Stroop” to identify the mediating neural substrate of cognitive interference. The Counting Stroop, allowed on-line response time measurements but speech was prohibited. The reason for that was because speaking produced head movements that could prevent the collection of vital performance data by the functional magnetic resonance imaging (fMRI). During this task, subjects reported by pressing the button when the number of words one through four came on the screen. Interference trials contained number words that are incongruent with the correct response (e.g., ”one” written three times), while normal trials contained a category of common animals (e.g., ”dog”). Nine normal adults took part on this experiment and their reaction times were measured.

In this Counting Stroop experiment there are also, two conditions. First one is the Number Condition and the second one is the Control condition. Here, the interference trial contained numerical numbers that are incongruent with the correct response (e.g., “1111” written four times), while normal trials contained symbols or letter (e.g. “bbb”). Then the participants had to take part in both conditions and their times were measured.

Method

Design

Students were exposed to two conditions: one containing numbers and the other control characters (symbols or letters). These are the independent variables. Since all the participants competed both tasks this is a within subject design. Each participant was asked to write down the number of characters in each condition separately. Then, their time of response was measured, so time is our dependent variable.

Apparatus and Materials

The test consisted of one double sided paper with a list of 20 characters for each condition. The first condition contained numbers (e.g. “444”) and the second condition contained symbols and letters (e.g. “@@” or “bbbb”). However, for each condition there were some rules. In the Number Condition, lines could not contain the same number written as the number of the characters (e.g. “4444”). In Control Condition, characters could not have letters or symbols mixed together (e.g.”aaa??”, “aabb”).

Participant response times were measured by a chronometer.

Participants and Procedure

Ninety three undergraduate students at Swansea University were tested (69 female and 24 male) aged between 18 and 47 (M= 19.77; SD=4.97).

The subjects were separated into groups of four. They were asked to write down numbers (Number Condition) in the first paper and control characters (Control Condition) in the second paper. Then they had to count how many characters or numbers each line contained and to write it down as fast as they could. While one participant was completing the questionnaire, another one was measuring the time. Each participant had to test his own model to another two participants, so each one had to complete the questionnaire twice.

Results

Table 1 shows the mean number of time in seconds that took for the participants to complete each condition and relatively the number of errors made.

Table 1. Reaction time and number of errors in each condition.

Mean

SD

Number of errors

% of errors

Number Condition

31.63

9.32

56

3.0

Control Condition

30.21

9.32

22

1.18

In this experiment it was found that t=2.25 with df 92. Comparing it to the critical value of t=1.984 we can conclude that the difference in the means of the scores in each condition is significant. Therefore, participants’ reaction time was significantly quicker when they had to count control characters.

Discussion

The main result of this study was that participants reaction times under the number condition were slower than in control condition and also there was a greater percentage of error when completing the number condition task (3%), while control condition was (1.18%).

Our results did not appear to be different from what Bush et al. (1998) obtained. In their experiment the results showed that there were longer reaction times in the interference blocks than for neutral ones, and they also found that there was a decrease in reaction times with practice during interference trials (diminished interference effects) which indicated that learning occurred. However, the percentage of error was slightly higher in our experiment. We can assume that the reason for that was that the participants in Bush et al. (1998) experiment were instructed to answer as quickly as they could but without sacrificing accuracy for speed and they had previous knowledge about the experiments when they were doing the trials.

Although we have demonstrated that the counting Stroop produced a cognitive interference as revealed by the results in number condition, our experiment probably would not be considered very reliable. Every participant did not have the same tests, since all ninety three subjects made their own version of the test. Also they knew before hand that the number in Number Condition could not be the same as the number written down (eg. “22” twice), and this probably raised their judgment. The noise level in them room was very high and this may have had affected their attention while completing the task. The participants before starting to complete the task they had a view of the first few numbers or symbols, letters so this could have an affect in their overall reaction times. Finally there was no prior knowledge of the condition of the participants and if they had any defects (eg. Vision problem).

In future studies it would be useful to take all of the above into consideration. All participants should complete the same model of the test. The test should be taken in an isolated room without any distraction or interference. Any additional information that would help the participants’ judgement should not be given before taking the test. The ratio of male to female should be even so we can have a comparison between the two genders. Finally all the participants should be healthy and should be in the same level of knowledge. For further researches could be analyzed making distinctions between participants ages and, maybe also, between answer of males and females to see if they differ in time of response.