The Stroop Effect and Colour-Related Words

An experiment was carried out to test if the Stroop effect occurred when a small but significant modification to the conditions was applied to the classic Stroop experiments. Previous evidence suggested that although automatic and controlled processes can work simultaneously, they can cause undesired interferences. In this experiment, colour names were replaced by colour-related words in the Stroop condition and it was found that the Stroop effect still occurs, suggesting that characteristics of the meaning of the word are automatically processed and perceived. INTRODUCTION

Our senses (visual, auditory, tactile and so on) detect external stimuli. This sensory information is processed by internal cognitive processes and result in perception (Edgar, 2007, p. 3). We have limited brainpower capacity to process all the information available to us through our senses (Edgar, 2007, p11) and for our brains to allocate appropriate resources, this information gets filtered by cognitive processes. One of these processes is attention. Attention acts like a filter between what is sensed and what is perceived and allocates sensory information processing resources as appropriate.

Several studies have been conducted in the area of attention. In 1954, Donald Broadbent suggested that the attention system works like a “bottleneck” that only allows a limited amount of sensory information through (as cited in Edgar, 2007, p. 17). He further suggested that this “bottleneck” chooses the information based on its type and physical characteristics, such as location or sound pitch. Posner (1980) suggested that “spotlight” would be a more appropriate analogy (as cited in Edgar, 2007, p. 15), as the area illuminated by it receives priority of processing resources.

Eriksen & Murphy, 1987, suggested that a “zoom lens” was still a more accurate analogy as attention can be zoomed in on a small area and thus its information processing would be focused or zoomed out to process a wider area of sensory stimuli (as cited in Edgar, 2007, p. 15). Schiffrin & Schneider (1977) conducted a series of experiments which led them to distinguish these attentional processes between controlled and automatic. Controlled processes are those which we can consciously control whereas automatic processes are those which do not require conscious

awareness (as cited in Edgar, 2007, p. 20). Shiffrin and Schneider suggested that automatic processes require little or no attentional resources and can therefore occur without conscious awareness (as cited in Edgar, 2007, p. 20). Although this unconscious perception happens under the conscious level of awareness, it may still influence our reactions and behaviour (Edgar, 2007, p. 6). As controlled and automatic processes can operate simultaneously, one or the other can more strongly influence our behaviour and reactions thus causing the Stroop effect.

In 1935, Stroop devised an experiment to demonstrate this interaction between the two processes where participants are presented with a list of words written in different coloured inks (Edgar, 2007, p. 21). In the Stroop condition, the words are colour names and the ink colour and name colour do not correspond. In the other condition, the words are colour neutral and printed in the same colours as in the Stroop condition. The participant is to name the ink colours as quickly as possible. Although this is quite easy to do in the colour-neutral condition, the Stroop condition encounters interference from the colour name, an automatic process.

In this experiment, we will be investigating whether colour-related words in the Stroop condition have a similar effect to that found with colour-name words, as used in classic Stroop experiments. Do characteristics associated with the meaning of the word (precious knowledge of the word) slow down the participants’ responses? Does the knowledge of the word trigger automatic processes overriding controlled processes? Or perhaps the word meaning has no effect on automatic processes. METHOD DESIGN In this experiment, a within-participant design was employed to investigate a variation of the Stroop effect.

The independent variable was interference in controlled processing by the automatic processing of external stimuli. Two conditions were used. Participants had to work through two lists of thirty words each naming the colour in which the words were written. In the experimental condition, the list consisted of colour-related words printed in incongruent colours and, in the control condition the words used were colour-neutral, printed in the same colours as the corresponding word on the experimental condition’s list. The dependent variable was the time taken to name the colours of all the words on the lists. The time was measured by the

researcher to the nearest second with a stopwatch, producing interval data. To minimise practice effects, the conditions were presented to the participants in alternate sequence. The words used in both lists were matched in length, initial letters and colour. The words were also each repeated the same amount of times in both lists. PARTICIPANTS Sixteen of the twenty two participants that participated in this experiment were recruited amongst colleagues at The Open University, including family members and friends. The remaining six participants were recruited by asking for volunteers at the researcher’s place of work and residence.

The ages of all twenty two participants ranged from 18 to 69 and there were 12 women and 10 men. With regards to the nature of this research, all participants were informed that they would be asked to name the colour of words written on two lists and be timed in order to investigate automatic processing and to provide data for the researcher’s TMA. The participants were informed of their right to withdraw from the experiment at anytime during and/or after the experiment without providing a reason and that their data would be anonymous.

None chose to withdraw. Twenty two participants gave fully informed consent. The participants were not aware of the hypothesis of this experiment. All participants were debriefed after their participation. (PROCEDURE) MATERIALS In this experiment, the stimuli presented to the participants were two lists, each with thirty words printed in six different colours. Each list consists of six words repeated five times, each of the times printed in a different colour. For the experimental condition, the colours used were incongruent to the word, i. e.

the word sky was printed in red, yellow, green, purple and orange, but not in blue or the word lemon was printed in all colours except yellow. For the control condition, a list of colour-neutral words was used. Great care was taken to make the lists as similar as possible in order to minimise any arising confounding variables. The length of the words was matched to the corresponding words in the experimental condition list, as well as the colour the word was printed in and its initial letters. To measure the time the participants’ took to go through the list to the nearest second, a stopwatch built into a mobile phone was used.

Standard instructions were devised to be read out to participants prior to their participation. These stated the order in which the experiment would take place and what was required of the participants. Consent forms were also created for the participants to sign. PROCEDURE Participants were individually approached and asked to volunteer to take part in a psychological experiment that would take no more than 10 minutes. All participants who agreed were briefly explained what the experiment would entail and what they were required to do.

Participants were told this experiment was being conducted to collect data for the researcher’s TMA and that it would take place individually and not in group. They were informed of their right to withdraw and that their data would be anonymous. They were asked for their age and gender for classification purposes only and given a consent form to sign. Once the consent form was signed, participants were told they would be presented with an A4 sheet containing two columns of words written in six different colours.

The six colours used were named and the participants told they were required to go through the list of words naming out loud the colours the words were written in as fast as they could. An example was given with two words – chair (written in blue) and house (written in red). When the participants confirmed they understood what was required of them, an A4 sheet with the first condition was placed in front of them. They were timed with a stopwatch while following the instructions given. Once completed, the participants were asked to repeat the process with another A4 sheet, containing the second condition.

The sheets were presented to participants in alternate order, i. e. participant 1 was presented with condition 1 sheet and then condition 2 sheet, participant 2 was presented with condition 2 sheet and then condition 1, and so on. The times taken with each condition were annotated. Participants were individually thanked and debriefed with an explanation of what was being measured and the research hypotheses. They were also asked if they had any questions or concerns about the experiment. None did. RESULTS

The research hypothesis of this experiment was that conflict can be caused between automatic and controlled processes by colour-related words and not only colour-named words as found in classic Stroop experiments, and, as a result, participants would take longer to perform the task with the Stroop condition. The performance in each condition was measured to the nearest second. As can be seen on Table A in appendix 3, there is a difference between the mean responses of the two conditions. The mean of the Stroop condition was 2. 8 seconds, just over 12 percent longer than the colour-neutral condition.

The data collected was entered into SSPS software and a paired sample t-test was conducted. This test revealed that the difference between these conditions was statistically significant (t(21)=4. 109; p=0. 001; d=0. 6750) thus suggesting that the Stroop effect also occurs with colour-related conditions and so, the null hypothesis, that there is no difference between conditions, can be rejected. Descriptive statistics also show that the response times measured during the Stroop condition were longer than during the colour-neutral condition. DISCUSSION

The results obtained through this experiment suggest that the automatic processing of reading colour-related words interferes with the controlled processing of naming the colour of the ink the words are written in. The results also allow for the null hypothesis to be rejected. They indicate a difference between conditions and dismiss the cause of this difference to be due to sampling error. The data collected was statistically significant. The average time taken to complete the task with the Stroop condition was longer than with the colour-neutral condition by almost 3 seconds.

The results seem to indicate that the automatic process of reading does not only perceive the exact synonym of the word being read but also characteristics of the word. It seems that when a participant automatically reads the word “sky” they will not only perceive it for what it means, the sky above us, but also the colour blue, or in the case of the word “grass”, not just the vegetation in our gardens, but also the colour green. The longer time taken to complete the task with the Stroop condition suggests that more attention is required to prevent the automatic process from taking over.

The results of this experiment support Schiffrin and Schneider’s distinction between automatic and controlled processes and two-process theories. When collecting the data from participants, it was noted that, towards the end of the Stroop condition task, participants would significantly slow down and even briefly interrupt their responses. This seems to indicate that although the participants start the task by focusing on the ink colour, the focus widens, or weakens, and the automatic processing of reading the word starts taking over.

This suggests that controlled processing does require a lot more processing resources than automatic processing and thus attention cannot be sustained for long periods of time when both processes are in conflict. Eriksen and Murphy’s “zoom lens” analogy can be applied in this situation as the controlled processing being the zooming in on the specific task of naming the ink colour and the zooming out being when automatic processing starts trying to take over and slows the controlled responses down. Word count: 2001 words