Conditions of congruency and effect on response times
Vision is the most important sense modality and this is why visual attention is studied vastly. LaBerge (1983) cited in Eysenck, 2005 carried out a study involving five-letter words, whereby a probe was put in place after or instead of the word. This probe was put in any of the spatial positions between the five letters of the word and two conditions were implemented. One of the conditions involved the participant focusing on the whole word and the other condition was to focus on the middle letter of the word by categorising it. The results showed that when the probe was placed in the central attentional beam the response was faster than when it did not. This means that the attentional spotlight has a broad beam from the word task or a narrow beam from the letter task.
Our eyes are capable of obtaining large amounts of information in parallel, as when they are directed at the computer screen while typing they can see, for example a bird fly past the window. This means that people are able to be influenced by objects within the environment that they did not realise they had seen. It is thought that letters and words that are short can be seen immediately, in a short period of time. This was examined by Sperling (1960) cited in Braisby, 2005 by presenting the participants four letters in three rows and were showed for a period of 50ms. The participants were able to report three or four of the letters and the others were unrecorded in that time.
Evett and Humphreys (1981) cited in Braisby, 2005 carried out a study involving two words, which both of them were masked meaning that it makes the target word unnoticeable. Results showed that when the second word was related (congruent) to the first word, it was reported more accurately.
The Stroop effect (Stroop, 1935 cited in Braisby, 2005), consisted of participants being able to read words faster and automatically than naming colours. For instance, if a word is printed in red ink but the word actually says "green" the participant will more than likely say the colour green more readily than the colour in which it is displayed. This is incorrect; the participant should read out the colour that the word is actually written in. This cognitive mechanism is called directed attention, whereby the participant needs to stop one response to carry out another response. This may occur because of selective attention, where naming colours requires more attention than reading words.
The flanker effect shows the differences in response times for congruent and incongruent conditions on a choice reaction task. Many recent studies using the flanker effect, used letters in the experiments to test response time and one of these was carried out by B.A. Eriksen and C. W. Eriksen, 1974 cited in Eriksen, 1995, . Eriksen used the letters H and K for one lever and S and C for the other. Three displays appeared on the screen where one of them was compatible, the second incompatible and the third was neutral. The first display, compatible consisted of the target letter being flanked by letters that were the same as the target. The second display, incompatible had flankers that were opposite to the target and the neutral display had two types, one being of similarity to the H and K letters and the other being similarity to the S and C letters. It was concluded that response time was high when the target was flanked by letters that were different to the target letters and low when the flankers were similar to the target letters. However, in the neutral condition the neutral flankers shared feature similarity to the target letters there was no effect. This was compared to when the flankers showed feature similarity to the target letters in the other response, the response time was similarly increased for the opposite response letters, but not as much. This is to examine what irrelevant information is taken in, in a visual task.
Schaffer and LaBerge (1979) cited in Braisby, 2005 discovered priming effects within their study. The study involved words instead of single letters appearing on the computer screen and the participant had to decide what category the word belonged to, which measured the participant's response time. The participant had to press a key on the keyboard to say which category it belonged to, for example, food and another key to associate it with the other category; places. The target word was surrounded by two flanker words, one below it and one above it making it in the centre. Within the study two conditions were implemented where the flankers were either congruent; same category as the target word or incongruent; opposite category to the target word. Results showed that when the flanker words were congruent with the target word the participant's response time was quicker than when the words were incongruent because they were related to each other. This is very similar to the Stroop effect, where the impact of the colour of the words and what the word actually displayed slowed down the response time.
The aim of the study was to see whether congruency or size of the target word has an effect on the response times for participants.
Based on past research there are two hypotheses, one is that if the flanker words are congruent to the target word, then there will be a faster response time recorded by the participant and if they were incongruent to the target word, a slower response time will be recorded. The second hypothesis is that when the target word is bigger in font size than the flanker words, a faster response time will be recorded by the participant, than when the target word is smaller.
The type of design method that took place within the experiment was a within subject design method, which consisted of two Independent variables which were manipulated and one Dependent variable that was measured. The first Independent variable was congruency; whether the target and flanker words were congruent or incongruent. The second Independent variable was size; whether the target word stayed the same size or was enlarged. The Dependent variable that was measured was response time.
Twenty participants took part in this study, who were Psychology Undergraduates from the University of Central Lancashire. The undergraduates were selected by using the opportunity sampling technique, meaning the participants who were taken were there and ready at the time of the study. There was note of the participant's gender and age split at the time of the study.
A computer was used within the experiment where the screen measured 100cm and the participant sat 50cm away from it. Coloured stickers were used to put over the letters 'z' and 'm', where the colour blue was used for the letter 'z' and the colour red was used for the letter 'm'. For the participant to remember which colour represented which category name card, were slotted into the gaps of the keyboard above the coloured stickers. The name card that showed animal was placed above the red sticker and the food name card was placed above the blue sticker. The size of the monitor was 15 inch's and was an HP L1906. The list of words for both the animal category and the food category were decided and ten words were chosen for each (refer to appendix 1).
Participants were first asked to enter a room and sit in front of a computer, approximately 50cm away. They were then read a set of instructions saying that they were to be involved in a computer based task, involving deciding which category the target word belongs to. The experimenter told the participant that they were allowed to withdraw from the study at any time and if they would consent to taking part in the study. The participants who did consent were asked to place their right index finger on to the red key on the keyboard which was the animal category. Their left index finger was to be placed on to the blue key on the keyboard, which was the food category. The participants were also reminded of which colour represented which category via the name cards; one showed animal for the red key and the other food for the blue key, which were placed above the coloured keys. Each participant was then asked to read the instructions, which were displayed on the screen that said what colour represented what category and were asked to press space bar when ready. On the computer screen a blank screen was displayed for one second, followed shortly by another screen containing a fixation cross, which showed where the middle word was going to be shown (target), for one second. A target display was then shown containing three words directly below each other. The middle word was the target word and the other two either side it were the flanker words, which lasted for a maximum of three seconds or until the participant responded to the word. The participant was then meant to press either the red or blue cotton to show whether the middle word was showing a word in the animal or food category. The study contained eight practice trials, which lead straight into the eighty main study trials without warning, which all the participants took part in. Throughout the study there were two factors involved. One was the size of the target, which was increased to size 28, or stayed the same at size 18. The other factor was whether it was congruent or incongruent, where the flanker words were either in the same category as the target word, which made it congruent or different to the target, which made it incongruent. Once the study was completed, a screen came up displaying thank you and they were then debriefed.
The raw data consists of the reaction times recorded for each participant from the experiment (refer to appendix 2). The errors and outliers above or below 2 standard deviations, from the experiment were removed. This meant that anything below 2 standard deviations <124 were removed as these were deemed too quick and anything above 2 standard deviations >1328 were removed, as these were deemed too slow. The test that was performed was a 2 x 2 within subject (repeated measures) one-way ANOVA (refer to appendix 3).
Table 1: Mean response times for the congruency of the target and flanker words and size of the target word
Target Display. Response Time: Mean by Congruency and Size
Mean Target Display. Response Time
The table above shows the mean response times for the size of the target word and whether the words were congruent or incongruent to each other. When the target word is the same size as the flanker words and is congruent the mean response time from the participants was slower than when the words were the same size and incongruent. When the target word was larger than the flanker words and congruent the mean response times from the participants were quicker than when the word sizes were different but incongruent. The table shows that when the target word is larger than the flanker words and the words are incongruent, the mean response times are slower from the participants than when the target word is the same size as the flanker words. When the target word is the same size as the flanker words and congruent with the flanker words, the mean response times are slower than when the target word is larger than the flanker words but are congruent.
When referring to the repeated measures one-way ANOVA calculations output, the results show that there is an insignificant result for the size of the target (p=0.888). This shows that there is no significant relationship between the size of the target word and the participant's response times. The relationship between whether the words are congruent or incongruent and the participants reaction time is also insignificant (p=0.465), which shows that whether the words are congruent or incongruent between the flankers and target it has no effect on the participants response time. The final relationship between the interaction of the size of the word and congruency, has no significant effect with the participants response time (p=0.265).
The present study has suggested that when referring to the statistical analysis results, there was no effect when the size of the target word is different (larger) to the flanker words or the same as the flanker words, for the participant's response time. There was also no effect when the target and flanker words were congruent or incongruent with each other, for the participant's response time. The final result showed that there was no effect for the interaction between size and congruency of the target and flanker words, for the participant's response time.
By referring to the statistical analysis results, both hypotheses are incorrect as the statistical analysis results do not support the hypotheses. This is compared to when referring to the means table 1, the mean response times do show some support for the hypotheses as when the target word is of a different size (larger) to the flanker words and is congruent the mean response times are faster than when they are incongruent. More support for the hypotheses is shown via the means table 1, as the mean response times for when the words are congruent, but the target word is larger, is faster compared to when the target word is the same size and also congruent.
One limitation within this study was that no thought was taken into consideration about participants may be having disabilities in reading computer screens, such as impaired vision or being dyslexic, so were not able to read the words properly. This could have been taken into consideration on the computer screen before the experiment took place, when inputting age and gender. This could come under the heading of visual disabilities. Another limitation could be that a keyboard was used within the experiment and it takes time for participants to look for the colours on the keyboard, as there are other keys around. An improvement could be to use single, larger buttons by themselves so the participants will not have to take time over finding the coloured buttons. Also instead of pressing a button, the participant could squeeze something to respond to the words, which would be faster than pressing a button as it is in the participants hand already.
For future research the conditions could be changed within the study to improve it, this could then improve the results of the response times from the participant. One way of changing this could be to change the colour of the target word, then the target word would be a different colour to the flanker words or the target word could be the same as the flanker words and see what the results would be. Another condition that could be changed for future research is the timing in which the target or flanker word is displayed on the screen. For instance, the flanker word could appear 1 second after the target word to see if any effect would occur or vice versa to observe response times from the participants. A study has been conducted to show this by Broadbent and Gathercole (1990), which was adapted from the original experiment of Schaffer and LaBerge (1979). They carried out the original experiment by using the flanker effect and adapted it by delaying the flanker words by 40ms and presenting the target word first. This was to enable a priming effect to occur even more than the original experiment. The results of this study showed that the flankers had a great effect when the category of the flankers and targets were the same. A third condition could be to see whether the length of the word could affect the results of the response time from the participants, for instance if the flanker word was long and the target word was short, would this have an affect on the time to respond.
A past study has been conducted to show whether colour has an effect on the response times or not. This study was carried out by Cohen and Shoup, 1997- Journal of Experimental Psychology: Human Perception and Performance, Vol 31(6), Dec 2005. pp. 1493-1509, Selective Attention and the Perception of an Attended Nontarget Object, Chen, Zhe, where the target was in a colour and the flanker words were either congruent or incongruent in colour to the target word. The results showed that the flankers influenced the performance, but only when the flanker words and target word was defined by the same colour and when they were on different dimensions the effect was minor.
In conclusion, the results showed that in the means response time's table there was support for the hypotheses as there was an effect on the response times when the conditions of congruency and size of the target and flanker words were put in place. However, when the statistical analysis results were calculated these did not support the hypotheses, as all the results did not have an effect on response time. This therefore means further analysis is needed to make a distinction between whether congruency or size has an effect on response size or not.
If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal: