Word Superiority Lab - Full lab report PDF

Preview

Summary

Full lab report...

Description

Running Head: THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

The Word Superiority Effect: Speed and Accuracy of Recognition Differences Between Non-Words and Words

1

THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

2

Abstract This experiment was conducted as a test of the word superiority effect and differences between accuracy and time of identifying words vs non-words. This experiment also sought to make a connection between accuracy and time of response for identifying words, and the participants tendency to have a maximizer or satisficer decision-making style. Study participants were 9 female students at Fordham University, with a mean age of 20.67, and mean of 2.78 years of college completed. Participants completed 48 trials on the website CogLab, where they had to determine as quickly as possible whether the two stimuli- either words or non-words- were the same or differed by 1 letter. Paired-samples t-tests indicated that participants accurately identified more words as being the same than non-words, but did not find any statistically significant difference in the amount of time it took participants to respond to words vs. nonwords. Independent samples t-tests indicated that a participants decision-making style had no significant effect on their percentage of correct words reported, nor the amount of time it took them to respond to words. Study results highlight the importance of the word superiority effect on accuracy of reporting words, but not the other variables considered.

THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

3

The Word Superiority Effect: Speed and Accuracy of Recognition Differences Between Non-Words and Words It is important to understand how people recognize letters. Some research has provided evidence that bottom-up processing- which involves the recognition of an objects features in order to recognize the whole object- is involved in the recognition of letters (Barry, 2018). Research has also shown that the feature detection aspect of bottom-up processing is used to recognize letters (Barry, 2018). For example, a study conducted by Kinney et al. (1966) showed that people are most likely to confuse the letters and numbers, “C and G,” “B and 8,” “O and G,” “G and 6,” and “C and 0” (Barry, 2018). This is because all of these letters and numbers share very similar features. But, it is important to further study the mechanisms behind letter perception, because the word superiority effect provides entirely different evidence for the way in which people perceive letters. The word superiority effect (WSE) states that it is easier to detect a letter that is presented as part of a word than when it is presented alone or in a meaningless string of letters (Barry, 2018). Research conducted by Reicher (1969) involved presenting participants with 3 types of stimuli, a single letter (K), a 4 letter word (WORK), and a 4 letter non-word (OWRK) (Barry, 2018). The results of this study found that letter recognition was better in the word condition than in the single letter or non-word condition, implying that the word superiority theory has nothing to do with memory or guessing (Barry, 2018). Further study of the word superiority effect is important because it has real world implications for people with dyslexia. People with dyslexia might be exhibit less of the word superiority effect because they are more likely to display letter-by-letter reading (Barry, 2018). It is important to understand how the word superiority effect is demonstrate in individuals with dyslexia, as it can give insight into how the word superiority effect works in non-dyslexic

individuals. One study on word superiority in dyslexic individuals that used the ReicherWheeler Paradigm that was previously mentioned found that dyslexic children actually THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

4

performed at levels similar to children of the same chronological age and children of the same reading-level age (Grainger, Bouttevin, Truc, Bastien, & Ziegler, 2003). This information is important as it shows that an absence of the word superiority effect in these children could not be attributed to poor lexical knowledge. Another interesting study sought to provide evidence that top-down processing of letters was not restricted to words but to meaning as well. This study conducted by Woollams, Evans, and Lambon Ralph (2017) found that words showed an advantage of non-words and consonant strings in terms of the amount of time the participant had to be exposed to the word, and the amount of time it took for them to react to it. This evidence is important to the experiment that will be conducted, as reaction time will be compared between words and non-words. This study also revealed that priming participants with a related word before exposing them to the word they have to identify, caused them to take longer in their decision-making process (Woollams, Evans, & Lambon Ralph, 2017). These results are interesting as they demonstrate that meaning has an effect on word recognition. The rationale behind conducting this experiment is to further support the idea that the word superiority effect involves the use of top-down processing for letter detection. Because of top-down processing, accuracy and speed of perception is affected by context and expectations (Barry, 2018). One of the overarching questions that thing study will try to answer is whether there is evidence of the word superiority effect based on accuracy of identifying words vs. nonwords and response time when identifying words vs. nonwords (Barry, 2018). As a supplementary analysis it will also seek to determine whether the percentage of correct responses and response time for identifying words is associated with people’s tendencies to be either

maximizers or satisficers. The hypothesis of this study states that participants will display greater accuracy and shorter response time when identifying words rather than non-words. It also states that participants with a maximizer decision making style will have a greater percentage of correct THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

5

responses when identifying words, while participants with a satisficer decision making style will identify words in a shorter amount of time.

Method Participants All participants were recruited for this study by their lab instructor in a Cognitive Psychology lab class. There were 9 participants, 100% of which were female. All participants were between the ages 20-21 (M=20.67, SD=0.50). All participants were enrolled at Fordham University. Materials CogLab. Participants used the website “CogLab: The Online Congition Lab” to complete this experiment. Students used CogLab to complete a test of the word superiority effect. The test involved the presentation of either two words or two non-words. The words or non-words were either the same or differed by one letter. Participants had to indicate as quickly as they could, whether the stimuli were the same or different. The stimuli were presented for an indefinite amount of time, until the participants made a decision. The participants indicated that the stimuli were the same by pressing the “m” key on the computer keyboard, and that they were different by pressing the “z” key. This test consisted of 48 trials. 24 trials were words and 24 trials were nonwords. 24 trials consisted of words or non-words that were the same, and 24 trials consisted of words or non-words that differed by one letter. The independent, categorical variables were the type of stimuli, either words or non-words. The continuous dependent variables were the

accuracy with which the participants indicated that the stimuli were the same or different, and the amount of time (in m/s) that it took to respond. MaxStat. This measure was used to determine whether participants used a maximizer or satisficer decision making style (Barry, 2018). A maximizer is defined as someone who THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

6

consistently tries to find the best option, while a satisfier is someone who is content with the option that is good enough (Barry, 2018). This measure consists of a response range from 1-7. 1 indicates “completely disagree,” and 7 indicates “completely agree” (Barry, 2018). If a participants average rating is greater than 4 they are considered a maximizer, and if it is below 4 they are considered a satisficer (Barry, 2018). Procedure The participants completed 48 trials of this experiment in which they were presented with 2 stimuli and had to determine whether they were the same or different. All of the data was then entered into SPSS. A paired-samples t-test was conducted to compare the average percent correct for word vs. non-word trials. A paired-samples t-test was conducted to compare the average response time for word vs. non-word trials. An independent-samples t-test was used to compare the average proportion correct in words trials for maximizers compared to satisficers. An independent-samples t-test was used to compare the average response time in word trials for maximizers compared to satisficers.

Results Three participants were missing word superiority data. There was a statistically significant difference between the average percentage of correct words and the average percentage of correct non-words, (t(5) = 5.00, p < .01). This means that the average percentage of correct words reported (M = 0.94, SD = 0.04) was greater than the average percentage of

correct non-words reported (M = 0.87, SD = 0.03). There was not a statistically significant difference between the average response time for words and the average response time for nonwords, (t(5) = -1.25, p > .05). This means that the average response time for non-words (M = 1558.51, SD = 1136.86) is not greater than the average response time for words (M = 1076.06, SD = 338.21). There was no significant difference between the percentage of correct words THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

7

reported by maximizers and satisficers, (t(4) = -4.80, p > .05). This means that the average percentage of correct words reported by maximizers (M = 0.96), was not significantly greater than the percentage of correct words reported by satisficers, (M = .93, SD = .05). Standard deviation information was missing for percentage of correct words reported by maximizers because only one participant was classified as a maximizer. There was no significant difference between the response time for words for mazimizers and satisficers, (t(4) = .50, p > .05). This means that the response time for words was not significantly less for satisficers (M = 1109.34, SD = 366.98) than for maximizers (M = 909.65). Again, there was no standard deviation information for response time of maximizers because only one participant was classified as a maximizer.

Discussion The part of the hypothesis that stated that participants would display greater accuracy when identifying words rather than non-words, was supported. These findings support the prevalence of the word superiority effect. These results are consistent with the findings of Reicher (1969). It is possible that the findings of these two studies were similar because they used similar methods. The part of the hypothesis that stated that participants would take less time to identify words than non-words was not supported. These findings were not consistent with those of

Woollams, Evans, & Lambon Ralph (2017). There are a few possible reasons behind the discrepencies in these findings. It is possible that the sample size was not large enough in the current experiment to yield significant results, because the mean response time for words was about 500 milliseconds less than it was for nonwords. It is also possible that word superiority effect only contributes to the accuracy of word recognition and not the amount of time it takes to

THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

8

recognize letters and words. For example, if bottom-up processing plays a larger role in letter recognition in words, then the length of the whole word would affect the amount of time it takes for recognition to occur. The part of the hypothesis that stated that greater accuracy when identifying words is associated with a maximizer decision-making style was also not supported. This could mean that a person’s decision-making style does not actually have an effect on their ability to accurately identify words. The part of the hypothesis that stated that shorter response time when identifying words is associated with a satisificer decision-making style was also not supported. One of the biggest limitations of this study was study was its small sample size. Also, out of the 9 participants, only 6 completed the experiment. The fact that all participants were female means that the results findings should not be applied to males. There was also little variation in the ages of the participants. This means that the findings should not be generalized to older adults or young children. This study could have been stronger if more participants of different ages had completed the experiment, as it may have also yielded age-based differences in the word superiority effect. One strength of this study was that all words and nonwords were the same length. Otherwise, the length of the words could have had an effect on the time it took for participants to respond.

Future research should look further into the ways in which bottom-up processing versus top-down processing effects letter and word recognition. Future research should also be conducted on the differences in word superiority effect between dyslexic and non-dyslexic children, because the study conducted by Grainger et al. (2003) found there to be very little difference in the word superiority effect for dyslexic and non-dyslexic children.

THE WORD SUPERIORITY EFFECT SPEED AND ACCURACY

8

References Barry, M. (2018). Word superiority [Powerpoint slides]. Grainger, J., Bouttevin, S., Truc, C., Bastien, M., & Ziegler, J. (2003). Word superiority, pseudoword superiority, and learning to read: A comparison of dyslexic and normal readers. Brain and Language, 87(3), 432–440. Woollams, A. M., Evans, G. A., Lambon Ralph, M. A. (2017). Seeing the meaning: Top-down effects on letter identification. Frontiers in Psychology, 8....