Prototype Experiment - Research Paper Example

Prototype Research al Affiliation) The use of prototypes has is common to arrive at several judgments. This is known to make work easier since all that one has to do is to relate the various prototypes to the available data before arriving at a definite conclusion (Hampton, 1995). Several authors have written extensively about this topic as indicated in the literature review each presenting a different point of view in regards to this matter. In this paper, I carried out an experiment to try and find out the validity of this statement. This involved the use of 22 participants. They had to undergo a trial phase followed by the testing phase. The experiment was carried out online and it involved the use of prototypes and the non-prototypes. For this instance, the prototypes were represented by dots that were in two categories A and B while the non-prototypes, also known as variants were the dots that did not fall in either of these categories. The participants had to go through the training phase, followed by the testing phase before moving on to the trial phase. The role of the participants in this experiment was to categorize the various dot patterns as either A or B and at the end of the process they had to save their work in the systems. The data was evaluated using various models and represented in the form of graphs to illustrate the connection between these two factors. The conclusion was able to prove the point that indeed the use of prototypes was very crucial in arriving at certain conclusions. Introduction The use of prototypes is common in our everyday judgments. A prototype refers to an existing model or formula whose role is used to test and evaluate an existing design or new concepts. This is based on the fact that research has already been done in regards to these areas and therefore what is needed is just to fit the new data into existing models. Prototypes are currently used in a variety of fields such as medicine and in manufacturing fields and they have become critical in our day to day activities. This field is of crucial importance to scientist, physicists and even doctors. However, the degree of this importance is yet to determine and we have to consider whether it is possible to survive without these elements. I have therefore established a hypothesis in regards to the matter that states, is the use of prototypes key in decision making? This paper therefore considers the various publications on the matter and tries to justify them by carrying out an experiment in regard to the matter while trying to prove the validity of the hypothesis. Literature review There are several studies by different authors in regards to the prototype theory of concepts. Each of these authors had a different opinion to write in regards to the topic and these are outlined in the literature review below. According to James Hampton there were 4 experiments specially designed to test the two predictions of the prototype theory. The initial prediction claimed that there are certain necessary elements of a concept that can only be matched on several instances; in this case, there are certain unnecessary features of the given concepts that can affect their categorization. The next hypothesis was in relation to the various independent features that were used to determine similarity. Previously, the existing versions of prototype theory assumed linear relationship of the feature matches; thereby the change in any given feature on the membership category is higher when the categorization probability is closer to 50%. Hampton found out that this was not the case but rather the effects of changing the various features was at a higher level when all the other features were positive, resulting in a maximum categorization probability. Evidence in regards to this topic was also in an article by Solomon et al that talked about the importance of concepts in relation to cognitive processes. They state that although the study of various concepts had been related to the study of categorization, this is only a single conceptual function out of several others. As a result, categorization cannot be the only means of studying concepts because they serve various functions that interact in various ways, therefore affecting conceptual processing and structure. Another reason is that by studying the various functions in isolation one can only identify with the various cognitive processes that are connected to it. It therefore makes it hard to discover other processes that may be common to the various multiple functions. It is the view of these authors that the study of concepts should be done in relation to other interrelated functions. The widespread use of concepts has been related to real life examples such as the paper by Smith in regards to the application of these concepts in the courtroom. In courtrooms, the jury is expected to come up with a fair ruling and they have to rely on the evidence presented in the courts plus the instructions that have been issued by the judge. It is assumed that none of them has ever practiced law before. Research in regards to these assumptions revealed that several jurors usually had preconceived ideas about crimes and this had an effect on the choices that they made. This was despite the fact that they were expressly given instructions not to let their ideas influence their decision making process. It was therefore found that for instructions to be effective they must be able to change the jurors’ pre conceived concepts about crime. The individual differences in prototypical judgments in regards to trait categories were discussed extensively by Hershey. This was by the use of three experiments that clearly outlined the individual differences in the judgments in regards to trait categories. In the 1st study, the subjects had to rate nine different categories and it was found that the degree of agreement among the individuals was much lower than the reliability of retesting for individual ratings. The 2nd and 3rd study involved the rating characters prototypicality in a given film. Once more, there was a lower level of agreement among the various individuals than the retest reliability. According to this study, it was proven that differences in various individuals in regards to perception of prototypicality is a major effect when it comes to comparing consistency of behavior in relation to different situations. Therefore trait ratings based on certain behavior was found to be unreliable. There is therefore need to reconsider existing theories such as the implicit personality theory. Cognitive representations of Semantic categories were studied by Rosch of the University of California, Berkeley. This was by the use of the priming technique to determine the type of cognitive representation that was due to superordinate semantic category names. This was by the use of 10 different experiments that resulted in four findings. First was that the internal structures of these superordinate semantic categories was a pervasive aspect in regards to the way these various categories were usually processed in regards to the specific tasks. Secondly, it was revealed that there exist different levels of perception processing. Thirdly, the underlying representation had an effect on the superordinate categories with pictures taking less time to prepare than words. Lastly, there was revealed a comparison in the representations of the various superordinate categories with several kinds of categories such as color and other categories at various levels of abstraction. This was able to provide methods used to collect data in regards to several issues that were not resolved due to the development and nature of abstraction. The effects of category prototypes on the accuracy of performance rating was studied by Kinicki Et Al. they used two studies to try and determine the influence of performance prototypes on performance appraisals of employees. It was shown that in case of rating of instructors, most students had already established prototypes that they used to rate their instructors. The results showed that the accessibility of these prototypes had minimal effects on the overall results while cognitive primes that were based on vignettes had a much stronger influence on the performance ratings than videotaped stimuli. Mervin and Rosch used prototypes to try and relate the resemblances among members of various families. This was done by the use of 6 different experiments. The hypothesis in their research was that the members of a similar category who are considered to be more prototypical are those with similar attributes to other members of the same category and very few common features with members of other categories. This implies that the prototypicality was a function of the various attributes of these given items. These experiments were able to show that the resemblance in family members can be used as an alternative to criteria features when it came to defining the different categories. Method Participants The experiment employed 22college students. Each participant judged different versions of the concept in the experiment. All the participants acted in this one experiment once. The participants in this experiment were largely drawn from college students. These were aged between 20 and 27. Mean age of participants was 22.52, and standard deviation was 2.04.15 were female, while the rest were male. Also, 18% of Caucasian, 23% of Asian, 45%, Hispanic, 14% of other race were in the experiment. Materials The concept to be used in the experiment was drawn up based on the method on previous studies on the topic. This was one of the earliest studies that were conducted to explore the representation of concepts in a controlled manner. All the participants were required to access the Cengage Cog Lab, which is an online computer lab. This is where they would find the instructions for the experiment, and participate in the experiment. Each participant used a desktop computer individually in the classroom.The design of the experiment was the same throughout the experiment. Each participant participated in the trial stage and the testing stage of the experiment. Most important is that there was the use of prototypes and non-prototypes in the experiment. These were in the form of dots, which the experiment used. The prototypes were those dots that would fall in either category A or B of dots. On the other hand, the non-prototypes were also called the variants, and these were the dots that neither belonged to category A or B. However, each of the non-prototypes was a specific variant for either category A or B. Procedure The participants were instructed to log into the Cengage Cog Lab in order to participate in the experiment. First was the training phase, where the subjects were trained on how to go about the instructions, and second was the testing phase. After logging in the online computer lab, the participants were asked to look out for a black rectangle. After this, each participant participated in the trial phase the experiment. A fixation point was displayed for one second before being replaced by different dots. A participant was expected to classify the dot pattern as A or B. At this stage, each dot was a variation of one or two fixed prototype random dot patterns. In order to make a variation, the subjects were asked to randomly take ten dots from 25 dots in a prototype and change their position. Each participant made 60 trials before proceeding to the testing phase. In the testing phase, the dot patterns were of four types. One dot was a prototype and corresponded to category A, while another was a prototype that corresponded to category B. Each of the remaining two types of dots was a variant of each prototype. Each dot pattern was presented four times in a random order.The role of the participants was one throughout the whole experiment. They were expected to use the shortest time possible to categorize the dot pattern as A or B. To continue to the next trial, the participants were to press the start button on their keyboard. At the end of the experiment, each participant was instructed on how to save their data to a set of global data. The results and conclusions of the experiment were to be made. Experiment Result Data Statistics Prototype Mean Variants Mean N Valid 21 21 Missing 0 0 Mean 909.3750 1038.0298 Std. Deviation 375.99484 437.34604 Skewness 1.245 .906 Std. Error of Skewness .501 .501 Kurtosis 1.366 1.009 Std. Error of Kurtosis .972 .972 Range 1458.88 1769.00 Prototype Mean Frequency Percent Valid Percent Cumulative Percent Valid 377.13 1 4.8 4.8 4.8 442.88 1 4.8 4.8 9.5 655.13 1 4.8 4.8 14.3 666.50 1 4.8 4.8 19.0 676.00 1 4.8 4.8 23.8 689.25 1 4.8 4.8 28.6 707.38 1 4.8 4.8 33.3 719.25 1 4.8 4.8 38.1 736.88 1 4.8 4.8 42.9 746.00 1 4.8 4.8 47.6 771.38 1 4.8 4.8 52.4 819.75 1 4.8 4.8 57.1 909.13 1 4.8 4.8 61.9 919.75 1 4.8 4.8 66.7 938.38 1 4.8 4.8 71.4 992.38 1 4.8 4.8 76.2 1136.25 1 4.8 4.8 81.0 1298.00 1 4.8 4.8 85.7 1305.25 1 4.8 4.8 90.5 1754.25 1 4.8 4.8 95.2 1836.00 1 4.8 4.8 100.0 Total 21 100.0 100.0 VariantsMean Frequency Percent Valid Percent Cumulative Percent Valid 419.50 1 4.8 4.8 4.8 421.88 1 4.8 4.8 9.5 587.00 1 4.8 4.8 14.3 594.38 1 4.8 4.8 19.0 735.75 1 4.8 4.8 23.8 788.50 1 4.8 4.8 28.6 822.00 1 4.8 4.8 33.3 844.50 1 4.8 4.8 38.1 917.50 1 4.8 4.8 42.9 960.88 1 4.8 4.8 47.6 963.88 1 4.8 4.8 52.4 979.50 1 4.8 4.8 57.1 1039.00 1 4.8 4.8 61.9 1075.63 1 4.8 4.8 66.7 1095.13 1 4.8 4.8 71.4 1214.63 1 4.8 4.8 76.2 1403.63 1 4.8 4.8 81.0 1514.88 1 4.8 4.8 85.7 1556.00 1 4.8 4.8 90.5 1676.00 1 4.8 4.8 95.2 2188.50 1 4.8 4.8 100.0 Total 21 100.0 100.0 Paired Samples Statistics Mean N Std. Deviation Std. Error Mean Pair 1 PrototypeMean 909.3750 21 375.99484 82.04880 VariantsMean 1038.0298 21 437.34604 95.43673 Paired Samples Correlations N Correlation Sig. Pair 1 PrototypeMean &VariantsMean 21 .945 .000 Paired Samples Test Paired Differences t df Sig. (2-tailed) Mean Std. Deviation Std. Error Mean 95% Confidence Interval of the Difference Lower Upper Pair 1 PrototypeMean – VariantsMean -128.65476 147.71509 32.23408 -195.89387 -61.41566 -3.991 20 .001 Results There was a clear difference in the results in both scenarios. The measure refers to the different types of variables that can be used to collect data. These can be nominal, ordinal, interval or by the use of ratios in this case, the measure is by the use of ratio data. One scenario involved the use of prototypes while the other involved the use of non-prototypes (Rosch &Mervis, 1975). The mean and the standard deviation of the variants were much higher than those of the prototype. The mean for the variant was 1038.03 while the mean for the prototypes was 909.37. on the other hand, the standard deviation for the variant was 437.34 while the standard deviation for the prototype was 375.995. The prototype had a higher level of Skewness and Kurtosis than the variants. The level of skewness for the prototype was 1.245 while that of the variant was 0.906. The level of Kurtosis for the prototype was 1.366 while that of the variants was 1.009. Discussion and Conclusion The results of this study show the difference between the use of prototypes and non-prototypes in regards to the decision making process. The mean and the standard deviation of the variants were higher in the case where the variants were used than in the case where the prototypes were used. This implies that when the prototypes were used it was easier to generate results that were slightly similar unlike the case when the variants were used. This can be because by using the prototypes there is already a course of direction and certain guidelines within which the study is conducted (Beck &Seagal, 1988). As a result, ant variance from this course of action is noted early and the changes are made to ensure that the data that is captured falls within the required scope. In the case for the variants, the researcher has to begin his study from the very beginning and there is lack of a supporting structure to guide their decision making. Furthermore, this is indicated by the use of the standard error mean which clearly indicates that the error when using the variants was much higher than when using the prototypes. The use of prototypes is therefore crucial if the researcher of any given topic wants to produce accurate results. This means that in any field of research the use of prototypes if we are to produce reliable information that is required in the decision making process. In the field of research, developed prototypes act as a guideline and they should be used effectively by researchers. This conclusion is important in regards to future research. It is not only important in highlighting the importance of prototypes in regards to the decision making process but also due to the fact that present research can be used to build on past prototypes. When existing prototypes are proven to work seamlessly they become more validated and there is an inherent increase in their use in various fields of research (Solomon &Medin, 1999). There is a constant change in various factors that affect research and this includes features such as technology and this may have an impact on future research particularly the use of prototypes. It therefore means that for research to be valid in future it has to incorporate these changing features such as technology. Given that the use of prototype has already been proven to have an impact on the outcome of any given research, there should be a further analysis on the extent to which the prototype has an influence on the outcome of any given piece of research. This is important so as to establish the relevance of new data in regards to any subject matter and to reduce over reliance on existing information in regards to thee particular subject (Kinicki, Wade&Trost, 1995). In the course of this research there were various problems that were encountered in the process and these had an effect on the outcome of the research. First of all, the research was carried out by only using 22 respondents at a particular location. This figure is too small to be used to come up with results that represent such a wise field. As such the research should be carried over a wide area with many other respondents (Smith, 1991). When the scope of the research is increased the findings that will be presented will not be biased to a specific area. The results will be a summary of the outcomes of a wide range of data and therefore the final results will have more credibility.Another limitation to this research was that some people may already have preconceived notions in regards to the use of prototypes since the might have used it in their daily course work (Rosch, 1975). In academics, the use of existing methods and concepts is common and therefore this background knowledge might have an influence on the output of these respondents. As such, the data collected might not be a direct representation of what is collected by the respondents. Finally, the use of only one method to collect data may have an influence on the accuracy of the results. There should be a combination of qualitative and quantitative means in data collection so that the data collected is much more accurate. The findings in this paper indicate the importance of the use of prototypes in the field of research. It further indicates the difference in the outcomes when they are not used and this shows the key role that they play in guiding any type of research. As a result, the use of prototypes should not be an option in regards to future research but should be made compulsory in any given field so that the result outcomes are within an acceptable range. In this way, the information can be reliably used in the decision making process by interested parties. References Beck, L., McCauley, C., & Segal, M. (1988). Individual Differences in Prototypicality Judgment About Trait Categoriess. Journal of Personality and Social Psychology (pp. 286-292). Pennsylvania: American Psychological Association. Kinicki, A., Hom, P., Trost, M., & Wade, K. (1995). Effects of Category Prototypes on Performance-Rating Accuracy. Journal of Applied Psychology (pp. 354-370). Arizona: American Psychological Association. Rosch, E., &Mervis, C. (1975). Family Resemblance. COGNITIVE PSYCHOLOGY (7 ed., pp. 573-605). California: Academic Press. Hampton, J. (1995). Testing of Prototype Theory of concepts.Journal of Memory and Language, 34, 686-708. Rosch, E. (1975). Cognitive Representations of Semantic Categories.Journal of Experimental Psychology, 104(3), 192-233. Solomon, K., Medin, D., & Lynch, E. (1999).Concepts and categorization.Trends in Cognitive Sciences (pp. 99-106). Pennsylvania: Academic Press. Smith, V. (1991).Attitudes and Social Cognition.Journal of Personality and Social Psychology (pp. 857-872). Pennsylvania: American Psychological Association. Read More