The Effect Of Semantic Category Psychology Essay

The Effect Of Semantic Category Psychology Essay Tuving, suggested that in order to attain good memory banks, one way of retrieving information is to organize words that have a relationship with one another. For example, clustering words together in one category could help achieve better recall, rather than words that are randomly mixed. Clustering could help re-organize memory, therefore, attaining information more effectively. The aim of this study was to replicate Cofer, Bruce, and Reichers (1966) experiment of presenting semantic category instances in blocked lists. This method was used and it was predicted that memory for words will be superior, when words are visually blocked by semantic category, as opposed to being blocked randomly. This hypothesis was fully supported. Thus, the null hypothesis, there will be no difference when words are blocked by semantic category or blocked randomly across categories, was rejected. An unrelated design was used and opportunist sample of 40 1st year psychology undergraduates at City Univer sity took part in the experiment. The independent variable, blocking by semantic category had two levels. Half of the participants were assigned to the first level of the independent variable, blocked by semantic category. They were presented with six blocks of related category list words, which consisted of six related instances in one category. The second half of the participants was assigned to the second level of the independent variable, blocked randomly. The participants were presented with six blocks of listed words of randomly mixed instances, of several categories. The dependent variable was memory for words that were visually presented. The memory for words was measured under each condition, by the mean score of the written free recall, out of 36 words correctly remembered, after a 90 second period. The results indicated that people in the blocked by semantic category condition recalled more remembered words, with a mean score of 24, than those in the blocked randomly cond ition. The blocked randomly condition mean score were much lower; their mean score was 19. It was therefore concluded that this current research was very consistent with Cofers et al (1966) findings that clustering related words together in one category could help re- organize memory, so information could be attained more efficiently. Introduction: word count: 824 According to Kalat (2009) memory is organized by a complex network of interrelated neurons within the brain; that can hold millions of pieces of independent data. It is this ability of our mind to store detailed, organized memories of past experiences that makes us capable of learning. These experiences stored in the form of memories help us learn from mistakes, protect us from danger, and achieve the goal that we set, by harnessing the power of our memory. We are better able to learn life lessons that help us avoid mistakes in the future, based on our own past, and the malfunction of others. Tulving (1972) recognized that individuals had two types of long-term memory, episodic and semantic memory. According to Tulving (1972) episodic memory is autobiographic, which is a memory system that consists of episodes of individuals past experiences, of particular objects. Normally these objects are stored in our long term memories and can be retrieved, at any given time. For example, remembering loved ones, or a representation of a meaningful event, or even what one might eat for breakfast, are all forms of episodic memories. Semantic memory however, is referred to the memory of meanings, and understandings. Eysenck and Keane (2000) suggested one way of thinking about memory organization is coding. Semantic memory could be held for a few minutes or several years. This model suggests that some information is stored and coded visually. Yet to gain access to our memory banks and retrieve relevant pieces of information, is trying to gain access to our memory banks of the learned inf ormation. One way of achieving this, is with organization and practice. Craik and Tulving (1972) described this as deep semantic processing in their levels of processing approach. They suggested that when a particular stimulus is processed, we need to concentrate on word meanings to help aid our memory banks as opposed to paying too much attention to visual and phonemic features. When people use this approach for a particular task or even revision for exams in real life, empirical evidence has shown that deep semantic processing could even be achieved when one is not trying to remember, as demonstrated by Craik and Tulving (1975), incidental task. In order to attain good memory banks, one way of retrieving information is to organize words that have a relationship with one another. For example clustering words together in one category could help achieve better recall, rather than words that are randomly mixed. Clustering could help re-organize memory, therefore, attaining information more effectively. Gross (2005) suggested that while poor memory can sometimes be the result of a mental handicap or disability, it most often has to do with a lack of attention or inability to concentrate, poor listening skills, and other types of bad habits. Fortunately, you can re-train yourself with proper habits to develop and fine-tune your memory. The basic tool for developing better memory is the clustering technique. There has been many experiments that have demonstrated this particular organization for attaining memory. One study was Bousfeilds (1953) experiment; words that were placed into categories were more accurately remembered than a list of unrelated words. Another particular study conducted by Cofer, Bruce, and Reicher (1966). They identified that presenting semantic category instances in blocks actually increased recall. The remembered words of instances of one category, was much higher than the recall of the blocks of instances of the randomly mixed categories. Subsequently, this current study replicates Cofer, et als (1966) clustering technique experiment. The independent variable was the presentation of blocking by semantic category. There were two levels of the independent variable, blocked by semantic category, and blocked randomly. The manipulation of the independent variables was the way the participants corresponded to the presentation of 36 words, grouped into six blocks of six words. Half of the participants were assigned to the first level of the independent variable, blocked by category. The participants were presented with six blocks of listed words, consisting of six instances of one category. The second half of the participants was assigned to the second level of the independent variable, blocked randomly. The participants were presented with six blocks of listed words of randomly mixed instances, of several categories. The dependent variable was memory for words that were visually presented. The memory for words was measured under each condition, by the mean score, of the written free recall, out of 36 words correctly remembered, after a 90 second period. If the reasoning behind Cofer, et al s (1966) experiment was correct then we can then expect that memory for words will be superior, when words are visually blocked by category, as opposed to when they are blocked randomly. This is the experimental hypothesis and is directional, as it predicts the results. However, if Cofer, et als (1966) experiment cannot be explained, then there will be no difference in recall when words are blocked by semantic category, or blocked randomly across categories. Design: word count 215 The independent variable was the presentation of blocking by semantic category. The first level of the IV was, blocked by category, and the second level, blocked randomly. The manipulation of the independent variables was the way the participants corresponded to the presentation of 36 words grouped into six blocks of six words. An unrelated sample design was used. The advantage of this design was that both samples were seeing the same materials (words). However, the disadvantage of the design was individual differences. To control for this, random assignment to conditions to one of the two levels of the independent, (refer to appendix for allocation). Half the participants were assigned to the first level of the IV, blocked by category. They were presented with six blocks of listed words, consisting of six instances of one category. The second half of the participants was assigned to the second level of the IV, blocked randomly. They were presented with six blocks of listed words of randomly mixed instances, of several categories. The dependent variable was memory for words that were visually presented. The memory for words was measured under each condition, by the mean score of the written free recall, out of 36 words correctly remembered after a 90 second. Participants immediately recalled the words after the presentation. Participants: word count 72 In total there were 40 participants. 34 females and 6 male participants were 1st year psychology undergraduates at City University. Participants ranged in age from 18-40, with a mean age of 21-22. Each participant was currently taking part in a laboratory course in the social sciences building. Therefore the sample used was an opportunist one. However, due to the sample design, random allocation to conditions was very relevant due to individual differences. Apparatus and Materials: word count 215 Each participant was presented with a plan sheet of A4 sized paper. The experimenter, Dr. Stuart Menzies, randomly allocated half the participants material that consisted of either 36 nouns that were presented visually together into six blocks of listed words on one page. Each block consisted of six instances of one category. This condition was the first level of the independent variable and was named the blocked by category group. The other half of participants were randomly given, 36 nouns that were also visually presented together into six blocks of listed words on one page. This time however, each block consisted of randomly mixed instances of several categories. This condition was the second level of the independent variable, and was referred to as the blocked randomly group. (Please refer to the apparatus and material section for the blocked lists categories, in the appendix). The response mode was the written free recall of the remembered words after a 90 second period. The ti me was monitored by a stop watch. Participants immediately recalled the words in any order after the presentation (refer to appendix, instructions and words lists). The results were taken away and recorded by the experimenter, who noted them down on a data sheet. (See results section in the appendix for the raw data sheet,). Procedure: word count 252 The experimenter handed out material in a random fashion to participants. A list of 36 nouns were visually presented in six blocks, consisting of either six instances of one category on one sheet of sheet of paper, or randomly mixed instances of several categories on one sheet of paper. Blocked by semantic category consisted of six instances of one category. This condition became the first level of the IV. Randomly blocked categories consisted of randomly mixed instances of one category. This condition became the second level of the IV. The experimenter verbally translated the instructions of the task ahead, (please refer to apparatus and material section for instructions in the appendix). The dependent variable, memory for words was measured under each condition, by the mean score of the written free recall, out of 36 words correctly remembered. Each participant had 90 seconds to remember as many words from the six blocked lists. A stop watch was used to count down the time, after t he 90 second period expired, the experimenter then informed the participants to stop! Both word lists from the two conditions were taken away. Immediately after the presentation the participants were then asked to note down all of their remembered words in any order that they preferred. The experimenter had given a further 90 seconds to recall the remembered words. The written free recall scores was recorded, and measured by experimenter. The mean score of the two calculated differences were then obtained, (please see the mean scores in the appendix). Results: word count 313 The dependent variable was memory for words that were visually presented. The memory for words was measured under each condition, by the mean score of each condition of the written free recall, out of 36 words correctly remembered, after a 90 second period. Table 1 shows the difference of the mean scores of the two conditions, blocked by category and blocked randomly. The results indicated that more people in the blocked by category had remembered more words in the six blocks of listed words, consisting of six instances of one category, than the randomly blocked category lists condition.. (Refer to the results section in the appendix for the raw data, graph of the mean scores, and the full calculations of the Mann -Whitney U test). Memory for words visually presented. Measured by the written free-recall-score X/36 words Table. 1 Blocked by category condition Blocked Randomly condition Mean Spread Range 24 15-33 18 19 13-27 14 Mean score from the two calculated differences Since the samples were unrelated and can be ranked on an ordinal scale, a Mann Whitney U test was employed. A 1 tailed test was employed because the experimental hypothesises was directional. Using the statistical formula and tables, obtained from QM PS1007. Runyon and Haber (1976) The Mann -Whitney U test indicated N1=20, N2 =20 participants, Uobs = 0, p = 0.05 (1 tailed test). Ucrit = 138 =p= .005. Since the Uobs Discussion: word count 688 The aim of this study was to replicate Cofer et als (1966) experiment of blocked presentation, of presenting category instances in blocked lists. This method was used and the prediction that memory for words will be superior, when words are visually blocked by semantic category, as opposed to whose words are blocked randomly, was fully supported. Thus, the null hypothesis, there will be no difference when words are blocked by category or blocked randomly across categories, was rejected. The results indicated more people in the blocked by category condition recalled more remembered words, with a mean score of 24 than those in the blocked randomly condition. The blocked randomly condition had much lower recall scores; the mean score was 19. Eysenck and Keane (2000) suggested that gaining access to our memory banks is best applied through organization. Tulving, (1972) recognized that there are two memory system for long term memory. One is semantic memory and the other is episodic. Semantic memory, as mentioned in the introduction, is referred to the meanings and understandings of general objects based in the world we live. Semantic memory could be held for a few minutes or several years. This model suggests that some information is stored and coded visually. Yet to gain access to our memory banks and retrieve relevant pieces of information, is actually gaining access to our memory banks, of the learned information. One way of achieving this, is with organization and practice. Summarizing the previously learned literature has helped to illustrate how memory could be obtained and re- trained by applying clustering techniques for memory organization. This could be just some of the possibilities of why the blocked by category condition had produced a much higher recall of remembered words, than the randomly blocked condition. It would seem from the results of Bosfeilds (1953): Cofer, et als (1966) and this current study could be achieved by organization. Semantic memory does liked to be organized, and if this organization is not in sight, then the less likely it is, for us to attain good memory banks for recall. Consequently, Cofer, et als (1966) experiment identified that presenting category instances in blocks, actually increased recall, and the remembered words of instances of one category was much higher than the recall of the blocks of instances of the randomly mixed categories. Therefore, this current study was very consistent with Cofer, et als (1966) findings. The data suggests that there was an overall majority of recalls made in the blocked by category condition. Thus, there were some methodology limitations in this current study. Individual differences was a concern however, randomization to conditions was effective. However, larger randomized controlled trails could provide more definitive evidence, the bigger the population, the better the results would indicate, (Miller, 1984). A number of caveats need to be noted regarding this present study. Firstly, there was the unrepresentative sample, such as more females than males, and the unlimited range of ages. This study could be improved by having an equal amount of males and females and a wider range of ages. Further research may want to look at gender differences in clustering. Females may attain a good memory compared to men and vice versa. Also this study could also be improved by implementing more conditions, for example, blocked by category, followed by blocked randomly. Blocked randomly, followed by blocked by category. This could be achieved by comparing the related samples with both blocked presentations to find out if organization for words in categories is better than mixed word instances in memory recall. Another limitation could be that more time is needed to encode the words more effectively. Practice may help participants encode more information. With more time added several conditions could be c ompared. The blocked by category could have two or three conditions but with new participants each time. Each of the conditions could have varied times to encode the blocked category lists. One condition could have two minutes; the next condition could have three. The same procedure could then be applied to the randomly blocked condition. All of the mean scores could be compared to establish whether time has an impact on clustering words together. A further limitation could be the use of the stopwatch to time the participants. The time may not have been entirely accurate. Participants may have had more or less time to encode the information presented to them, or to recall the words. This could have confounded the findings. To control for this, future research could perhaps use a large digital clock counting down the time left. Many memory experiments find out if clustering is effective for memory recall. According to Tulving (1972), clustering could help re-organize memory, therefore, attaining information more effectively. The findings of this study have a number of important implications for future practice. Psychologists could perhaps look at children of different ages. The experimenters could investigate when memory organization emerges. This could help provide an insight to when our organization for memory begins. Words could perhaps be visually presented to children in the form of animal characters which are either related or not related. The children could then recall the remembered animal characters. If the children recall more of the related animal characters, than the unrelated characters, then this may determine when organization for memory begins. Another reasonable approach is to look at whether organization for memory declines with age. Youths and pensioners could be compared to see which ag e group has a better organization for words. It can be concluded that organization of memory does aid semantic memory. The research literature has shown effective techniques such as clustering words together to help aid our memories. The results of this current study did show that memory is aided well, if our organization of words is structured. References Bousfield, W. A. (1953.) The occurrence of clustering in the recall of randomly arranged associates. Journal of General Psychology, 49, 229-240. Cofer, C.N., D.R. Reicher, G. M. (1966) Clustering in free recall as a function of certain methodological variations. Journal of Experimental Psychology, 71, 858-866. Craik, F.I.M. Tulving, E. (1975) Depth of processing and the retention of words in episodic memory. Journey of Experimental Psychology, 104, 268-294. Eysenck, M, W. Keane, M, T. (2000) Cognitive Psychology- a students handbook. Psychology Press. Gross, R. (2005) Psychology -The Science of mind and Behaviour. Typeset by GreenGate publishing services, Tonbridge Kent. Miller, S. (1984.) Experimental Design and Statistics. London, Typeset LTD. Kalat, J. W. (10Eds,), (2009). Biological Psychology. USA: Wadsworth, Cengage Learning. Tuving, E (1972) Episodic and semantic memory. In E. Tuving W. Donaldson (Eds.), Organization of Memory. London: Academic PressAppendices: Table of context Page Appendix 1: Design section: P12 Allocation to conditions. Appendix 2: Apparatus and Material section: P13 Instructions, word lists. Appendix 3: results section: P15 Raw data graph showing mean scores Appendix 4: Mann- Whitney U test: P17 Full calculations Appendix 5: Stimulus and response: P19 Participant 9, recalled words Appendix 1: The table below is showing the allocation of participants (Ps) to conditions. 15 females and 6 male participants were 1st year psychology undergraduates at City University. Each participant was currently taking part in a laboratory course in D112 of the social sciences building at City University. Therefore, the sample used was an opportunist one. Because there were two unrelated samples, an independent measures design was employed. Blocked by category condition Blocked randomly condition P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 P26 P27 P28 P29 P30 P31 P32 P33 P34 P35 P36 P37 P38 P39 P40 Ps =20 Ps =20 Appendix 2 Your investigation The aim of you experiment is to investigate whether presentation of category instances to participants in a blocked or in a randomly mixed fashion affects recall. You are thus attempting a replication of the study by Cofer, Bruce and Reicher (1966), and the basis of their finding and the other evidence mentioned above it would seem appropriate to state a directional (one tailed) hypothesis. State the null hypothesis as well. Remember that both the aims and the hypothesis need to be formally stated in a very clear and specific manner at the end of the introduction section of you report. Employ an independent groups design, with at least 10 participants in each of the two conditions (the larger you overall sample size the better). The dependent variable (the measure) will be the number of words correctly recalled. The independent variable (the manipulation) will have two levels corresponding to the different ways of presenting words to participants in two conditions. Describe the two conditions and any controls against confounding variables. Remember that, apart from the manipulation of the independent variable, everything should be held constant (standardized) across the two conditions of the experiment. Materials consists of 36 words (6 from each 6 categories) presented visually together as 6 blocks (lists) on one page; each block consisting of either instances of one category or randomly mixed instances of several categories. Condition A Condition B Apple Tulip Chair Apple Buttercup Cabbage Banana Buttercup Bed sofa Bed Giraffe Pineapple Rose Wardrobe Snake Cauliflower Wardrobe Orange Carnation Table Aeroplane Pineapple Lily Lemon Pansy Sofa Pea Elephant Motorbike Pear Lily Desk Carnation Pansy Lemon Dog Potato Car Ship Parsnip Lion lion Pea Bus Table Rose Chair Elephant Cauliflower Train Onion Desk Pea Monkey Cabbage Aeroplane Car Orange Bus Snake Parsnip Motorbike Tulip Monkey Dog Giraffe Onion Ship Banana Train Potato Instructions in a moment you will be asked to turn over the sheet of paper in front of you. On this sheet will be 36 words. You will have one and a half minutes in which to try and memorize all the words. After this period of this time the words will be removed. You must then immediately recall as many of these words as you can. Write them down on the piece of paper in any order you like. Results section should contain: Summary descript statistics comparing the two samples ( ie measures of central tendency nad dispersion, and perhaps graphs/charts) Summary of inferential statistics analysis by an appropriate test for the two independent samples ( e.g. Mann Whitney U Test ). The observed value of the statistic must be compared to a critical table value in order to determine its significance. Statement of the decision to retain or reject the null hypothesis will conclude the results section. Remember that both the raw data (individual scores) and all statistic calculations and formulae must appear in the separate Appendix (at the end of the report). They must not appear in the results section Appendix 3: Raw data Blocked by semantic category Participants Written free recall x/36 Blocked free recall randomly Participants Written Free recall x/36 P1 13 P2 30 P3 27 P4 26 P5 21 P6 22 P7 17 P8 21 P9 24 P10 24 P11 15 P12 25 P13 15 P14 28 P15 18 P16 19 P17 20 P18 15 P19 21 P20 27 P21 22 P22 21 P23 21 P24 26 P25 18 P26 17 P27 18 P28 29 P29 20 P30 23 P31 17 P32 33 P33 22 P34 21 P35 22 P36 23 P37 22 P38 29 P39 13 P40 20 Data of raw scores obtained from written free recall scores. Chart showing the two differences of mean scores between the two levels of the independent variable. The graph above shows the differences in mean scores for the two levels of the independent variable. The blocked by category condition showed that more participants recalled more words than the blocked randomly condition. Appendix 4 Full calculations of the Mann-Whitney U test Free -Recall Mann-Whitney U Test Blocked by Category Rank1 Blocked Randomly Rank2 N=20 N=20 13 1.5 30 19 27 20 26 13.5 21 13 22 8 17 5.5 21 6 24 19 24 11 15 3.5 25 12 15 3.5 28 16 18 8 19 3 20 10.5 15 1 21 13 27 15 22 16.5 21 6 21 13 26 13.5 18 8 17 2 18 8 29 17.5 20 10.5 23 9.5 17 5.5 33 20 22 16.5 21 6 22 16.5 23 9.5 22 16.5 29 17.5 13 1,5 20 4 R 1 = 210 Mann -Whitney U Test calculations U = N1 N2 U = 2020 + 400 + 400+210-210=400 N1N2-U = Uobs = (0) U = 400 = 400 400 = 0 Uobs = 0 = Ucrit = 138 = p =0.05 (1 tailed test) Since Uobs Reference: Runyon and Haber (1976)