The Correct Recognition Of Words Based On The Depth Of Processing - Research Paper Example

 The correct recognition of words based on the depth of processing The correct recognition of words based on the depth of processing The definition of memory shifts from one aspect to another when topics are changed. Memory is a pool and its components are the lanes of the pool; there can be as many lanes as you can fit in this pool if there is room for one to swim in. Same theory applies to memory, one can add as many components and altered definitions to memory if it pertains to the evidence that supports this component or definition. Memory can be defined as three components, sensory memory, which is when there is incoming information from any of our sensory functions. There is working memory or short term memory, which is the memory that can be held for a short period of time and has a capacity limit. Lastly, long term memory, which is defined to be the component of memory where information is never lost and can be retained at any given moment. However, these three components are not the only components of memory, a different breakdown of memory is included. This breakdown includes the idea of how deep one thinks about information. There are many variables that can help one remember information, capacity and duration are one, but depth is also another component. How deep someone thinks about a piece of information is critical in being able to retrieve it later. How deep someone thinks about information can be portrayed in many ways depending on how the information is presented. It can be in ways like in a sentence or next to another word that has the same definition, or even the number of vowels and consonants is presented. How deep one thinks about the information depends on how well they will retrieve it in a later task. Experiments prior to this one tested how recognition is affected by the depth of processing. The article by Craik and Tulving (1975), profoundly describes how processing words in different depth levels affects the retrieval of those words in a recognition task. Craik and Tulving used three levels of encoding, shallow, medium, and deep. Their shallow level included typescript words which was asking if the word was in upper case or lower case. The medium level of encoding was asking rhyme questions, if the words rhymed with other presented words. The deep level of encoding included fill in the blank questions. Each participant was shown these question formats in a sum of 60 trials. They were required to answer yes or no to each question, after the 60 trials were over, a pool of 180 words were presented and participants were asked to check all the words they saw in their first phase of the experiment. Participants recognized 18% of typescript words which was the shallow encoding. The mean percentage of words recognized for rhyme decisions were 37%, and mean percentage of recognition for the sentence decision was 65. The interpretation of this data would be that the deeper the level of process is, the better the performance of recognition. Our aim for this experiment was to replicate Craik and Tulvings’ (1975) experiment and see if we can retrieve the same results. We questioned the theory of whether the deeper levels of encoding will be recognized better than the shallow levels of encoding. We predicted that because we are deeply processing information and understanding its actual meaning, it will be easier to recognize in a retrieval task. Method Participants This experiment consisted of seventeen college students of both sexes. They were at a varied age and diverse ethnicities. Design and Materials The Participants were required to complete a two-phase process, in which phase one was different types of encoding levels, and phase two required them to recognize the words in phase one. This experiment was initially adapted from Craik and Tulving (1975). The independent variable were levels of processing, which included three levels, the shallow processing, the medium processing and deep processing. The shallow processing had the participant distinguish between the word given and a string of letters that consisted of v’s, for a vowel and c’s, for a consonant. This level did not require the participant to think about the word deep enough to understand its meaning, e.g., (the word sheep was presented and ccvvc was presented next to it asking the participant if this word matched this pattern, but at times a wrong pattern would also be presented to allow the participant to lightly think about the word like, cvcvc.) The second level of process was the medium encoding which consisted of rhyme words. The participant was required to determine if the given word rhymed with a second word that appeared next to it, e.g., (the word snake would appear and the word bake would be presented next to it asking the participant if both words rhyme, sometimes it would be the opposite where the word car and tool would be presented and the participant would need to determine if those words rhyme or do not.) The last level of processing was the deep level which consisted of synonym words, e.g., (the word generous and the word kind would be placed next to each other allowing the participant to think deeply into what both words mean, at times the words car and bottle would appear next to each other also requiring the participant to distinguish if they are synonyms or not.) The participants were required to determine if the initial word is a synonym to the second word presented. This level would allow the participant to think about the word deeply and understand its meaning leading them to be able to recall or recognize it easier in a later task. A one-way ANOVA test was conducted to measure all three levels together and three between subject t-test were also conducted to see if there is any significant difference between consonant/vowel level and rhyme level, the rhyme level and the synonym level, and lastly the consonant/vowel level and the synonym level. Procedure This experiment was conducted using the Cog Lab 2.0 manual, which showed the participants how the experiment is conducted. Instructions were given and participants were to press the space bar to start. There were two phases to this experiment, phase one included the levels of processing. With each pair of words that showed participants were required to either agree or disagree with what was presented. They had to press y for yes and n for no. When phase one was over with all three levels completed phase two started, and this phase consisted of a recognition task. Individual words appeared and asked the participants to press the same y, n buttons if they recognize these words from phase one of the experiment. Results The proportion of recognized words in phase two is associated with the levels of processing. The null hypothesis stated that there are no significant differences between all three levels of the independent variable. What this means is that all three levels of processing are equal in mean percentage for recognition. The alternative hypothesis states otherwise, that there is a significant difference between the three levels of processing. The rhyme level of encoding has a higher mean recognition than the consonant/vowel recognition, and the deep level of processing which was the synonym level, has the highest recognition mean percentage compared to the shallow encoding, consonant/vowel level, and medium level encoding, rhyme level. Figure 1 represents the participants mean recognition for the consonant/vowel level which was 56% with a (SD= .20). The mean percentage for the rhyme level, 60% with a (SD = .15). The last level was the synonym level which has the highest mean recognition of 79% with a (SD= .12). These results show the difference between all three levels. Clearly that deep level of encoding which was the synonym level had the highest mean recognition. The consonant/vowel level which was the shallow level had the least mean recognition. The analysis of these results tells us that the deeper level of encoding the better recognition there will be. To be able to provide any meaningful data further statistical analysis is required to show any significant differences between the three levels. The results that we obtained required us to use two statistical analyses to confirm the prediction that the recognition of words is associated with the levels of processing. A one-way ANOVA test was conducted and the results were, F (2, 32) = 15.875, p< .001, this means that the correct recognition of words was significantly affected by the levels of processing. Three t-test were also conducted to see which one of the three levels are producing the significant difference. The first t-test was between the consonant/vowel level and the rhyme level, results were, t Read More