Study Bay Coursework Assignment Writing Help
The aim of this experiment was to investigate the impact of levels of processing (Independent Variable) on our memory (Dependent Variable). We hypothesized to find that a deeper level of processing lead to a better memory and in turn, greater recall. The experiment controlled variables such as environmental conditions, age of the target population and educational background of the participants. The experiment uses the Independent Measures design and the participants are a convenience sample of teenagers. We reduced Practice Effect and Fatigue Effect by using the Single-Blind Technique. The results proved the hypothesis: Semantic encoding lead to greater retention and a deeper trace of memory whereas Structural encoding lead to a shallow processing of memory. These results are replicated in Craik and Lockhart’s (1972), Levels of Processing model. The researches done by Hyde and Jenkins (1973) and Craik and Tulving (1975) also prove the same results: deeper levels of processing lead to better recall. The Mann-Whitney U test also made a clear distinction in the number of words recalled due to deep and shallow processing. The study also raised some limitations such as generalizations, rehearsal, ecological validity and others. In totality, the experiment effectively manipulated variables and reproduced accurate results.
Introduction
Research Question: To what extent do levels of processing influence memory of words?
Learning occurs through many different processes. The Levels of Processing are used to explain why we have a deeper trace of some things and a shallow trace of the others.
The research by Craik and Lockhart (1972) on levels of processing serves as a backbone for cognitive psychology. In contradiction to the Multi-Store Model by Atkinson and Shiffrin (1968) that divides memory into different stores (STM and LTM), the levels of processing model by Craik and Lockhart suggests that memory occurs through different processes.
Shallow Processing is divided into Structural (visually appealing words) and Phonological (sound words) encoding whereas, Deep Processing is based on Semantic (meaning of words) encoding. They proved through their research that “trace persistence is a positive function of the depth to which stimulus has been analyzed” which means, Elaborative Rehearsal (analyzing the words) lasts longer than Distinctiveness (distinguishing between words).
Another research by Craik and Tulving (1975) is similar and furthermore validates the former research. They conducted an experiment where the participants were divided into 3 groups. Each participant was given a list of 60 words through one of the three tasks, which tested the three levels of processing. The memory recognition test concluded that the words semantically processed had a greater recall than the rest.
Hyde and Jenkins (1973) also carried out an experiment using the incidental learning (informal learning) technique. They divided their participants into different groups and conducted one of five tasks on each. After the surprise recall test, they concluded that those tasks, which involved semantic processing, had a better recall and in turn deeper processing.
Elias and Perfetti (1973) performed a study using the same technique. It involved a rhyming task for a list of words along with finding synonyms for the same. Participants in the synonym task were able to recall more words than the rest, drawing the same conclusion as above.
Hence, I chose to investigate the relationship between levels of processing and the recall of words by conducting two asks: One in which the first group of participants need to identify the word color and the other in which the second group needs to make sentences with the given words.
HYPOTHESES
Null Hypothesis: There will be no influence of the levels of processing on memorizing words.
Non-directional Hypothesis: There will be a significant difference in the recall of words semantically processed to those structurally processed.
Directional Hypothesis: The number of words structurally processed will be leave a deeper trace than those semantically processed.
METHOD
Design
The participants were given a Consent Form and were not forced to be a part of the investigation. They were given clear instructions at the beginning of the task and debriefed at the end.
Independent Variable:
Deep and Shallow Levels of Processing
Dependent Variable:
Number of words recalled
Control Variables:
Environmental Conditions
Age of the student
Educational level of the student
Number of words
We reduced Demand Characteristics by using the Single Blind technique as we conducted a surprise recall test at the end of the task before which the participants were not allowed to interact with each other. It was also mentioned in the consent form that their identity will remain hidden. Hence, this reduced the Fatigue Effect and Practice Effect.
We also used Independent Measures in our design as each participant was part of either group and were allowed to perform either one of the tasks involving deep or shallow levels of processing. Therefore, this increased External Validity.
The Mann-Whitney U Test was conducted as it is a non-parametric test and the sample participating in the research does not exceed thirty, hence it proved as a suitable calculation test representing statistical data for our test.
Participants
We used Non-Probability Sampling under which we used Accidental Sampling or Convenience Sampling keeping the age group constant. It was also the most convenient for an experiment to be carried out in a school setting.
We conducted the experiment on 20 participants in total, and we divided them into two groups of 10 participants. In consideration of gender, the participants were randomly mixed and each group consisted of an uneven number of males and females.
The target population included 16-year-old children having a similar educational level and a part of IB Year-1. As a control, the participants in Group 1 (Deep LOP) completed the task together at the same time and participants in Group 2 (Shallow LOP) did the same. This reduced the chances of interaction between participants.
Materials
Pen
Stopwatch
Appendix A- Consent Form
Appendix B- Parents Consent Form
Appendix C- Standardized Instructions
Appendix D- Word lists and attached Task sheets
Appendix F- Debriefing Letter
Blank sheets of paper
A classroom setting was required to conduct the experiment.
Procedure
The procedure was identical for both groups, Group 1 (Deep LOP) went first and then followed by Group 2 (Shallow LOP)
The participants were not allowed to interact at any point during this experiment.
Each group was taken to a comfortable classroom with sufficient chairs and tables for 10 participants.
They were then handed the Consent Form (Appendix A) and Parent’s Consent Form (Appendix B) for those above the age of 16.
Thereafter, the Standardized Instructions (Appendix C) for each group respectively, was handed out. The participants were allowed to withdraw at this point. For those who did not clearly understand, the instructions were explained individually.
They were then given the Word lists and Task sheets (Appendix D) but were not allowed to begin until told.
Group
Time Limit
Task
Group A (Deep LOP)
8 minutes
Make sentences with the given words.
Group B (Shallow LOP)
2.5 minutes
Identify the color of the letters of each word and write how many were blue, green and red in the given blanks.
The time limit was maintained using a stopwatch and after the time was up, the participants were asked to keep their pens down while we began collecting the answer sheets. If a participant completed early, the rest of the time was used to revise and recheck their answers.
Then after, we announced the surprise recall test and a blank sheet of paper was given out to each participant.
They were given 1 minute to complete their recall test and the time was kept using a stopwatch.
The answer sheets were collected after the completion of this task and they were asked to maintain silence until they left.
The Debriefing letter (Appendix F) was then handed out. After it was read, we collected it and the participants of this group were asked to remain seated until we gathered the nest group into another classroom.
Results
Group A (Deep LOP)
Group B (Shallow LOP)
Mean
7
3.8
Median
7
4
Mode
7
3,4 (Bimodal)
Range
4
2
Standard Deviation
0.748
1.154
Discussion
The aim of the experiment was; to what extent do levels of processing affect our memory of words. This was proved by the results of our investigation as the participants who made a meaningful sentence with the words had a deeper level of processing compared to those who merely skimmed through the words by identifying the color of the letters.
This study reproduced the same results as that of Craik and Lockhart (1972). The variables were the same for all the researches mentioned in the Introduction (Pg. 4). The experiment conducted by Craik and Tulving (1975) reinforced that the depth of processing increases as you go further from Structural processing to Phonological and the deepest at Semantic processing. Our research tested only two of these conditions, the Structural encoding was the task where the colors of the letters were required to be identified and this proved to be a Shallow level of processing. The second task required the participants to make sentences with the same word list, with this the words were processed deeper in the brain. The cognitive function of memory is affected by many conditions such as these. The research by Hyde and Jenkins (1973) also proved the same hypothesis with 3 added tasks in comparison to our experiment. They also concluded that more words were recalled as a result of Semantic processing.
The statistical results also proved the same; the mean, median and mode for Group A was a greater number compared to Group B showing a greater number of words recalled due to deeper processing. According to Craik and Lockhart, ‘memory is a by-product of perception’ and if words in our brain are perceived deeper, our memory is stronger in the long term. The results also replicated that of Craig and Tulving’s experiment as they proved the same with the help of their analysis. Hyde and Jenkins conducted a wider, in depth analysis reproducing that the recall is directly proportional to the depth of processing. We also incorporated incidental learning in our test; with the help of a surprise recall test and Convenience Sampling.
By using the same classroom setting in both the cases, we made sure that the participants are not only comfortable but also in an appropriate condition to answer the task. With this the environmental conditions were controlled. The age of our target population was the same, and all the children were from the same grade. This kept the age and educational background of the participants constant. The independent measures design ensured the elimination of the Practice Effect and Fatigue Effect. The Single-blind technique also ensured that Demand Characteristics were avoided.
Our participants were given a Consent form (Appendix A) before the investigation and were thanked and debriefed at the end (Appendix F). They were allowed to withdraw at any point during the experiment was being conducted. Their anonymity was protected even after we released the results, which they chose to view or obtain. The experimenters were also conscious not to disrupt the participants in any way, which would hinder their completion of the test and in turn the results.
A laboratory experiment like this one questioned ecological validity as the variables could have had an impact on the responses in participants. We also avoided taking the task sheet before the entire group completed the experiment and this may have resulted in rehearsal, affecting our analysis to an extent. The direct comparison of the results between the two groups might have also created some uncertainties as the participants were different and their cultural and social background could have impacted. As we have studied, cultural and social factors affect our memory and one tends to remember things they can relate to. Some words might have evoked this resulting in inaccuracies.
It can be very difficult to make generalizations by using Opportunity Sampling, as it may not have met the exact requirements for the target population. Random Sampling could have been a better options but it is more widely distributed and therefore time consuming.
There was a conflict between concrete and abstract words affecting the recall of a participant. As suggested by Walker, I., & Hulme, C. (1999), concrete words (objects/events that avail to our senses) are recalled more than abstract words (an idea/concept). To avoid this, the word list could have included either one of those sets. Another study by Weldon, M.S., & Bellinger, K.D. (1997) concluded that words learnt collaboratively are leaving a deeper trace in the memory compared to individual processes in remembering. This contradicts our research as the task was given to one participant, separately. To avoid this, the test could have been carried out collectively, in order to compare the results.
The difficulty of words was maintained to a minimum but this cannot be judged, as a participant might have been familiar with some and not with the others. To avoid this, we could have consulted the word list to a control group and an English teacher.
In conclusion, this experiment effectively manipulated variables and obtained accurate results to prove the hypothesis. The deeper level of processing (Semantic encoding) leaves a deeper trace of memory, resulting in a better recall. These results support the Levels of Processing theory by Craik and Lockhart (1972).
