Effects of Deep and Shallow Processing in Memory Recall

Abstract

            Memory is how people draw their past experiences and use the information in their present. Memory can also be defined as the processes and structures involved in retrieving and storing necessary information. Memory is thereby a by product of the depth of processing information; thus, there is no concise difference between long term and short term memory. The processes in memory are phonemic processing, structural processing, and semantic processing. The objective of the report is to study how shallow and deep processing affects memory recognition. It aims to process a memory recognition test. The methodology of the research follows repeated measure design and random sampling strategy. 153 participants were selected in the research, who filled the assessment papers. The research results indicate more significant results in the titled and the upright questions than the man made questions. Also, the analysis indicates that the memory retention score is more significant for the deep memory than the shallow one. So, the findings of the research were consistent with the hypothesis. As far as the discussion is concerned, many previous researches supports the findings of the research, which was the research’s strength. The research limitation includes unequal distribution of gender in a sample by which the gender differences cannot be counted. This research can be applied to develop various strategies for enhancing memory recognition and learning. The results were interpreted by using one paired t test after doing the descriptive analysis. Several errors were counted and compared as well in both deep and shallow learning. This is how the research is conducted. However some improvements need to be made for future research.

Introduction

            Memory is how people draw their past experiences and use the information in their present ((Hanslmayr and Staudigl, 2014). Therefore, memory is defined as the processes and structures involved in retrieving and storing necessary information (Cowan, 2008). Levels of processing models are focused on the depth of the processing involved in our memories, which then predicts the more in depth evidence. Depth in this scenario refers to the ideology extracted from the thinking rather than how many times the analysis has been performed. Therefore, the primary assumption of memory is what happens when the information is processed. Memory is thereby a by product of the depth of processing information; thus, there is no concise difference between long term and short term memory. The processes involved in memory are; phonemic processing, structural processing, and semantic processing. Structural processing deals with the appearance of the objects (i.e., the physical qualities of something). Phonemic processing deals with the sounds encoded. Sematic processing is when we encode a words meaning and relate it to other similar meanings or words. Elaboration rehearsal is in deep processing where there is a more meaningful analysis (e.g., association, thinking, and images) that leads to better recalling.

            The objective of the report is to study how shallow and deep processing affects memory recognition. So, the report aims to process a memory recognition test. Similarly, it also indicates whether deeply studied things are well remembered like the shallow studied things. The hypothesis of the research study is that memory retention scores better in the deep condition.

Method

Participants

            The participants used in this context were from the previous years. There were 153 participants in total, out of which 19 were males and 134 females who were undergraduate psychology students at Middlesex University. None of the participants were forced to be a part of the research study, and they were requested to take part in the study course. At the same time, the others do not specify their age and sex. There were both males and females whose ages ranged from 17 50. The standard deviation of the age is 5.26, with a mean of 2.47. A random sampling strategy has been used in the allocation of participants.

Table no: 01 Descriptive Statistics

Design

            A repeated measures design was used. Two exploring styles were examined, which are deep and shallow memory scores. The dependent variable is the memory score, while the independent variable is required to complete the task.

Materials

            The participants were required to have a paper and a pen where they would record their answers. The questionnaire was a sematic diagram showing various images. The objects used were a standardised set of 260 pictures: norms for visual complexity, familiarity, image agreement, and names agreement. Thirty two objects were presented with 16 deep and 16 shallow. Twenty four of this appeared in the memory test. The eight were removed as a primary buffer (i.e., the first and last object in each block), along with the new twenty four objects. The images were in black and the middle of a white computer background screen, size 400x400 pixels.

            The memory score sheet had 48 objects (24 new and 24studied) on a piece of paper with a Y/N response.

Figure 1: an example of an object shown in the deep condition within the experiment.

Procedure

            The participants were required to state whether the images were either natural or artificial. The test was required individually as each participant was given their score sheet to fill. The research was conducted in a lecture theatre where the individuals were given assessment papers to fill. Due to many participants, the independent variable was derived from the average time required to complete the task. This was done by giving a limited amount of time per question. The participants responded using a questionnaire. The objects were presented for 500 milliseconds. In the memory test, the participant had 5 minutes to decide which objects are natural or artificial or whether they were tilted or not. Informed consent was taken before experimenting. Also, the information regarding debriefing was provided to the participants to let them know about the participants intentions. The participants had to circle Y (yes) for objects they thought were natural and N (no) for objects they thought were artificial.

Results

            The scores were presented in the score sheet and evaluated by overlooking the participants correct answers. All the scores of 153 participants were counted based on memory retention. The following graphical illustration presents the frequencies of the incorrect rejections and the frequencies:

Figure no: 01 Total Incorrect misses

The correct answers were ticked and counted. Questions that were ticked on both the Y and N part were excluded.  A paired sample t test showed that most participants scored a higher score in the tilted and upright questions (M= 17.56; SD = 4.43) as compared to the natural or man made questions where the scores were (M = 2.24; SD = 2.36), t (152) 11.7, p < 0.001, 95% Cl [1.86, 2.61]. 

Table no: 02 Paired Sample T test

Discussion

            The research findings reported that the participants showed greater results in the titled and the upright questions than the man made questions. Also, the analysis indicates that the memory retention score is more significant for the deep memory than the shallow one. So, the findings of the research were consistent with the hypothesis.

            Several studies have been conducted over time to assess memories retention using tests and level of cognition. In Craik and Tulving (1975) study, the findings outlined that memory cognition level determines how well an individual retains a memory. If the memory is shallow, then the individual is less likely to retain it. Deeper memory encodings are more comfortable to remember but harder to process and encode. Many researches are present in support of previous literature, indicating that memory retention score is more excellent in deep learning than shallow learning. Marzi (2010) conducted a study to analyse the deep and shallow learning. His research findings reported that deep learning involved deeper level of memory processing, so memory retention is more in deep learning. Another theoretical based research finds out the same pattern of memory retention and focuses on the retention interval and processing level. He find out that memory retention requires some period to shift towards long term memory. So, deep learning leads to memory retention. Both these researches supports the thesis statement and the findings of the research.

            The evidence to the above findings can also be supported by research in which the researchers use deep learning strategy to enhance the memory retention in between students. The theoretical background of the research showed that deep learning associates with a greater level of memory processing and retention. A survey was also done by Liu and colleagues in 2020 to see the deep learning and memory retention. The survey finds out that memory retention is a powerful strategy to enhance learning and convert STM to LTM. So, the directions of this research were also in support of the research hypothesis and findings. The unexpected findings analysed in this research involve different techniques used in the case of deep learning. All the previous literature finds the practical implications of in depth learning strategies, which can be used for students to enhance their memory.

            As per the critical assessment of the research, the strengths involves that the theoretical approach is an improvement on Atkinson and Shiffrins account of transfer STM to LTM. For instance, the elaboration rehearsal can lead to recalling information rather than just maintaining the rehearsal. The different levels of processing models have changed the direction of memory research. It has shown that encoding is not a simple and straight forward challenge. This has widened the focus from observing the long term memories as simple storage units and seeing it as a complex processing system (Maki and Schuler, 1980). Craik ideas have led to diverse experiments that have confirmed the superiority of deep semantic processing for remembering information. This explains why people remember some things much better than others. The explanation of memory is essential in todays life as it highlights how elaboration, which requires deeper processing of information, can help memory. Regardless of the strengths, there are several criticisms of the level of processing theory. First, the concept of depth is vague and cannot be observed. Thus, it cannot be objectively measured. Second, deeper processing takes more effort than shallow processing. However, the depth of processing makes it more likely people will remember something. Lastly, it does not explain how the more in depth processing results in better memories. Researchers state that the level of processing theory describes rather than explaining. (Bradshaw and Anderson, 1982). Another weakness of the research is that the sample is limited and involves social desirability. Also, the study cannot be generalised. The research could have been even better if it involves an extensive range of sample from different cultures. The amount of processing done is ideally how the individual did much analysis or assessment. The individual stops the analysis as soon as the task analysis in question has been achieved. Therefore, once the individual has determined whether the object is tilted or uptight, the processing stops. Therefore it can be stated that the processing of a stimulus is done over a particular series. Simple physical attributes are processed first, while other qualitative attributes that are difficult to process are processed later. This suggests that the time taken to process memory can be used as a scale to determine how deep or shallow the memory is (Craik and Lockhart, 1972). Ideally, since an objects physical attribute, if determined faster, the individual will determine the objects orientation first. This will take a short time when the individual first assesses the stimulus when they make the judgment. The individual will take less time to determine whether the object is tilted or upright than the time is taken trying to discern if the object is artificial or natural.

            Further studies can be improved by enhancing the sample from different cultures. A cross cultural study will provide more generalised results and will reduce the social desirability factor. Involving all these factors will improve the research further. The reason for conducting the future result is to enhance the reliability and validity of the research. In future study, gender differences in memory retention can also be seen. This will let enhance the literature based on the gender differences in between memory retention. Changing this aspect of the research will positively contribute to analysing the memory retention between male and females.

            To conclude, the findings of the research are in support of the hypothesis. Also, the previous literature goes in support of the research findings. Deep learning is associated with more excellent memory retention as compared to shallow learning. A critical assessment of the research indicates that it has both strengths and limitations. The research needs to be improved by overcoming the limitations of the research.         

Conclusion

            The need to better understand the dynamics of memory retention is still imperative. Studies have already outlined that the level of retention of memory does not affect the individual willingness to learn, how difficult the task is, or the effort involved. The retention of memory is directly affected by the qualitative aspect of the object. In other words, events with many meanings are better remembered than others (Hanslmayr and Staudigl, 2014). In this case, whether the object is tilted or uptight does not give the meaning of the object itself or the individual who is observing it. However, if the qualitative object attributes are reviewed, it will create a better and meaningful memory aspect. It is essential to pursue further studies in this field because it cannot be stated that memory retention is determined by how meaningful something is. This statement is too vague to build a theoretical framework upon.

            Other studies have outlined other factors and aspects that may affect how memory or stimulus is processed. Aspects like rhyme in a sentence make it easier to remember, and if a particular series of objects have a pattern, the pattern is simpler to remember. Ideally, some scholars have argued these rhythms and patterns may affect how an objects general information is processed. For instance, in the given test, the alternating sequence of the natural or tilted object may lead to a pattern that may be easily remembered by the subject (Loaiza and Camos, 2016). This may affect the results that may result from the study. Several implications also arise from conducting the study, as outlined by other scholars. For instance, it has been argued over time that the viewpoint of analysing both the structural aspects and meaningful aspects of an object can be the same (Craik and Tulving, 1975). This means that the two aspects are qualitatively similar, and possibly how a shallow memory is created may be similar to how a deep memory is created.

            Further implications include the fact that humans are not proficient in memory when they perceive objects under learning instructions. This means that the subjects are given individual instructions to guide them towards achieving specific analysis tasks in their mind. If the subject could have started the operation independently, scholars suggest that they report higher memory retention and recall performance. In this experiment, instead of instructing individuals to state if a particular object listed is natural and artificial, the participants would select an object of their own and discern if it is natural or artificial. In fact, in such a case, the participants analysis range would be broad to identify all the objects physical and qualitative aspects. This implicates the experiment as individuals would have a lower performance while the experiment is being conducted.

            To avoid such implications in the future, more research should be done to establish a solid theoretical framework to build the research. Nonetheless, each study adds a particular aspect of memory retention to the literature review. This study concludes that memory retention is higher when a deeper level of processing is required to assess the qualitative attributes. In this case, there is higher memory retention where the participants were told to state if the object was natural or artificial. The lower performance in memory retention while listing whether the object was uptight or titled is explained by the low amount of processing needed to analyse its physical attributes. Memory retention is also high when the individuals interaction with the object in question is high. Even so, extensive research carried out in the future may shed some light on the exact operations done when encoding a memory to determine if the memories will be shallow or deep.

References

Bradshaw, G.L., & Anderson, J.R. (1982). Elaborative Encoding as an Explanation of Levels of Processing. Journal of verbal learning and verbal behaviour, 21, 165 174.

Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal behavior, 11, 671 684.

Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268–294. https://doi.org/10.1037/0096 3445.104.3.268

Equivalently following elaborative vs. non elaborative processing. Memory, 22(5), 470 480.

Cowan, N., 2008. What are the differences between long term, short term, and working memory?. Progress in brain research, 169, pp.323 338.

Forrin, N. D., Jonker, T. R., & MacLeod, C. M. (2014). Production improves memory

Hanslmayr, S., & Staudigl, T. (2014). How brain oscillations form memories—a processing based perspective on oscillatory subsequent memory effects. Neuroimage, 85, 648 655.

Kelley, C. M., Alban, M. W., Lindsay, S., Roediger, H. L., & Yonelinas, A. (2015). Constrained retrieval in recognition memory (pp. 198 212). New York, NY: Psychology Press.

Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2020). Deep learning for generic object detection: A survey. International journal of computer vision, 128(2), 261 318.

Loaiza, V. M., & Camos, V. (2016). Does controlling for temporal parameters change the levels of processing effect in working?

Maki, R.H., & Schuler, J. (1980). Effects of Rehearsal Duration and Level of Processing on Memory for Words. Journal of verbal learning and verbal behaviour, 19, 36 45.

Marzi, T., & Viggiano, M. P. (2010). Deep and shallow encoding effects on face recognition: an ERP study. International Journal of Psychophysiology, 78(3), 239 250.