One of the most intriguing puzzles of life is the inner working of memories. Countless scientists have attempted to unravel the mysteries of how memory is formed, stored, retrieved and in many cases, forgotten. One interesting finding about memory is the Serial Position Curve or Effect (SPC, SPE), which can be explained using developed theories and principles of memory. The study of the Serial Position Effect is important because it relates to general memory usages such as improving short-term memory, as well as translation of short-term memory into long-term memory.
The Serial Position Effect was observed by Glanzer & Cunitz (1966) when they conducted a study which presented participants with lists of items, usually 10, 15, 20, or 30 words, one at a time, after which participants were asked to recall as many words as they could. When recall was immediately after the last word, plotted results showed that the percentage of correct recall for words at the beginning and end of each list was much higher than mid-list words, giving a U-shaped graph. These results are known as the Primacy Effect (better recall of earlier words) and the Recency Effect (better recall of most recent words). However, when recall was delayed by just 15 to 30 seconds, and participants were asked to perform an interpolated task (e.g. counting) to prevent rehearsal, the Recency Effect disappeared while the Primacy Effect was still present (Passer, Smith, Atkinson, Mitchell, & Muir, 2004, p. 299). Some memory theories and principles can be applied to examine what causes these results.
Under immediate recall conditions, the resulting SPC can be divided into three sections corresponding to the proportion of recall for each item: Primacy or Primary, mid-list words, and Recency. These differences in recall percentages can be explained by applying the Duplex Memory model, which is similar to the three-stage mode of memory, in conjunction with part of the Interference Theory and active rehearsal.
The Duplex Memory model depicts memory as two distinct but interrelated sections: short-term memory (STM) and long-term memory (LTM). Short-term memory is also known as ‘working memory’ because it holds information that one is aware of at any moment, and actively processes information. There are three components of working memory: auditory (phonological loop) which is when we repeat something mentally; visual-spatial -- a mental ‘sketchpad’; and the central executive, which directs our actions by deciding how much attention to give each of the two other components and recalls long-term memory. Short-term memory lasts just around 20 seconds, and has a limited capacity of 5-9 meaningful items. Long-term memory, on the other hand is more permanent. In fact, unless the brain is damaged, it is capable of forming new long-term memories until death, and so far, research suggests that its capacity is unlimited and memories may last up to a lifetime. (Passer et al., 2004, p.296)
As stimuli enter sensory storage, they are transferred as codes into short-term memory. Typically, there are four forms of codes: visual (mental images), phonological (sound), semantic (meaning of the stimulus), and motor (physical actions). Usually the form of the code is different from the original stimulus (Passer et al., 2004, pp 296-7). For example, when learning to play a piece of music, the original stimulus is printed sheet music (visual), while the code formed is usually motor, since it is impractical to memorize the position of every note on a scale. Instead, it one usually memorizes how fingers must move to produce the correct notes. Once in the STM, rehearsal of information occurs. The code may still not be encoded and therefore forgotten, encoded and transferred into LTM, or immediately recalled. Once in the LTM, a memory can be retrieved through long-term retrieval processes, or if not reviewed, eventually forgotten (Biggs, 2008).
In the Serial Position Effect study, as the first few words in a list are shown to the participants, they have the time to repeat the words to themselves until a new word is shown. Therefore, through rehearsal, the first few words enter participants’ long-term memory. However, as the list becomes longer, there are too many words for the participants to repeat before the newest word is presented. Therefore, the new words are not encoded into the LTM, and stays in the STM. Given the short “shelf-life” of short-term memory and its limited storage, the words following the first few are quickly forgotten and “knocked-off” as new words follow. Thus, the recency effect is present when recall is immediately following the last word, as they are still lingering in the participants’ short-term memory. However, when recall is delayed, and participants are prevented from rehearsal by simple tasks such as counting, the most recent words are forgotten because they are only stored in short-term memory and are replaced by the simple task. Delayed recall does not affect the primacy effect because those words are stored in the LTM. (Passer et al., 2004, p 299)
The Interference Theory is another possible explanation for the inability to transfer more words into long-term memory. Interference occurs when we forget information as a result of other items in the LTM preventing our ability to recall it. There are two forms of interference: proactive interference and retroactive interference. Only Proactive interference may influence the Serial Position study, since it occurs when material learned in the past interfere with recall of newer material (Passer et al., 2004, p 316). Given the fact that as the list of words get longer, it becomes harder and harder to rehearse words as more are presented. Hence it is likely the earlier words are ingrained more deeply than the mid-list words. So when the participants attempt to recall mid-list words, their minds can only recall the more familiar earlier words. Retroactive interference, which is a consequence of newly acquired information interfering with the ability to recall information learned at a earlier time (Passer et al., 2004, p 316), does not apply to the Serial Position Curve, despite resembling the recency effect. This is because participants are not actually learning the new words. Rather, when recall is immediate, participants recall a fresh image of the most recent words, which have replaced preceding words in STM.
Two theories, which do not directly impact the serial position curve, but affect overall memory, are Levels of Processing and the Duel-Coding Theory. Levels of Processing in memory is the concept that the deeper a concept or piece of information is processed, the better it will be remembered (Passer et al. 2004, p 300). There are three general levels of processes. The shallowest process is Preliminary Process, which involves feature detection, such as lines, angles, colours, etc. Slightly deeper, is Pattern Recognition; i.e. do features match previous knowledge? This is the level at which words in the serial position effect study are processed, since not enough time is given to participants to do much more than recognize the words. The deepest process is semantic analysis, which focuses on the meaning of stimuli (Biggs, 2008). The Dual Coding Theory is similar in that it says that memory can be enhanced by encoding information in both verbal and visual codes. Doing so would ensure a deeper consideration for information and concepts one is learning. One way to apply the Dual Coding Theory is the Method of loci, an ancient Greek process which assigns each concept one is attempting to learn to a section in an imaginary environment (e.g. a room in a house), then taking a mental ‘walk’ through that environment and connect each concept to it’s assigned place. The Dual Coding Theory is harder to apply to abstract concepts such as feelings (Passer et al., 2004, pp 303-4).
In learning new concepts, rehearsal is also useful. There are two types of rehearsal: maintenance rehearsal, and elaborative rehearsal. Maintenance rehearsal is the simple repetition of information. This is not overly effective when trying to learn complex concepts, but still necessary when learning rudimentary things like an alphabet, or as in the case of the Serial Position study, when not enough time is given to do anything else. Elaborative rehearsal focuses on the meaning of new information or relating it to something we already know. This is more effective in transferring information to the LTM because it utilizes a deeper level of processing since it involves semantic analysis (Passer et al., 2004, p 301). Neither Dual Coding Theory nor Levels of Processing explain the Serial Position Effect because participants are not given enough time to utilize either concept.
The SPC may be viewed as a representation of what likely happens when a student crams for an exam the night before, something most students have done at least once. When cramming, one often tries to get through the material as fast as possible, simply because of the sheer volume of material that needs to be covered. In doing so, the student is unlikely to stop and think about the material because he or she is running out of time. Hence, like the participants of the study, during the exam, the student can recall the first few lessons which he or she have reviewed repeatedly, as well as the most recent, since those are still in their STM. However, the student would probably experience difficulty recalling all the material in the middle of the course.
Therefore, to form long-term memories of course material for more efficient information retrieval, it is important to spread studying over several days, and use elaborative rehearsal methods. Once all the material is encoded into long-term memory, it is important to review the material regularly to keep it fresh in your mind, as the Decay Theory suggests that with time and disuse, the physical memory trace in the nervous system fades away (Passer et al., 2004, p 316). Therefore, by keeping in mind the limitations of our memory, as shown my the Serial Position Effect, and incorporating established memory models and theories to our study habits, we can greatly improve our academic performance in terms of recalling information.
References
Biggs, TC (2008) Chapter 8 notes. Course Website.
Passer, M.W., Smith, R.E., Atkinson, M.L., Mitchell, J.B., & Muir, D.W. (2004). Psychology: Frontiers and Applications (2nd ed.). Canada: McGraw-Hill Ryerson Limited.
