Example essay: Contrast two models of memory

Two models of memory that will be contrasted in this essay are Atkinson and Shiffrin’s multi-store model of memory and Craik and Lockhart’s levels of processing model. The primary difference in these two models is that one focuses on the structures of memory (the MSM), while the other focuses on processes (the LOP). Some classic studies will be used to demonstrate the claims of each model as well.

Let’s start chronologically with the multi-store model, proposed in 1968. Atkinson and Shiffrin devised this model by combining a range of ideas about memory that had come from studies during the cognitive revolution. The major claims of the MSM focus on the structures of memory, or the stores. There’s the sensory store (aka sensory register), the short-term store (aka working memory) and long-term store. A&S cite some classic studies that support their claim that memory is not one unitary store, but information flows between these different stores.

The classic case study of HM is one such study A&S cite as support for a multi-store model of memory. HM had his hippocampus removed to treat his epilepsy. While the surgery successfully reduced his epilepsy, it also resulted in him not being able to form new memories. He retained most of his memory from before the operation and he could hold information in his working memory, but he was never able to retain new learning in his long-term memory. This, A&S argued, is support for multiple stores of memory – if our memory was one single store in the brain then damage to one type (e.g. long-term) would also result in damage to another type (e.g. short-term). HM showed this was not the case.

But Craik and Lockhart wrote the first major critique of the MSM in the 1972. In response to the MSM, they proposed the levels of processing model. The central claim of this model is that the deeper information is processed the more likely it is to be remembered. There are three depths of processing – structure, phonological and semantic. Structural means processing based on physical qualities (size, shape), phonological is processing based on sound and semantic is processing based on meaning. Semantic has the deepest processing and information is remembered best. We can see this is one key difference to the multi-store model because it’s not concerned with the physical storage of memory (like the MSM), but rather what is happening during encoding.

Support for the LOP model comes from the 1975 study by Craik and Lockhart. The aim of this experiment was to see how processing affects memory. Participants were shown 60 questions and a five letter, one syllable, noun as a possible response (e.g. shark; cloud; crate). The questions were written in a way so that participants either had to select “yes” or “no” as the correct answer. The types of questions required participants to process the information at different levels.

For instance:
-Structural: “Is the word in capital letters.?”
-Phonological: “Does the word rhyme with WEIGHT?”’
-Semantic: “Would the word fit in the sentence: “He met a ……. in the street?”

After the questions, the participants were given a list of words and asked to circle those they remembered. The results showed that words processed semantically had the highest rate of accurate recall (80%), second was phonological (50%) and finally there was structural (15%).

This is evidence supporting the LOP and C&L’s critique of the multi-store model. They said that A&S focused too much on the way all information travels between stores the same way, regardless of other factors like encoding.

That being said, the MSM wasn’t just about structures. The model also focused on control processes, such as attention, rehearsal, search and retrieval. Information flows between the stores, they argued, based on these control processes (e.g. sensory to STS through attention, STS to LTS through rehearsal and LTS to STS through search and retrieval). But the MSM put too much emphasis on rehearsal as the primary way information is transferred, according to C&L, and that other factors like depth of encoding are important.

 Another interesting comparison is the chronology. The MSM came after the studies that supported the model, like HM’s case study, the primacy and recency effect studies by Glanzer and Cunitz and the trigram studies by Peterson and Peterson. The LOP, on the other hand, was proposed in 1972 and the main study that supports it was published three years later in 1975.

Perhaps another point of comparison is the origins of the theory. The MSM was an attempt to combine a range of ideas that emerged from studies in the years prior. It was hoped the model would consolidate and explain the range of findings coming from the cognitive revolution. The LOP, on the other hand, it would appear was devised solely as a critique of the MSM. The motivations for both models are different.

In conclusion, we can see the LOP and MSM both have empirical evidence to support their claims. Whereas the MSM focuses on structures and processes, the LOP is more focused just on processes. They also differ in how and why the theories and studies originated. (Approx. 850 words).

Disagree with my marking? Have other contrast points that could be added to the essay? Leave a comment.