Multi-Store Model of Memory

There are many theories which describe ways in which we remember things and why. 

The first proposal was put forth in 1968 by Atkinson and Sheffrin and was called the MSM. 

The sensory store gains information from the senses (eyes, ears, smell etc...). 

This information will only stay in the sensory store for 1-2 seconds, so it has a short capacity. 

Information gets transferred to the STM through attention, this is the process of repeating the information in the brain. 

When information is in the STM it is in a fragile and weak state.

The information is remembered for 18 seconds and so has limited capacity and duration.

If information is not rehearsed in the STM the new information will push it out.

Information gets transferred for STM to LTM through rehearsal, this is the process of thinking of a particular memory repeatedly so that when it's in your LTM it can be remembered for a long time as  the LTM is unlimited and information lasts forever. 

Evidence for MSM comes from Peterson and Peterson's study (1959).

They tested the duration of STM using syllables followed by a three digit number. 

After hearing the syllable participants had to count backwards from the number in 3's or 4's until told to stop after an increasing number of seconds and was then asked to recall the syllable.

They were told to count backwards to stop them from rehearsing the syllable.

It was found that participants remembered about 90% when there was only a 3 second interval and about 2% when there was a 18 second interval. 

This suggested that when rehearsal is prevented, STM lasts about 20 seconds at most. 

Conducted a natural experiment to test the duration of LTM.

They asked people of various ages to name people from their high school year book.

It was found that even 48 years later people were about 70% accurate. 

They persumably remembered the people as they were meaningful to the participants. 

The MSM suggests that the LTM is unlimited and information stays in there due to the process of rehearsing the information. The participants in this study would have rehearsed the names of people from their high school the entire time they attended school and so Bahrick's research supports the existance of LTM.

Brain scans have shown that the prefrontal cortex is active when individuals are working on a STM task (Beardsley 1997)

Whereas the hippocampus is involved when the LTM is engaged (Squire 1992).

This therefore shows the different memory stores. 

The validity can be questioned in the studies as they largly involved college students studying psychology and so they could have been subject to demand characteristics, that is, trying to guess the aim of the study.

It is likely that people aged 18-21 have been used and so results cannot be generalised to the population. 

The studies into MSM are good as they can be controlled but they tend to suffer from issues of validity such as demand characteristics because particpants used in research will often change their behaviour to please the experimenter. 

Lab studies related to MSM means that extraneous variables can be controlled, we can therefore be more certain that any change in the DV (e.g the lack of memory in Peterson et al's study) is due to the IV (e.g. the timed intervals between each of the syllables) and thus reach conclusions about cause and effect.

Presented particpants with a list of words that they could recall in any order.

He found that words at the beginning and end of the lists were recalled better than those in the middle.

This was because the words at the beginning (Primary effect) are recalled because they have been rehearsed and transferred to LTM and those at the end (recency effect) were recalled as they were still in the STM- this gives evidence for two seperate stores and the process of rehearsal. 

Found, in a patient with brain damage, that while his STM was seriously impaired, his LTM was intact suggesting information does not need to be rehearsed to go into the LTM. 

The MSM ignores individual differences and assumes everyone has the same memory-system structure.

However, the MSM is historically important in memory research as it was the first model that produced testable predicitions. 

The MSM led to s wealt of research to support the existance of 3 seperate stores. 

It was very important as from the MSM other theoris on how and why were remember things have been developed.