Decay Theory of forgetting from short term memory

The cognitive process whereby past experiences are remembered, probably the most fundamental constituent of human cognition, as without it other functions like perception, learning and language would not be possible.

Memory is often described as the capacity one has to retain information, recall it when it is required and to recognise it when one is exposed to it again. We remember things from the past and experience things in the present, therefore any past event that is evoked is evidence of memory.

Memory also is often referred to (though somewhat metaphorically) as a location where all these events, experiences and knowledge of a lifetime are accumulated. This is evident in theories of cognition which are reliant on divisions such as short term and long-term memory.

Some of the most compelling evidence that there is a distinction between short term and long term memory comes from patients with anterograde amnesia. This is a very insightful area of study. These patients frequently perform well when doing tasks which involve immediate serial recall of lists of e.g. words or digits, as long as they are not distracted between when they saw the initial stimulus and when they are asked to recall it. In general people find it easier to remember those words at the beginning of the list and at the end of the list than they do to recall words in the middle of the list. This is known as the primary and recency effect respectively. This occurs as the items at the end of the list are still likely to be present in short-term memory when recall begins. The anterograde amnesiacs show an intact recency effect but typically recall little else from the list.

Long-term memory differs in many ways. The capacity of LTM is obviously vast, compared to the “seven – plus or minus two” items that can be stored in short term memory. It is not apparent whether memories in long-term memory are also subjected to passive decay like in short-term memory.

Support that decay operates on active contents of mind was the empirical evidence that subspan material can be forgotten rapidly under conditions of minimal interference. This is demonstrated in work by Brown (1958) and Peterson and Peterson(1959).

The Brown-Peterson Task

They believed that some forgetting was simply due to the passage of time, and therefore decay. The method of the experiment was that participants where shown a simple three letter stimulus to try and remember, followed by a three digit number. The participants where first instructed to attend to the stimulus and then begin counting back from the three digit number shown by subtracting threes. The counting was to be done in rhythm with a metronome which was ticking twice every second. Participants where then asked to report the three letter stimulus they were first shown.

The results of this experiment were so surprising that many researchers wanted to replicate the study. What was so surprising was that memory of the simple three letter stimulus was only slightly better than fifty percent after three seconds of backward counting, then accuracy fell to just five percent after eighteen seconds of backward counting.

from Peterson and Peterson(1959)

While it appeared the Petersons had presented evidence of a simple decay function in short term memory, it was later discovered in subsequent studies that the essential ingredient – the distracter task which was considered to be necessary to eliminate or at least reduce rehearsal behaviour had in fact functioned as an interference.

Waugh and Norman (1965) Primary Memory questioned the assumption by the Petersons that there would be little or no interference from the distracter task to the memory test as the three letter word stimulus and the three digit numbers are categorically dissimilar. Waugh and Norman claimed the forgetting functions that the Petersons observed were not evidence of decay from short term memory, as the numbers spoken by the subjects during backward counting had interfered with short term memory trace then longer counting intervals would have provided more opportunity for interference as participants would have produced more numbers during the longer interval.

Waugh and Normans’ reanalysis of several short-term memory studies confirmed their suspicions. In their Probe digit task (1965) participants heard a list of sixteen digits read out at a rate of either one digit or four digits per second. The final item each list was a repeat of an earlier item and it served to the participants as a cue or probe to write down the digit which had followed the probe in the original list. For example, in the list 5 9 2 8 6 6 9, the probe digit would be nine and this would have cued the recall of two.

One important part of the experiment was the time taken to present all sixteen of the digits (sixteen seconds for group A and four seconds for group B). if forgetting was due to decay then there should have been a significant difference in the performance of each group to recall since more time had elapsed in group A. However the groups barely differed. Therefore Waugh and Normans results strongly indicated that forgetting had been caused by the number of intervening items between the critical digit and the recall test not merely by the passage of time. Therefore the Brown-Peterson task had not only prevented rehearsal but had also produced interference with the critical digit to be remembered ( which was not the intention).

Talland (1967) reproduced the Brown-Peterson task with two divergent distracter activities – one group performed subtraction (as before) and the other group simply read a list of the numbers that they would have had to say if they had been doing the subtraction. Despite there being the same retention interval for both groups, the group that did the subtraction did worse.

Peterson, Peterson and Miller (1961) tested different types of stimulus – nonsense words versus words, after 6 seconds (for both groups) of backward counting word recall was substantially higher than nonsense word recall. This again disagrees with simple decay explanations.

It is virtually impossible to test the simple decay theory adequately as a stimulus would have to be presented followed by a blank interval of time to enable decay to occur, then the recall test. However participants would then use the blank period of interval to rehearse the stimulus. But preventing rehearsal by introducing a distracter task causes interference. Decay theory would have to predict the same amount of forgetting for the same interval of time regardless of what distracter task was used since according to decay theory it is the passage of time that is the cause of forgetting.

It is difficult to determine whether or not information that is neither rehearsed nor displaced suffers any type of passive decay, however the fact that with rehearsal we can remember could show that it does, otherwise we would be able to remember, for example, a phone number just by sitting doing nothing.