Remembering and Forgetting

Introduction

Although some processes may apply to both STM and LTM it is convenient to consider the two types of memory store separately.

Explanations of forgetting in STM

Probably the two most well-known theories of forgetting in STM are decay theory and displacement theory.

Decay theory

Decay is the simplest explanation of forgetting. Decay is what supposedly occur when the passage of time causes us to forget. The memory trace (or engram) fades and the memory is no longer available. This theory is popular because it appears to fit with common-sense views of the way memory works. One of the most influential attempts to explain forgetting in terms of decay was by Hebb (1949).

However, it is very difficult to prove experimentally that decay actually occurs. For example, Peterson & Peterson’s results can be explained in terms of decay of the memory trace over the retention interval, but they can also be explained by interference by the distracter task (counting backwards in threes).

However, one observation to support decay is that STM can hold fewer words when the words are long (harpoon or cyclone) than if words are short (bishop & pewter). Note that we are referring here to how long the word takes to say (try it). According to Baddeley et al (1975) this effect occurs because as the words are presented, they are encoded acoustically. This takes longer for the long sounding words resulting in an increased possibility that the trace will have decayed for some of the words.

Displacement theory

Displacement occurs when competing information knocks out existing memories. This is similar to the concept of interference (see below), but the term displacement is usually preferred when talking about STM forgetting. Displacement fits in well with the notion that STM has a limited capacity. You can consider that items retained in STM are in a state of activation. The more that are activated, the less activation there is for any one of them. Seven items may be the maximum, any more than this means that the new item will displace one of the old items. Using the analogy from earlier of memory as a workbench, old items have to be removed to make room for new items that you might want to work on.

A study that is considered to support the idea of displacement in STM is one by Waugh & Norman (1965): the serial probe.

Forgetting in LTM

There are many different explanations of forgetting in LTM, each tending to focus on just one or two particular aspects of this complex phenomenon. Theories divide into two basic types.

• Availability explanations: These theories explain forgetting by proposing either that decay through disuse or interference from other learned material, result in the memory trace becoming unavailable.
• Accessibility explanations: These theories consider that information is never really lost from LTM. Unlike STM there is no practical capacity limitation for LTM and so no need for new information to displace the old. The basic cause of forgetting is not a problem of availability but finding the right cues needed to recall the memory. This explanation is usually known as cue-dependent forgetting although the term retrieval failure is also used. Freud first proposed another type of accessibility explanation: the idea of repression.

Decay versus interference in LTM

As was the case with decay and displacement in STM, it is very difficult to establish whether forgetting is due to the passage of time or interference. Jenkins & Dallenbach’s (1924) study suggested that interference is more significant than decay, since forgetting after a period of time spent sleeping was less than that following an equal period of time awake. However, their results can be explained in a number of other ways. For example, in the sleep condition participants learned the words in the evening and it is possible that the mechanisms that consolidate learning work better at night (see Baddeley, 1999). In fact participants who learn a list of words in the morning and then sleep, recalling the list later in the day, show as much forgetting as those that that remain awake.

A study by Baddeley & Hitch (1977) into recency effects in LTM could be more helpful in deciding between decay and interference. They asked rugby players to recall the names of teams recently played. Injuries and suspensions meant that most players they interviewed had missed some games, so for one player the last game might have been last week, while for another it was two months ago. They were able to show that the probability of correct recall was not dependent on the passage of time, as decay theory would predict, but on the number of intervening games. Recall for the last games was equally good whether that game played some time ago or last week.

Different types of interference in LTM

The study by Baddeley & Hitch suggests that interference may be a more important factor in forgetting from LTM. Interference theory suggests that there are two ways in which existing learning can interact to cause forgetting: retroactive and proactive interference. The process of interference has been extensively studied in the laboratory using simple tasks involving mainly verbal materials (paired associate lists). However, while such studies are often well controlled and give fairly clear results, they have been criticised for lacking ecological validity.

It has been suggested that interference can account for certain aspects of forgetting by eye-witnesses. Thus post-event questioning can interfere with their recollection of an event.

Retrieval failure: cue-dependent forgetting

According to Tulving, what matters in explaining forgetting is not whether a piece of information has been lost from memory completely (through decay, for example), but rather whether or not it can be found. Tulving suggested that success or failure at recall is determined by the retrieval cues that are present at the time of recall. The better, or more numerous the retrieval cues, the better the recall. Retrieval cues can be internal (demonstrated in the case of state-dependent forgetting), or external (context-dependent forgetting).

The retrieval cue hypothesis is a very plausible account of certain aspects of forgetting. According to the proponents of retrieval failure theory, it is unnecessary to use the idea of trace decay to explain the forgetting of information even over long periods of time. A change in retrieval cues can disrupt forgetting, and, as cues alter over time, memories will gradually become more and more inaccessible. For example, physical cues that are associated with our childhood years’ disappear, as houses, and even whole neighbourhoods, are pulled down.

The retrieval cue hypothesis has also been used to explain the phenomenon known as childhood amnesia, whereby we appear to have few reliable memories before the age of three and a half years. The environment of the infant and small child is very different from the world they will occupy some 10 to 15 years later (in viewpoint, scale, accessibility, etc.). Thus, whatever memories are stored at this time are encoded in a context that will not be available as retrieval cues in the adult environment.

Activity: Investigating childhood amnesia

Motivated forgetting: repression

Freud emphasised the emotional basis of forgetting and through the concept of repression suggests a means by which threatening material can be kept from consciousness — motivated forgetting. This account of forgetting is highly controversial. While there is general agreement that traumatic experiences have the potential to disturb people’s memories, there are many concerns over the implications of the theory of repressed memories. One area has been the suggestion that repressed memories from childhood can be recovered using appropriate therapeutic techniques. Freud believed that the hypnotic state produced regression. Under hypnosis the conscious control of behaviour is suspended and it becomes possible to return to childish modes of behaviour. Although Freud abandoned hypnosis as a method of gaining access to the unconscious, it is still used today to help people discuss memories whose inaccessibility is hindering therapeutic progress.

A number of attempts have been made to test the concept of repression in the laboratory (e.g. Levinger & Clark’s, 1961). However for ethical and other reasons this has proved difficult. In contrast, there has been no shortage of claims from clinical studies for the existence of recovered memories. For example, Herman & Schatzow (1987) claim to have found that 28% of female victims of incest reported major memory deficits from childhood and these repressed memories were more common among women who had suffered violent abuse. In another study, Williams (1992) indicated that in a group of abused African-American women, 38% reported repressed memories of that abuse. The abuse had taken place approximately 17 years previously.

However, an obvious problem with establishing the accuracy of repressed memories is that there is almost always no concrete evidence to support the claims. Sometimes evidence is discovered, for example there may be photographs or videos or it might be possible that close relatives may confirm memories not related to abuse. Loftus (1993) discussed the case of a 27-year-old man who retrieved repressed memories of seeing his mother trying to hang herself. The man’s father was able to confirm these memories by saying that his son witnessed this event when he was just three years old.

But there is also evidence that memories of past traumatic events can be highly inaccurate. Pynoos & Nader (1989) assessed children’s accounts of a sniper attack in the school playground, they found that misreporting of events was evident.

The fact that there are serious doubts over the accuracy of recovered memories raises a number of ethical issues. Psychotherapists have been accused of implanting false memories in the minds of patients. They have been accused of misleading vulnerable clients and of wreaking the lives of innocent parents.

Many psychologists accept that false memories could exist, pointing out the constructive nature of memory. Memories could be produced by combining actual memories with suggestions from therapists. Once established, such memories would be as real to the person as any other memory. As Loftus (1977) points out:

“We do not yet have the tools for reliably distinguishing the signal of true repressed memories from the noise of false ones…Psychotherapists, counsellors, social service agencies and law enforcement personnel would be wise to be careful how they probe for horrors on the other side of some presumed amnesic barrier. They need to be circumspect regarding uncorroborated repressed memories that return”.

Flashbulb memories

Flashbulb memories (FMs) are memories that occur in connection with highly emotional or otherwise significant events occur. A recent example would be hearing the news of the 7/11 attack on the World Trade Towers.

FMs are significantly different from ordinary memories in the amount of detail that they hold. Brown & Kulik (1977) suggest the FMs contain the following information:

• Who provided the information
• The place where the news was heard
• What was being done when the information was heard
• The person’s emotional state when hearing the information
• The emotional state of others
• The consequences of the event for others.

According to Brown & Kulik, FMs are very different from ordinary memories not just in terms of the amount of detail that they hold but also in terms of method of storage. They suggest that they rely on a special neural mechanism triggered by the experience of emotionally charged events. Cahil et al, (1994) present evidence that the stress hormones adrenaline and nor-adrenaline (see Unit 2.1) are involved in the storage of emotional memories but not everyday memories.

However, the idea of FMs is controversial. While it seems reasonable to suggest that it is the highly charged emotional content of the memories that gives them their vivid and seemingly accurate character, research does not actually suggest that emotion and memory recall are linked in any significant way. While there is a slight advantage for emotionally charged memories (particularly pleasant ones) over one or two years, the effect disappears after two years. So-called flashbulb memories turn out to be no more accurate than any other memory. A study that investigated the accuracy of FMs was Neisser & Harsch (1992): Memories of the Challenger space-shuttle disaster.

A conclusion of the study was that so-called ‘flash-bulb memories are no more accurate than other memories. The results suggest that what is different is the confidence that people have in their memories associated with significant events.

Weaver (1993) tested this idea by comparing his students’ memories of a very ordinary event (the circumstances when they next saw their roommate) with a dramatic event (the start of the Gulf War (‘Desert Storm’). Both events took place on 16th Jan. 1991. Followed up one year later, accuracy for the events was little different but confidence in accuracy was significantly higher for the dramatic flashbulb event.

Another area of research into the effect of emotional factors in memory is post-traumatic stress disorder (PTSD). Patients can sometimes suffer amnesia is often a consequence of being exposed to extreme psychological or physical trauma (e.g. in combat).

References

Back to AS Psychology FrontPage

Top