Theories Of Amnesia

At one time, most theorists tried to apply pre-existing theories of normal memory functioning to amnesics. For example, the evidence of Baddeley and Warrington (1970) and others seemed at one time to provide strong support for the multi-store approach discussed in Chapter 6. Cermak (1979) tried to apply the levels-of-processing approach to amnesia. He argued that amnesics typically fail to process the meaning of to-be-remembered information, and this lack of semantic processing causes the severely impaired long-term memory found in amnesic patients. This theory has been abandoned, because there is strong evidence that amnesic patients are well able to process meaning.

More recent theorists have considered the pattern of deficits shown by amnesic patients, and have then constructed new theories to fit that pattern. Some of these theories have been modified in the light of additional testing on normal individuals. Thus, theorists are increasingly inclined to use memory data from both amnesic patients and normal individuals in the construction and development of their theories.

The assumption that there is a single, unified long-term memory system has been rejected by all the theorists we will be considering. Most theorists have argued that there are at least two major types of processing associated with long-term memory. Other theorists have focused on memory systems, and have tried to identify the underlying brain systems involved. Many of the theories overlap each other. This fact, coupled with the imprecision of many of the theoretical approaches, means it is hard to decide which theoretical approaches are more promising than others.

Episodic versus semantic memory

As we saw earlier, Tulving (1972) drew a distinction between episodic memory, which is concerned with events or episodes happening at a given time in a given place, and semantic memory, which is concerned with general knowledge about the world. On the face of it, it seems reasonable to argue that amnesics have a severe deficit in episodic memory but essentially intact semantic memory. Amnesic patients have impaired episodic memory, as this description by Korsakoff (1889) of a typical amnesic patient reveals:

He does not remember whether he had his dinner, whether he was out of bed. On occasion the patient forgets what happened to him just an instant ago: you came in, conversed with him, and stepped out for one minute; then you come in again and the patient has absolutely no recollection that you had already been with him.

There is also little doubt that major parts of semantic memory are generally intact in amnesics. The most obvious examples of this are their largely unimpaired language skills, including vocabulary and grammar, and their essentially normal performance on intelligence tests.

In fact, there is a serious flaw in this argument, namely that like is not being compared with like. Language and the abilities required to perform well on intelligence tests are nearly always acquired before the onset of amnesia, whereas conventional tests of episodic memory are based on information acquired after the onset of amnesia. Thus, the findings described so far are consistent with the simple notion that amnesia mainly impairs the ability to acquire new episodic and semantic memories. Evidence that it can be very hard to establish new semantic memories after the onset of amnesia was reported by Gabrieli, Cohen, and Corkin (1988), who found an amnesic patient with an almost complete inability to acquire new vocabulary. In similar fashion, many amnesics do not know the name of the current prime minister or president, and have very poor recognition memory for the faces of people who have become famous fairly recently (Baddeley, 1984). It thus appears that most amnesics are impaired in acquiring new semantic memories as well as new episodic memories.

According to Wheeler et al. (1997), there is an important distinction between autonoetic or self-knowing awareness (found in episodic memory) and noetic or knowing awareness (found in semantic memory). The relevance of this distinction to amnesia was studied by Knowlton and Squire (1995). Amnesics and normal controls were given a test of recognition memory, and asked to divide recognised items into "remember" responses based on conscious recollection and "know" responses based on familiarity only. The amnesic patients performed much worse than the controls on both "remember" and "know" items, suggesting that the memory deficit in amnesia is not limited to one level of awareness.

Some recent evidence suggests that the distinction between episodic and semantic memory may have relevance to amnesia. Vargha-Khadem et al. (1997) studied two patients who had suffered bilateral hippocampal damage at an early age before they had had the opportunity to develop semantic memories. Beth suffered brain damage at birth, and Jon did so at the age of 4. Both these patients had very poor episodic memory for the day's activities, television programmes, telephone conversations, and so on. In spite of this, Beth and Jon both attended ordinary schools, and their levels of speech and language development, literacy, and factual knowledge (e.g., vocabulary) were within the normal range.

How can we explain the ability of Beth and Jon to develop fairly normal semantic memory in spite of their grossly deficient episodic memory? According to Vargha-Khadem et al. (1997, p. 376), episodic and semantic memory depend on somewhat different regions of the brain: "Episodic memory depends primarily on the hippocampal component of the larger system [i.e., hippocampus and underlying entorhinal, perihinal, and parahippocampal cortices], whereas semantic memory depends primarily on the underlying cortices."

Why do so many amnesics have great problems with episodic and semantic memory? According to Vargha-Khadem et al. (1997), many amnesics (including HM) have damage to the hippocampus and to the underlying cortices.


Most of the evidence has failed to indicate that the distinction between episodic and semantic memory is of fundamental importance to an understanding of amnesia. The fact that most amnesics have great difficulty in forming new semantic memories poses real problems for this theoretical approach. However, it is possible that partially separate brain systems underlie episodic and semantic memory, but both brain systems are typically damaged in amnesics. The findings reported by Vargha-Kardem et al. (1997) are consistent with this possibility, which should certainly be examined systematically in future research.

Context processing deficit theory

Long-term memory is generally better when the context at the time of the memory test is the same as that at the time of learning than when it differs (see Chapter 6). Contextual information is also important in allowing us to distinguish between otherwise similar memories Mayes (e.g., 1988) argued that amnesic patients can store information about to-be-remembered information, but find it hard to store and retrieve contextual information. This hypothesis is known as the context processing deficit theory As contextual information about time and place is found with episodic but not with semantic memories, this theory overlaps theories emphasising a deficit in episodic memory in amnesic patients.

Powerful findings related to context processing deficit theory were reported by Huppert and Piercy (1976). They presented a series of pictures on day 1 of their study, and a series of pictures on day 2. Some of the pictures presented on day 2 had been presented on day 1, and some had not. Ten minutes after the day 2 presentation, there was a test of recognition memory. On this test, participants were asked which pictures had been presented on day 2. The normal controls had no problem (see Figure 7.8a). They correctly identified nearly all the pictures that had been presented on day 2, and incorrectly identified very few of the pictures presented only on day 1. Korsakoff patients did much worse, correctly identifying only 70% of the day 2 pictures, and incorrectly identifying 51% of the pictures presented only on day 1.

Huppert and Piercy (1978) found that the recognition-memory ability shown by the amnesic patients was entirely due to the fact that the day 2 pictures were slightly higher in familiarity than the day 1 pictures, rather than to specific memory for the time of learning. Thus, Korsakoff patients showed practically no direct memory for temporal context, i.e., the day on which they had seen any given picture.

The most important finding obtained by Huppert and Piercy (1976) arose when they asked their participants to indicate whether they had ever seen the pictures before. With this test, it was not necessary to have stored contextual information about when the pictures had been seen in order to show recognition memory. The Korsakoff patients and the normal controls performed this task at a very high level, with the two groups hardly differing in their performance (see Figure 7.8b). Thus, information about the pictures themselves was stored in long-term memory by the Korsakoff patients, but very little (if any) information about the circumstances in which the pictures had been seen previously was available.

Context processing deficit theory has also received support from research on source amnesia, in which facts are remembered but not the source of those facts. Source amnesia in amnesic patients was studied by Shimamura and Squire (1987). Amnesic patients were more impaired than normal controls in remembering the source of trivia facts they were able to recall. Thus, amnesic patients have particular problems in remembering contextual information associated with their learning of trivia facts.

Increasing Your IQ

Increasing Your IQ

How Would You Like To Amaze People With Your Intelligence?  Increase your IQ and get prepared to receive accolades in every sphere of life. Do you feel dejected every time your boss praises a colleague for an intelligent professional move? Do you want to become a crucial resource to your company?

Get My Free Ebook

Post a comment