Long term memory and the amnesic syndrome Long term memory A wide variety of things is stored in long term memory (LTM). So it seems sensible to suggest that LTM is split up in some way. There are differing points of view on how far this can validly be taken. One distinction that is often made is between explicit and implicit memory (e.g. Graf and Schacter, 1985). Explicit memory is defined as requiring conscious recollection of previous experiences. Implicit memory is when something can be learned even without conscious recollection of specific experiences. Schacter and Tulving (1994) suggested that there were four major long term memory systems (in addition to working memory, which develops the idea of short term memory). Episodic memory: memory of specific events Semantic memory: a stock of knowledge about the world; “a mental thesaurus, organised knowledge” (Tulving, 1972, p. 386) Procedural memory: skills, including motor (riding a bike) and cognitive (e.g. how to read) The perceptual representation system: “a collection of domain-specific modules that operate on perceptual information about the form and structure of words and objects” (Schacter et al., 2000, pp. 635-636). Associated with priming. Episodic and semantic memory are often grouped together as declarative memory, which is often said to involve explicit memory. Procedural memory and the perceptual representation system are sometimes said to involve implicit memory. There is evidence that there are different kinds of LTM, but disagreement over how many kinds and what they are. Some important evidence that there are different kinds of LTM comes from the amnesic syndrome. The amnesic syndrome In its broadest definition, amnesia refers to any kind of memory impairment. In its narrower sense, it refers to what is more precisely called the amnesic syndrome. The amnesic syndrome is usually associated with damage to particular parts of the brain (the diencephalon; or the medial temporal lobe, especially the hippocampus). Common causes of this damage are stroke, closed head injury and alcohol abuse. In Korsakoff’s syndrome, amnesia is caused by alcohol abuse (together with inadequate diet). This syndrome usually has a gradual onset, and has other effects besides amnesia, which complicates interpretation of its effects. However, a lot of work has been done with Korsakoff‟s patients because it was a relatively common syndrome at the time that the study of amnesia was at its early stages. The study of amnesics helps us form theories of memory, and test such theories. The amnesic syndrome is particularly important for at least two reasons. As the label „syndrome‟ implies, they are a group of patients with similar symptoms. Secondly, it is obviously organic – that is to say it is associated with failure of particular parts of the brain, which suggests it can be related to particular mental modules. There are some famous cases of amnesia. HM has amnesia as a result of brain surgery which went wrong. NA‟s brain was damaged in a freak accident. Jimmy (described in Oliver Sack‟s book The Man who Mistook his Wife for a Hat) had Korsakoff‟s syndrome. All of these cases are classic because they all have very dense amnesia – that is to say, they have little or no use of the kinds of memory that are affected by the amnesic syndrome. Amnesics vary in how dense their symptoms are. In considering the effects of amnesia it is helpful to distinguish between anterograde and retrograde amnesia. Anterograde (forward-acting) amnesia is an impairment to the ability to remember new information, and is found in all cases of the amnesic syndrome (at least to some extent, and for some kinds of memory). Retrograde (backward-acting) amnesia is an impairment in the ability to remember old information (i.e. information which was known before the inset of amnesia). It is harder to study retrograde amnesia, but we know it exists in some amnesics and not others. Where it does exist, it seems to be greater for events which occurred just before the onset of the amnesia. One of the important findings about the amnesic syndrome is that it has different effects on different kinds of memory (e.g. Speirs et al., 2001). Short term memory (in the sense we have defined it) is not affected. Episodic memory is always affected. Semantic memory is often affected. Broadly speaking, procedural memory and the perceptual representation system are not affected. HM has been studied in great detail. His digit span shows that his short term memory is in the normal range (Wickelgren, 1968) and he enjoys doing crossword puzzles. He has little episodic or semantic memory for information subsequent to the operation (Milner, 1965). However, he has learnt new procedural tasks such as mirror-drawing (Milner, 1965) and rotary-pursuit (Corkin, 1968); and a task involving the perceptual representation system, the incomplete-pictures task (Gollin, 1960). Readings. A good treatment of long term memory is in Eysenck & Keane (2005, chapter 7). Amesia is dealt with in Esyenck and Keane and many general psychology textbooks. A specialist book on memory is Baddeley (1997). Oliver Sacks‟ description of Jimmie is chapter 2 of his book The Man Who Mistook his Wife for a Hat (1985). There is a good writeup of HM in Pinel (2003): Biopsychology (5th ed.). Mike Griffiths, Jan 2009 Implicit and explicit memory: Jacoby, Toth and Yonelinas (1993). Part 1. Participants read a list of words. E.g. „mercy‟. Full attention: Participants were told to read out the word and remember it (with the intention of creating an explicit memory) Divided attention: Participants read out the word while doing another task (with the intention that the participant would be too distracted to create an explicit memory). Part 2. Word completion. Participants were given the start of words they had read (e.g. M E R _ _). They had one of the following tasks: (a) Inclusion: Complete the word forming a word from part 1 (or failing that, any word you can think of) (b) Exclusion. Complete the word forming any word you can think of, except one from part 1. All of the words had several possible answers (e.g. mercy, merit, merge, merry). If participants had not done part 1, the average chance of using a word from part 1 was 30%. The results were as follows. 70% Chance of using a target word 60% 50% 40% Exclusion 30% Inclusion 20% 10% 0% Full Divided In the inclusion task Participants who had read the words in full attention gave them 61% of the time Participants who had read the words in divided attention gave them 46% of the time. The interpretation is that they had an implicit memory, even though they were too distracted to learn the words explicitly. How do we know that participants in divided attention did not remember some of the words explicitly? Because even in the exclusion task (when they were avoiding words they remembered) they still used a similar number of words.