An answer to this question should propose: the consolidation as a system for laying down new permanent memories; the long-term-memory store as a store of permanent memories. Permanent should be defined here as memory that may be retrieved after a period of interference or distraction of around 30 seconds to a minute. Short-term-memory (STM) is a temporary memory store that includes a part which is within consciousness (focus of attention) and a part which is not being attended to and will decay unless refreshed or rehearsed which would require attention. This brief and temporary story is vulnerable to interference and distraction. While throughout this book working memory has been broadly referred to as what can be "kept in mind", a more scientific description would be required here. It has two components: short-term-memory and an executive system. There is the assumption that such a working memory is actually "working" and is required within a process of mental manipulation (this is a slighty more demanding definition than the one sometimes used in the non-human literature). Later in this chapter these concepts will be described further.

At this stage of the chapter we can broadly say that HM was able to repeat back information and follow a conversation and so his STM was largely intact. However, the bilateral medial temporal lobe lesions meant that his consolidation system was no longer functioning. HM could therefore not make new permanent memories for the LTMS. Information already in LTMS at the time of the neurosurgery was mostly intact but because the medial temporal lobe is also important for retrieving recent memories, his most recently stored memories before the operation were also undermined.

As per text description but best to construct this form of memory if you are intending to take an exam.

Thiamine deficiency is associated with destruction of parts of the extended memory system; most notably the anterior thalamus. Alcohol toxicity is more noted for atrophy to parts of the central nervous system and in particular there is a noted damage to the frontal lobe and the cerebellum. It can be seen that thiamine deficiency is most noted as causing damage to the consolidation system while alcohol toxicity may undermine motor coordination (cerebellum) and the ability of patients to organise (prefrontal cortex) their own lives. There are of course patients with both WKS and signs of ARD. However, and this is controversial, there a rare group of patients who appear to have a relatively isolated WKS (the subject of my PhD) and this is seen by some as an indication that these individuals for reasons unknown have a particular genetic vunerability to WKS and as soon as they become hospitalised with this profound amnesia they stop drinking making the diagnosis of ARD less likely. Following a similar argument concerning the unusual individual vulnerability to WKS, the author has assessed a number of patients who show deterioration associated with their chronic alcoholism that deserve a diagnosis of ARD, but who have no signs of amnesia.

It is a common theme in this text that when a structure is damaged signs of dysfunction are found that relate to the presumed role of the structure but also to the main structures that they are connected with within a network. The anterior thalamus has connections with the prefrontal cortex. These patients' lack of insight in their amnesic state, their tendency to make false recognitions on memory tests, their poor source memory i.e. failure to remember the context of a memory and their tendency to confabulate are all features that may encourage a view that these patients have a frontal flavour to their memory performance.

The deterioration of a store is associated with the tendency to recognise an object or living thing by a more common representative e.g. a zebra might be called a horse. This is referred to as typicality. Alternatively it is assumed that a retieval difficulty may be compensated by a cue upon which the correct item is provided e.g. an animal with stripes that is from Africa. It is assumed that such cues are less helpful to patients with semantic dementia, but useful to persons with left posterior temporal lobe damage who are more likely to suffer from retrieval difficulties, but further evidence is required to support this view.

There is evidence that types of memories e.g. visual versus auditory are stored in cortical areas that are also responsible for processing that modality or function. For example, memory for actions would be processed in areas contributing to the analysis of movement. The polar areas of the anterior temporal lobe have been described as a hub for abstract multimodal memory, but if there is the assumption that such memories are also processed elsewhere within a network then a disconnection between the hub and the areas responsible for processing emotion would undermine the retrieval of emotional semantic memories e.g. people become angry or demeaned if you speak to them in a certain way.

It is has been an argument proposed here and more strongly in the first edition that the ease of storing new memories is dependent on the assimilation of new memories into memories that are already stored. This is in keeping with the Hebbian understanding of cell assemblies. Since already stored memories are assumed to deteriorate due to atrophy in the anterior temporal lobes in cases of semantic dementia, such assimilation and consolidation would be undermined.

The twigs of the tree represent more refined and discriminating concepts of semantic understanding e.g. guinea pigs compared to the more typically encountered cat. The branches that represent knowledge at a more commonly encountered store level Patterson refers to as typical. These are more basic, usual and redundant concepts e.g. cat, dog, horse.

When considering autobiographical memories there is a tendency for us to retrieve associated memories. We visit a stable of horses and recall the time we have fallen off a horse. The memory is then reconsolidated and because of this, it is more likely to be recalled at some point in the future with some unreliability or distortion. However, the context at the time of the retrieval may influence the nature of the memory. If the orginal accident was associated with teenage foolishness of attempting to ride a horse without a saddle and in an attempt to impress we might vaguely remember the incident when visiting a stable years later. If at the time we visit the stable a horse acts in an erratic way when we are attempting to feed it we would more likely reconsolidate the falling off the horse memory as confirming that horses are generally dangerous and we might overlook the foolish aspects of the original incident.

The leading nature of questions that someone may ask us will bias the retrieval of the memory e.g. "Did he severely hurt you when he was holding your shoulders?" Also, the police line-up procedure may have many opportunities for such retrieval errors that may be accentuated unless such procedures are carried out in an educated way. Some insurance against the unreliability of past memories is the immediate rehearsal of the memory shortly after the event, thereby strengthening the memory against future distortion.

Both theories seem to agree that when we reconsolidate memories the hippocampus is involved. But the standard theory believes, irrespective of the type of memory, that if a memory has been consolidated and stored after around three years then it is no longer reliant on the hippocampus for retrieval. We just retrieve it directly from its cortical storage area. With the multitrace theory (MTT) there is an agreement that memories are semanticised and turned into facts and do not need the hippocampus for retrieval, but the MTT position is that if we try and retrieve an event as it actually happened then the hippocampus is required. This autonoetic memory might be a blow-by-blow account of how you reacted to being beaten at school or the time you saw a girl or boyfriend for the last time.

The student should read this section to give a comprehensive reponse to this question and further reading will lend credance to any answer. But broadly, the idea that memory has multiple traces with each trace having a contextual cue based at the time of reconsolidation seems to be useful. But this is not necessarily argued against by the standard theorists. Some caution concerning the idea that all autonoetic memories require the hippocampus comes from the possibility that some recall of what actually happened at an event may be "sematicised". In other words we know what happened at the event, but we do not actually re-live the event when we recall the event. We recall rather the facts of what we think happened during the event and such retrieval does not require the hippocampus. Of course it is not always easy to know the nature of an individual's autobiographical retrieval. Does the question that you have devised for research test autonoetic memories or a semanticised version. Difficulties in deciding this interesting question concern the different and non-standard way questions have been posed when requiring autobiographical retrieval. Have they asked for the actual event or just knowledge of that event?

There seems to be something of a continuation between prospective memory (PM) in which in one extreme there is a specific cue to remember to do something in the future e.g. a salient reminder from an electronic device which is little more than cued recall. To another extreme in which the cue, if there is one, is not obvious e.g. remembering when you wake up that you need to go to the dentist that day without any external reminder. In between these extremes might be a self-cue when you say to yourself: "now what is it that I have to do today?", or you pass some vague contextual cue on a bill board with someone showing the results of a toothepaste.

Time-based PM is when there is no obvious cue to help us to remember to do something in the future. There is the assumption that there is a meta memory component in that there is a stored goal of remembering that oversees our everyday remembering.

The evidence for this proposal are studies that test STM or WM for information that is loosely associated e.g. houses and faces. One can see that holding such information within STM would be beneficial if we learned a connection between the two stimuli. The ease of assimilation and consolidation requires an association to be made within LTMS. Therefore the hippocampus would be required to make such an assimilation easier and as the consolidation takes place this would free up space within our limited capacity STM/WM. Some argument against the dependence of STM/WM on the medial temporal lobe e.g. hippocampus and other consolidation structures would be that this type of task is not a pure test of STM/WM. This may be the case but from a practical point of view it must be realised that patients with amnesia may be fine at holding information that is not obviously beneficial by consolidation process e.g. repeating a list of numbers but as soon as the task requires information where associative chunking is a clear advantage then the patient with a damaged consolidation will be at a disadvantage. The answer could be elaborated upon by further discussing the features of the task that could be defended as a test of STM and an argument that the consolidation system might be required to maintain information within WM and available to consciousness even without changing the permanent status of some of WM.

This is not an easy question and one that requires some speculation. We know that the implicit memory and priming may take place despite a damaged consolidation system. It is proposed by some theorists that the act of perception of a stimulus provides some temporary neural activity that makes a similar perceptual process easier and more likely in the future. Also, it is possible to see LTMS as an end point of perception and such priming may make a transistion from perceptual process e.g. vision, auditory and then activate the representation in LTMS of the stimulus. This would be an economic understanding of brain organisation and be in line with studies of binocular rivalry, for example. Priming would therefore require a representation within the model in Figure 18 that represented this process of perception and activation that was adjoining and activating LTMS. Perhaps positioned to the right of LTMS.

This is a question that is relevant to the perirhinal cortex with its connections to visual and emotional analysis with its role in face recognition, for example. Also the parahippocampal gyrus with its connections to such areas as the retrosplenial cortex with its prescribed role in imagery and the recognition of scenes, and the analysis of spatial location within the parietal cortex that lends support to the role of the parahippocampal cortex in the memorisation of location and scenes. A more comprehensive answer is, of course, possible.

There are some simple uses of mnemonics that can be directed to a specific purpose that does not ncessarily require generalisation. Such techniques may be generalised to the home situation given a supportive carer. However, with the more complex mnemonic technique difficulty sometimes arises. The patient has memory difficulties and so they must learn the technique to compensate but they can't learn the technique because they have memory difficulties (catch 22). This may also be the problem with applying electronic aids such as organisers and smart phones. If it is not easy to learn to use such devices in an intuitive way the patient may have difficulty benefiting from such techniques.

External aids are often simpler and intuitive methods of compensation that are likely to remain in habitual use for a longer period following therapy.

Errorless learning has been developed to overcome the problem of patients remembering their own mistakes. Patients with problems are less likely to remember the context of a mistake; in other words they will just remember the information and not that it was in error. The context of such a mistake might be the finding that they were told the information was an error and where, when and how the error was made. Without this associative contextual information the mistaken response of the information will be strengthened through the response making it more likely that the information will be retrieved in the future.

There is a basic tenet that people are more likely to complete and work hard at rehabilitation if they believe that they have some independence in the process and they have designed their own programme. This kind of ownership encourages understanding of the techniques by the patient. It is virtually impossible to take this approach without an understanding of the patient's problems and the patient ends up feeling that they have been heard — sometimes for the first time. Of course, the patient does not have the knowledge of the rehabilitationist but they can be given choices which allow them to feel empowered. They can also ask and be provided by the skilful clinicians with techniques that they know will fit their everyday circumstances and their future goals. This same kind of empowerment can be used in every kind of educational endeavour.

Continuing this approach it would be possible that you could show this flexibility in designing a programme with various degrees of memory impairment.