Organic Amnesia

The most intensively studied of the organic memory disorders is the global organic amnesia syndrome, which was first characterized in the 19th century. In this syndrome, there is impaired recall and recognition of facts and episodes encountered both postmorbidly (anterograde amnesia) and (more variably) premor-bidly (retrograde amnesia). These deficits often occur even when intelligence and shortterm memory are preserved. There is also preservation of overlearned semantic memories, of various forms of motor, perceptual, and cognitive skill learning and memory, and of at least some kinds of unaware memory for specific information (priming). The deficits can be produced by lesions to structures in the MTL, midline diencephalon, basal forebrain, and possibly ventro-medial frontal cortex as well as to structures that link these regions, such as the fornix.

There is growing evidence that the syndrome is functionally heterogeneous. First, it has been claimed that retrograde amnesia can be produced relatively independently of anterograde amnesia. Several studies have found poor correlations in patients between severity of anterograde amnesia and deficits in more remote premorbid memory. There have also been several reports of relatively selective or focal retrograde amnesia in which patients had a severe and enduring deficit in premorbid memory but a relatively preserved ability to acquire new memories (so that old memories could be "relearned" to some extent). The location of lesions that cause this condition remains to be accurately resolved, but damage often includes the anterolateral temporal neocortex, particularly on the left. There have also been reports ofpatients with focal retrograde amnesia who had damage to their tempor-opolar and frontal cortices, mainly on the left. These patients showed extensive retrograde amnesia, but their new learning abilities were relatively intact. The results suggest that old memories about autobiographical and public information may be disrupted by lesions to parts of the temporal cortex. Damage to prefrontal cortex can also disturb these memories, but it is unproved that such lesions alone can disrupt premorbid memory relatively selectively. It needs to be determined more precisely which temporal (and possibly parietal) lesions disrupt remote premorbid memories and how these relate to the damage underlying semantic dementia. There is also a particular difficulty with these focal retrograde amnesia patients in proving that they are not malingering or suffering from a psychogenic deficit. In such cases, the premorbid memory deficit is not directly caused by brain damage but may reflect the patient's conscious wish to achieve financial gain or avoid something unpleasant or the patient's unconscious avoidance of something traumatic. These cases can appear very similar to those arising from brain damage.

Second, it has been argued that amnesia is heterogeneous because lesions to different MTL structures impair memory in distinct ways. It is generally agreed that amygdala damage does not usually cause ante-rograde amnesia, in contrast to other MTL lesions. However, the amygdala probably plays a role in emotional memory by modulating the effectiveness with which other MTL structures store memories when emotionally evocative information is encoded. This role of the amygdala would explain why events associated with marked emotional arousal are so well remembered (flashbulb memory). The effects of amygdala lesions on memory have been very little explored in humans because of the rarity of selective amygdala lesions.

Some researchers believe that lesions of either hippocampus or perirhinal cortex within the MTL disrupt memory in the same way, but that hippocam-pal lesions may have a lesser effect. This view implies that free recall and item recognition are equivalently disrupted after MTL lesions. In contrast, it has been argued that hippocampal and perirhinal cortex lesions cause dissociable deficits. This view is based on animal and human evidence. First, animals with perirhinal lesions have shown item recognition impairments but intact spatial memory, whereas the reverse effect has been found with hippocampal lesions. Second, meta-analysis of human recognition data has suggested that there is a single dissociation in which patients with damage to the hippocampus or other parts of Papez circuit, such as the fornix, mammillary bodies, or anterior thalamus, are relatively unimpaired on item recognition but as impaired as more generally lesioned global amnesics (who are also impaired at item recognition) on tests of free recall.

Several patients with apparently selective hippo-campal damage were found to be clearly impaired on item recognition when an extensive battery of tests was given. However, patients with selective hippocampal sclerosis related to temporal lobe epilepsy were reported not to show impaired item recognition, and other patients with probable hippocampal damage caused by hypoxia showed a similar preservation. In addition, three patients with evidence of early selective hippocampal damage were shown to have completely selective free recall deficits because their item recognition was normal on a range of tests. Similar patterns of memory performance with relatively preserved item recognition have been found in patients with selective hippocampal damage acquired in adulthood, so it is unlikely that the pattern of selective free recall deficits reflects reorganizational processes following early brain damage.

Apparently selective hippocampal damage, sometimes causes severe item recognition deficits and sometimes leaves item recognition relatively or completely intact. Animal studies have shown that cerebral ischemia often produces damage that extends beyond the hippocampus. Because such damage may be both highly variable and difficult to identify either by structural imaging or by postmortem analysis, it is likely that those patients with more severe recognition deficits have damage extending into other brain regions important for item recognition, such as the perirhinal cortex. There is even evidence that extra-hippocampal damage is exacerbated by abnormal processes in the hippocampus that are triggered by ischemia. Extrahippocampal damage is probably best detected by measuring whether blood flow is abnormal in nonhippocampal brain regions so that it can be determined whether such abnormalities are more striking in patients with severe item recognition deficits. This kind of abnormality, invisible to structural magnetic resonance imaging, has been identified using positron emission tomography in patients who suffered hypoxia following heart attacks. These patients showed reduced blood flow in regions that appeared normal in structural magnetic resonance imaging (MRI). It is possible, however, that less used and more sophisticated structural MRI procedures would identify abnormalities in these regions.

Selective damage to other parts of the Papez circuit has also been found to cause little or no disruption of item recognition. Thus, relatively selective free recall deficits have been reported after fornix lesions caused by colloid cyst surgery and also following relatively selective mammillary body lesions. Lesions that affect only the anterior thalamic nucleus within the thalamus are very rare, but one patient has been described whose thalamic damage was confined to the left anterior thalamic nucleus, although she also had damage to the head of the left caudate nucleus and the left fornix. This patient had completely normal recognition for items and a free recall deficit for verbal materials.

Some animal research supports the existence of another memory system that comprises the perirhinal cortex, the dorsomedial nucleus of the thalamus to which it is reciprocally connected, and possibly the orbitofrontal cortex to which the thalamic nucleus links. The interconnections of this system with the Papez circuit system are illustratred in Fig. 1. As indicated previously, in animals, selective perirhinal cortex lesions disrupt item recognition but not some kinds of spatial memory. They also disrupt associations between similar kinds of items. Little relevant work has been done in humans mainly because selective perirhinal cortex lesions are probably nonexistent and selective dorsomedial thalamic lesions are extremely rare.

However, total bilateral destruction of the perirhin-al cortex (and other regions) in humans has been reported to show more severe impairments of recognition for complex patterns at delays of a few seconds compared to those of global amnesics without peri-rhinal cortex lesions. Interestingly, these patients performed normally at delays of 0 and 2 sec as well as when making judgments regarding when the stimulus was still present. This could mean that perirhinal cortex lesions selectively disrupt long-term memory because they leave short-term memory and visual perception intact. This interpretation is not undisputed because it has been argued, on the basis of animal research, that perirhinal cortex lesions disrupt the ability to represent complex conjunctions of stimulus features in perception, and that memory deficits may be secondary to these high-level perceptual deficits. If this argument is correct, then amnesia that is caused by perirhinal cortex damage does not constitute a pure memory deficit. In principle, the argument can also be extended to hippocampal lesions that might be causing disruption of the ability to represent other kinds of complex conjunctions (such as those between objects and locations or between faces and voices). Such high-level perceptual deficits would secondarily cause memory impairment, and they

Figure 1 The connections between the structures that may constitute the two memory systems as well as interconnections between the systems and between the systems and neocortical regions. Shading with lines indicates structures that may form the perirhinal cortex-dorsomedial thalamic memory system, whereas pale gray shading indicates structures that may form part of the Papez circuit memory system. Arrows indicate that there is evidence that connections exist. Thick solid lines indicate connections that can confidently be related to the memory systems. Thin solid lines indicate connections that are known to link the two memory systems. Both systems connect to prefrontal association cortex, but it is uncertain what role this cortex plays in memory and it is also unclear whether the two memory systems interact in the prefrontal association cortex.

Figure 1 The connections between the structures that may constitute the two memory systems as well as interconnections between the systems and between the systems and neocortical regions. Shading with lines indicates structures that may form the perirhinal cortex-dorsomedial thalamic memory system, whereas pale gray shading indicates structures that may form part of the Papez circuit memory system. Arrows indicate that there is evidence that connections exist. Thick solid lines indicate connections that can confidently be related to the memory systems. Thin solid lines indicate connections that are known to link the two memory systems. Both systems connect to prefrontal association cortex, but it is uncertain what role this cortex plays in memory and it is also unclear whether the two memory systems interact in the prefrontal association cortex.

would be very difficult to detect because of the likelihood that affected patients would use compensatory strategies. It cannot therefore be confidently claimed that major forms of amnesia are unrelated to very specific forms of encoding failure.

The very severe long-term memory deficits that have been reported following perirhinal cortex lesions may be dependent on this damage being nearly total. Support for this possibility is provided by a subgroup of epileptic patients with selective damage that includes only part of the perirhinal cortex who show normal performance on standard tests of anterograde amnesia but accelerated forgetting over delays of weeks. By these delays, their memory is badly impaired. Partial perirhinal cortex damage may therefore cause a much less severe and much delayed item recognition deficit than does total damage to this cortex. The mnemonic effects ofdorsomedial thalamic lesions may also critically depend on the extent of damage. Thus, patients with marked destruction of this nucleus show impaired item recognition as well as free recall, whereas patients with only a small portion of the nucleus damaged show little evidence of memory deficits.

Papez circuit lesions and perirhinal cortex system lesions both cause retrograde as well as anterograde amnesia. Although severe retrograde amnesia can extend back to memories acquired decades before the causative brain damage, there often appears to be a temporal gradient in which earlier acquired memories are less disrupted. Considerable evidence suggests that Papez circuit lesions cause a less severe, more steeply temporally graded retrograde amnesia than do larger lesions that involve the whole of the MTL. This would be consistent with the hypothesis that the hippocampus (and perhaps other MTL regions) only stores facts and episodes for a limited time until reorganization results in neocortical storage in the sites that presumably represent the stored information. An alternative view that challenges this claim is that there is no reorganization of the location of long-term memories as a result of processes such as rehearsal, at least for some kinds of information (such as spatial). The simplest interpretation of this view is that it should predict no temporal gradient, although more complex interpretations are also possible that make it more difficult to distinguish between the predictions of the no change and the change views of what happens to long-term memory as it ages. Therefore, additional work is needed to determine whether memories cease to depend on the MTL as they age. Future work must also explore whether selective Papez circuit lesions minimally disrupt premorbid item recognition memory.

Exactly what processes are disrupted in patients with the amnesia syndrome? If the syndrome is heterogeneous, there must be several such processes. It is widely believed that many global amnesia patients process facts and episodes normally because their intelligence is preserved and normal kinds of information are available to them at input when memory load is minimized. As discussed previously, however, it has not been shown conclusively that subtle high-level perceptual processes may not be disrupted following MTL lesions. Evidence from transient global amnesia (TGA) indicates that it is unlikely that retrieval is impaired in global anterograde amnesia. TGA is a form of global amnesia that lasts for only a few hours and is usually caused by a reversible abnormality in the MTL, which has been revealed by showing that blood flow to the MTL (and sometimes other brain regions) is reduced during a TGA attack but usually returns to normal when tested later. Upon recovery, although all premorbid memories apart from those acquired in the few minutes or hours prior to the incident typically return, as does the ability to lay down new memories, no memories for the incident return. Because these memories do not return even when retrieval must be normal, global anterograde amnesia is probably caused by a failure to consolidate facts and episodes into long-term memory in the minutes following input.

If Papez circuit lesions do cause a selective amnesia, they will disrupt the consolidation of different kinds of information from those disrupted by perirhinal cortex lesions. The available evidence suggests that patients with selective hippocampal lesions can be relatively normal not only for recognition memory for single items (e.g., words or faces) but also for associations between the same kind of items (e.g., word-word or face-face associations). However, they are severely impaired at recognizing associations between components that would probably be processed in different cortical regions (e.g., object-location and face-voice). Similar and possibly more severe spatial memory deficits have also been reported following parahippo-campal cortex lesions. The evidence is more conflicting about whether patients with selective hippocampal lesions show relatively preserved remote memory for overlearned facts. Patients are unquestionably impaired at the initial acquisition of new facts (such as vocabulary) but it remains to be shown that their impairment does not occur because their factual memory cannot be facilitated by the retrieval of associated contextual information (which involves episodic memory and is disrupted). Also, some patients may learn to compensate for their learning deficit by rehearsing factual information much more frequently than people with normal memory. Probably, hippocampal and Papez circuit lesions disrupt consolidation of both factual and episodic associations, the components of which are represented in different cortical regions.

Although lesions that include the perirhinal cortex drastically impair item recognition, little is known about the effect of selective lesions of this structure in humans. Animal studies indicate that selective damage to this cortex disrupts recognition of items and associations between similar components but not memory for some kinds of spatial information. It is uncertain in humans whether perirhinal cortex lesions disrupt memory for all the kinds of information affected by hippocampal lesions and memory for some other kinds of information as well, or whether there are some kinds of memory (such as some forms of spatial memory) that are only disrupted by hippocampal lesions.

Was this article helpful?

0 0
All About Alzheimers

All About Alzheimers

The comprehensive new ebook All About Alzheimers puts everything into perspective. Youll gain insight and awareness into the disease. Learn how to maintain the patients emotional health. Discover tactics you can use to deal with constant life changes. Find out how counselors can help, and when they should intervene. Learn safety precautions that can protect you, your family and your loved one. All About Alzheimers will truly empower you.

Get My Free Ebook

Post a comment