Processing Memory in the Brain Glenn Mason-Riseborough (16/5/1997) Introduction “Memory is the property of the entire neocortex of the brain: There is no evidence supporting theories of memory that assign specific memory abilities to specific regions of the brain.” The statement quoted above is an extremely strong view on memory acquisition and storage in the brain, and also on the results of past experimental studies. This argument actually consists of two separate statements with the second intended as a premise for the first. To prove the falsehood of the second statement we need only find one study that shows significant correlation between a specific type of memory and a specific part of the brain. Thus the bulk of this essay will examine research of subjects with specific memory deficits due to lesions in specific areas of the neocortex. The first statement above proposes both the principle of equipotentiality and the principle of mass action ? it is a statement supporting diffusion of memories rather than localisation. These principles are respectively the concept that the storage of memories is conducted equally by all parts of the neocortex, and that this memory is then stored in all parts of the brain. The second statement argues that all experimental research conducted to date shows no significant evidence that specific memory abilities are correlated to specific areas of the brain. This essay will critically evaluate these propositions. It will provide evidence both for and against the first statement and show research on lesions in specific areas of the neocortex that dispute the second statement. Due to ethical considerations, research conducted on memory deficit is from observational studies of patients, or animal experimentation. It must be noted that observational studies show correlation but cannot reliably show causation. Thus by simply observing people with memory deficits, we cannot conclude that lesions in their brain in certain areas caused these deficits, only that the two events may be correlated. Research favouring diffusion of memories Much of the data which supports diffusion of memories was gathered by Karl Lashey over a 35 year period up until the middle of the 20th century (Pinel, 1993). This helped popularise the belief that although some functions such as movement, perception, attention and language were localised, memory was not (Kandel, 1992). Lashey’s research consisted largely of training various animals (such as rats, cats and monkeys) to perform certain tasks. These animals would then have their brains surgically lesioned in specific areas and then reassessed, performing the tasks they had previously learned. Lashey’s findings showed that no particular site of lesioning caused memory failure, similar size lesions in different areas caused similar results. The data also showed that only sufficiently large lesions in any area had an affect. From this data Lashey concluded that while some parts of the neocortex may be more important for some memories (for example the visual cortex processes visual memory), both the principle of equipotentiality and the principle of mass action held true. At the time Lashey’s research seemed so completely thorough that no-one presumed to question his results. Research favouring localisation of memories Penfeld’s use of electrical stimulation on the temporal lobes In the 1940s, Wilder G. Penfeld, a neurosurgeon at Montreal Neurological Institute started using electrical stimulation to map the motor, sensory, and language functions of patients about to undergo surgery for epilepsy. This mapping process could be achieved because the patients were still fully conscious and only had local anaesthesia. This enabled the patients to report any sensations when certain areas of the neocortex were electrically stimulated. From more than 1000 patients, Penfeld found that stimulation in the temporal lobes occasionally resulted in the patient having a flashback to an earlier experience (Kandel, 1992). This was the first direct evidence that the temporal lobes played a part in memory storage. The case of H.M. and medial temporal lobe lesions Perhaps the most well known case study in the study of memories is that of H.M. In 1953, H.M. was a 27 year old assembly line worker who had suffered for the past 11 years from extremely large epileptic seizures centred in the medial area of both temporal lobes. To reduce these seizures, a bilateral medial temporal lobectomy was performed by William B. Scoville of Montreal Neurological Institute (Kandel, 1992). Recent MRI scans show that approximately 5.5 cm of each medial temporal lobe was removed, this included the amygdala, most of the hyppocampus, and most of the surrounding cortex (Tippett, 1997). The operation significantly reduced the number of seizures, however H.M. was left with severe memory deficits. From studies of H.M. over the past 40 years, a lot has been learnt about different types of memory. Post surgery, it has been found that H.M. has severe anterograde amnesia ? he can not form new memories. He also has retrograde amnesia in which he can not remember events up to 11 years prior to the surgery (Tippett, 1997). H.M. has retained a normal short-term memory ? his digit span is 6 digits. He is able to perform tasks such as the repetition of digits or sequences of blocks up to his digit span. However he is not able to increase the sequence length by repetition of the same sequence (as is normally the case of people with intact long-term memories) (Pinel, 1993). He has not suffered from any intellectual loss and in fact his intelligence quotient has risen slightly over the years (possibly due to fewer epileptic seizures) (Pinel, 1993). Perhaps the most interesting feature (and the one most relevant to this essay) of H.M.’s memory is the discovery that he could learn new skills. This was first discovered when H.M. was asked to perform a mirror-drawing task. In this task he was asked to trace around a pattern inside a border by only looking in a mirror. As the number of trials increased, the number of times he went outside the border decreased. As expected however, H.M. claimed each time that he had never before performed the task (Pinel, 1993). From the study of H.M. we have learnt many things about memory. Firstly, we now know that the medial temporal lobes play an important part in the acquisition of memories. Secondly, there is a distinct neurological difference between short-term and long-term memories. Damage to the medial temporal lobes results in impaired long-term memory, however the short-term memory stays intact. The problem seems to be an inability to transfer short-term memory to long-term storage. Thirdly, there is a distinction between explicit and implicit memories. Damage to the medial temporal lobes leaves a person unable to recall specific events, however they can still acquire new skills through elementary kinds of learning such as habituation, sensitisation or classical conditioning. Other cases (for example Milner, 1965 cited in Pinel, 1993) have shown the removal of one medial temporal lobe may result in amnesia similar to H.M.’s. A possible explanation as to why this has occurred in only some cases, is that the other temporal lobe of these patients may not have been functioning fully. Further cases show the association between memory and the medial temporal lobes in patients with viral encephalitis (Squire, 1991). The case of R.B. and damage to the hippocampus In 1978, at the age of 52 R.B. became amnesiac when, during cardiac-bypass surgery, the blood flow to his brain was interrupted (Tippett, 1997). According to Pinel (1993), R.B.’s amnesia was similar to H.M.’s, though less severe. R.B. died in 1983 and a post-mortem revealed the only observable brain damage was to the CA1 subfield of the pyramidal cell layer of both hippocampi. Since this was an area removed during H.M.’s medial temporal lobectomy, it is possible to suggest that this was the area that caused H.M.’s amnesia also. Korsakoff’s syndrome and lesions in the medial diencephalon S. S. Korsakoff, a Russian physician of the late 1800s first described a syndrome in which severe memory loss was the primary symptom. It was generally assumed that it was caused by brain damage due to alcohol poisoning, but it is now known that it is caused by thiamin (Vit B1) deficiency that is associated with excessive alcohol consumption (Pinel, 1993). Patients with Korsakoff’s syndrome typically have retrograde and anterograde amnesia. The retrograde amnesia often includes memories that occurred many years prior to the onset of the syndrome becoming noticeable. Brain damage due to Korsakoff’s syndrome is often extremely diffuse, and it was difficult to determine which specific areas of the brain (if any) were associated with the memory loss of the patients. Initially it was thought that the memory loss was caused by lesions in the mammillary bodies, however extensive post-mortem studies of patients with Korsakoff’s syndrome by Victor, Adams and Collins (1971, cited in Pinel, 1993) pinpointed other areas. Their studies showed that in all cases with memory deficits, there was damage in the mediodorsal nuclei of the thalamus. Other studies (for example Brion and Mikol, 1978 cited in Pinel, 1993) however, have shown some patients with Korsakoff’s amnesia had no noticeable damage to their mediodorsal nuclei. Typically though, Korsakoff’s syndrome patients have damage in their thalamus and hypothalamus (medial diencephalon) as well as diffuse damage in other areas of the neocortex and cerebellum (Pinel, 1993). The case of Jimmie G., reported by Oliver Sacks (1985) is a classic example of Korsakoff’s syndrome. While Sacks makes no mention of specific areas of the neocortex that were damaged in Jimmie G., this case is typical in the type of memory deficits that a patient with this syndrome incurs. Jimmie G. was a navy man who served in the second world war then stayed in the navy until 1965. After he left, Jimmie started drinking heavily and around Christmas of 1970 became confused and developed memory deficits. He was sent to a few different hospitals and nursing homes before Sacks first met him in 1975. Sacks discovered that Jimmie’s memories stopped at 1945; Jimmie believed that he was 19 years old and the second world war was just over. Jimmie’s short-term memory was intact but he had profound anterograde amnesia and a retrograde amnesia that had deleted 25 years of memories. The case of N.A. and thalamic lesioning N.A. (reported in Pinel, 1993) also received damage to the mediodorsal nucleus of the thalamus, however in his case it was not from Korsakoff’s syndrome. In December of 1960 N.A. was injured when he was stabbed through the right nostril with a fencing foil. The foil travelled up and left into the forebrain after penetrating the cribriform plate. Immediately after the accident, N.A. had a retrograde amnesia of approximately two years. Two and a half years later he still had no memory for events up to about two weeks prior to the injury. For the first six to eight months after the accident N.A. had trouble remembering day-to- day events, however at times he would spontaneously recall other past events. A CAT scan taken in the late 1970s showed lesions in the left mediodorsal nucleus of the thalamus; there was no other visible damage alone the path the foil had taken. Alzheimer’s disease Alzheimer’s disease is a terminal disorder which is characterised by progressive dementia. In its early stages it is generally detected by a deterioration of memory. Studies have shown that there is an abnormally high proportion of “tangles” (twisted filaments inside individual neurons) and “plaques” (accumulations of degenerating neurons and abnormal protein) in the brains of sufferers of Alzheimer’s disease (Tippett, 1997). It has also been seen that tangles disproportionately affect the medial temporal lobes. There also seems to be a reduction in cholinergic activity in axons that supply acetycholine to the cortex, hippocampus and amygdala, (Tippett, 1997). Post- mortems show that amongst other things, there is large amounts of damage in the cortex, hippocampus and basal forebrain. Results from animal experimentation Research on animals allow neurologists to perform tightly controlled experiments. Specific areas of the brain in animals can be surgically lesioned and then the performance of the animal compared to the performance of control groups. This allows causal links between brain and function to be made, however it could still be questioned whether these results can be generalised to humans. A recent series of experiments on rats by Trond Myhrer and Katrine Wangen (1996) provide interesting results regarding the topic of this essay. Previous experiments had shown that cutting the fibres connecting the temporal cortex (TC) to the lateral entorhinal cortex (LEC) resulted in deficits in the retention of visual information. It had also been shown that cutting the fibres of the hippocampal perforant path (PP) caused deficits in the ability to acquire visual information. It had been assumed that both these lesions together would result in some combination of the deficits. Trond and Wangen’s results confirmed that separately TC/LEC and PP lesions caused the expected deficits. However, for TC/LEC + PP lesions, results showed that either there was no effect or the effect vanished as testing proceeded. These results seem to suggest that there is some sort of ‘compensatory mechanism’ that produces the recovery of function. Conclusions It seems from the cases reported in this essay and others, that both the medial temporal lobe and the diencephalon are important for explicit memory formation. The early work of Lashey is no longer regarded as sufficient evidence for the diffusion of memories. The work of Penfeld, and cases such as H.M. have shown that the medial temporal lobes play a part in explicit memory acquisition, and cases such as R.B. show that it may specifically be the hippocampus. Studies of patients with Korsakoff’s syndrome and Korsakoff-type symptoms show us that the diencephalon (thalamus and hypothalamus) also play an important role in memory. The case of N.A. shows us that memory loss can be associated with thalamic lesions. Alzheimer’s disease is an extremely complex disorder, however it seems that the memory losses associated with it may be attributed to damage to the medial temporal lobes. Experiment on animals confirm these case studies of human patients. Contrary to Lashey’s work, recent experiments lesioning the brains of animals show that there are specific areas which affect memory. However, it often not simply the case that increasing the size of the lesion increases the size of the memory deficit. All of these studies show that there is a large amount of evidence that supports theories of memory that assign specific memory abilities to specific regions of the brain. The popular belief at present is that while there is no single centre in the brain for memory, the formation and storage of memories takes place in specific areas and not the entire neocortex. References: Kandel, E. R., & Hawkins, R. D. (1992). The biological basis of learning and individuality. Scientific American, September, 53 - 60. Myhrer, T., & Wangen, K. (1996). Reduced cognitive dysfunctions in rats when the temporal-entorhinal cortices and hippocampal region are denervated simultaneously. Psychobiology, 24 (4), 281 - 293. Pinel, J. P. J. (1993). Biopsychology (2nd edn.). Boston: Allyn and Bacon. Sacks, O (1985). The man who mistook his wife for a hat. London: Pan Books. Squire, L. R., & Zola-Morgan, S. (1991). The medial temporal lobe memory system. Science, 253, 1380 - 1386. Tippett, L. J. (1997). Unpublished lecture notes for University of Auckland paper 461.230.