Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Mapping recognition memory in the primate brain: why it's sometimes right to be wrong

Aggleton, John 1999. Mapping recognition memory in the primate brain: why it's sometimes right to be wrong. Brain Research Bulletin 50 (5-6) , pp. 447-448. 10.1016/S0361-9230(99)00127-6

Full text not available from this repository.


It has been acerbically observed that the importance of a piece of research can be measured by the length of time that it holds up progress in that field. Although the comment was supposed to be sardonic, it does hold a grain of truth. This is because being wrong can be enormously helpful in advancing a field, just as long as the ideas are clearly formulated and testable. My ‘Twentieth Century Highlight’ is a study [4] where the findings were correct but their interpretation was utterly wrong. By the 1960s it was accepted that the temporal lobes contain at least one structure vital for episodic memory, and that the hippocampus was the most probable candidate. Much of the evidence came from amnesic patients with damage in the temporal lobes, including the famous case H.M [9]. Unfortunately, the neuropathological data could not resolve the issue and it became apparent that lesion studies with nonhuman primates might offer a more direct approach. A series of studies in the 1960s and 1970s studied the cognitive effects of hippocampal lesions in monkeys but the impairments on tasks like visual discriminations and delayed response seemed very mild when compared to the clinical condition, and it seemed impossible to model anterograde amnesia. The breakthrough came in 1978 with a paper by Mortimer Mishkin who reported that separate removals of the hippocampus and amygdala led to very mild deficits on a test of recognition memory, yet joint removal of the two structures had a devastating effect [4]. Parallel findings were reported for tests of associative memory [7] and tactile recognition [8], so strengthening the case that joint limbic pathology was responsible for temporal lobe global amnesia. This was a totally new concept, yet it seemed to make sense of a great deal of the neuropathological data and rapidly became the dominant viewpoint. The 1978 study was a landmark because it not only showed that animal studies could model aspects of anterograde amnesia but it also highlighted the importance of selecting the right behavioural test for a given question. Thus a behavioural assay for amnesia did not simply consist of any task involving learning—what was needed was a test that measured a function severely disrupted in anterograde amnesia, and such a test became possible with the development of a behavioural test of visual recognition [2] , [5] and [6] . Recognition memory is typically lost in amnesia and this study was one of the first to examine this aspect of memory in animals. Nowadays, so much more is known about spared forms of learning in amnesia that the need to tailor behavioural tests more precisely seems obvious but this was not always so. There is, however, a twist in the tale. As researchers pioneered ways of making more and more selective cortical and subcortical lesions in the temporal lobes a quite extraordinary fact emerged. Painstaking work by Murray and Mishkin at NIH and by Zola and Squire in San Diego [10] and [11] showed that circumscribed lesions of the amygdala and hippocampus alone had little, if any, additive effect contrary to the interpretation by Mishkin [4]. In fact, the critical regions for recognition proved to be not the hippocampus and amygdala but the perirhinal cortex and, to a lesser extent, the entorhinal cortex [3]. These rhinal regions, which lie just ventral to the amygdala and hippocampus, had been completely overlooked when interpreting the original study even though they had been removed. This was because almost all of the previous evidence had pointed to the hippocampus itself and next to nothing was known about any independent functions of the rhinal cortices. Since then the perirhinal cortex has proved to be critical for more than one mnemonic function, indicating that the amnesic syndrome can be further fractionated [1]. The discovery that the hippocampus and amygdala need not be critical for recognition in monkeys reaffirms one of Sherlock Holmes’ most famous maxims that “when you have eliminated all which is impossible, then whatever remains, however improbable, must be the truth”.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Psychology
Neuroscience and Mental Health Research Institute (NMHRI)
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Publisher: Elsevier
ISSN: 0361-9230
Last Modified: 04 Jun 2017 02:43

Actions (repository staff only)

Edit Item Edit Item