Peter M. Milner (1999) Cell Assemblies: Whose Idea?. Psycoloquy: 10(053) Lashley Hebb (4)

Volume: 10 (next, prev) Issue: 053 (next, prev) Article: 4 (next prev first) Alternate versions: ASCII Summary
Topic:
Article:
PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).
Psycoloquy 10(053): Cell Assemblies: Whose Idea?

CELL ASSEMBLIES: WHOSE IDEA?
Book Review of Orbach on Lashley-Hebb

Peter M. Milner
Department of Psychology
McGill University
1205 Dr. Penfield Ave.
Montreal, QC
CANADA H3A 1B1

Ps64@musica.mcgill.ca

Abstract

Orbach raises the possibility that Lashley anticipated the theory advanced by Hebb (1949) in his book The Organisation of Behaviour. According to Orbach, Lashley's publications provide evidence that he was thinking of a theory similar to Hebb's before about 1945, when Lashley first saw a draft of Hebb's book. Orbach reproduces a number of Lashley's papers which he believes provide support for this thesis. I consider the evidence far from convincing. Hebb's assemblies are learned representations of concepts. Lashley's resonators are innate aids to perception. Hebb's assemblies, but not Lashley's resonators, can be associated with each other in explaining thought.

Keywords

cell assembly, central autonomous process, engram, equipotentiality, Hebb, Hebbian learning, Lashley, localization, memory trace, nativism, reverberatory circuit, Vanuxem Lectures
1. Orbach's (1998, 1999) book is mostly by and about Karl S. Lashley. Hebb makes an appearance mainly as the lucky disciple whose writing skill, it is suggested, brought him the acclaim that should really have been Lashley's. The greater part of the book consists of a selection of Lashley's publications plus a series of lectures, never before published, that he delivered at Princeton University in 1952.

2. Lashley is greatly to be admired. He retained a sensible approach to psychology as it passed through a bad period, and raised important questions that are still valid. Orbach has performed a useful service by making available some of Lashley's previously inaccessible or elusive writing. He also shows us the less admirable side of Lashley, his stubbornness, lack of experimental rigor and, by implication, a tendency toward self-deception.

3. The first part of the book is mainly concerned with the question of whether the vital ingredients for the cell assembly theory presented by Hebb (1949) were part of Lashley's thinking before 1942. Some of the publications of Lashley reproduced by Orbach in the second part of the book were intended to demonstrate this thesis, but only the first three were written before Lashley had read drafts of Hebb's book.

4. The first of the selected papers (Lashley, 1924) is of little direct relevance to the thesis, though there are indirect implications that will be considered later. The paper is a report of what by present day standards would be considered an extraordinarily poorly conceived experiment, purporting to show that useful information is stored by neurons that do not participate in the learning process. Rats learned a brightness discrimination with one eye covered and retained the learning when the blindfold was moved to the other eye.

5. Seven years later, in the second paper chosen by Orbach, Lashley (1931) describes the experiment again but partially retracts the conclusion saying: "Of course the corresponding cells of the two retinae may excite the same cells of the central system so the experiment is not crucial." Instead, he substitutes a stimulus-equivalence experiment in which a rat is taught to jump to a solid white triangle and later jumps to a small outline of the same figure. Lashley assumes that because different receptors are stimulated, learning must have taken place in cells not fired during the learning. But, as in the earlier experiment, he ignores the possibility that some cells may be fired by both the positive figures. He did so despite the fact that he spent much of his life looking for pathways through which this could occur.

6. The main topic of the second of the reprinted Lashley papers is his theory that the function of much of the neocortex is to facilitate subcortical integrative processes. This theory accounts for learning deficits that are proportional to the mass of cortex damaged, but independent of its location. Although there is a brief mention of a stable interference pattern arising in a homogeneous cellular matrix as a result of sensory input, I can find nothing in this paper that in any way foreshadows Hebb's neural representation of a concept.

7. Orbach's third selection is Lashley's (1942) major attack on the problem of stimulus equivalence. Lashley was apparently a pushover for Golgi stained sections of the brain. He himself recounted the story of finding a collection of Golgi stained frog-brain slides when he was a young laboratory assistant and thinking it would be possible to discover how the frog worked by tracing the connections. He was disappointed to discover that the Golgi technique stained only a minute fraction of the neurons, but the lesson did not stick. In this paper we find him stating, in relation to one of Lorente de N's (1934) Golgi diagrams of the rat cortex, that "From such a structural organisation functional properties may be inferred with some confidence." He confidently goes on to fantasise a cortex made up of millions of resonators tuned to a wide range of frequencies through which waves produced by sensory impulses spread and mutually interfere, giving rise to standing wave patterns.

8. In Lashley's optimistic view, the interference patterns would be replicated many times throughout the visual cortex and would remain fixed even when the stimulus pattern moved from one set of receptors to another. Any physicist could have disabused him of that idea, but Lashley conceded that it was just an oversimplified analogy to set people thinking along the right lines. For the present purposes, the question is not whether Lashley's ideas have merit, but whether they contain the seeds of Hebb's cell assembly. I can discern no sign of any.

9. It is true that Lashley appreciated the reverberatory possibilities of Lorente de N's recurrent circuits before Hebb did, but he saw the circuits as innate resonators capable of spreading the effect of a stimulus across the cortex, not as structures established by learning to represent a stimulus. Lashley used the reverberatory circuit to solve the stimulus-equivalence problem, essentially as part of an innate perceptual mechanism. For Lashley, learning came later to associate the fixed interference pattern with a response. As we saw from his earlier papers, Lashley, at this stage, was opposed to the suggestion that synapses could be changed by activity, and only began to entertain that possibility after he had encountered Hebb's work on the cell assembly.

10. Hebb's great leap forward, for which he received all the glory (glory that Orbach implies should go to Lashley), was that he proposed a neural solution to the concept problem. Ideas had been considered to be purely mental by most psychologists up to that time (even by behaviourists, which is why they refused to talk about them). Hebb gave ideas a neural basis and opened the door to a cognitive neuroscience. There is no hint of this development in Lashley's use of Lorente de N before about 1945, when Hebb told him about it.

11. Like Orbach, I found it surprising that Hebb (1980) should have acknowledged Hilgard and Marquis (1940) as the source of his information about Lorente de N's recurrent circuits. Surely it must have been impossible to work with Lashley for any length of time in the 1940s without hearing about resonators and interference patterns. But perhaps Hebb did not pay too much attention to Lashley's attempts to explain stimulus equivalence. He took enough physics at Dalhousie to recognise this as a pipe dream. Hilgard and Marquis introduced Lorente de N in a learning context, and that may well have been the moment of Hebb's epiphany. It is possibly significant that in his 1942 paper, and even as late as 1949, Lashley cites only Lorente's anatomical papers, while Hilgard and Marquis (1940) cite only his electrophysiological papers of 1938, as does Hebb (1949).

12. Hebb and Lashley worked on different problems. Both showed great ingenuity, both were equally impractical. The cell assembly would never have worked as Hebb claimed, any more than Lashley's loosely sketched model would have worked. I think Lashley was right that the main learning takes place after the perceptual process is complete, but neither he nor Hebb saw that it is impossible to solve either the perceptual or the conceptual problem in a vacuum, considering nothing but sensory input (Milner 1999).

13. Lashley's papers after 1947 show a definite influence of Hebb's ideas, though as Orbach points out, Lashley never (or hardly ever) refers to him. It is possible that Lashley did not understand the radical step that Hebb had taken, and really believed that the theory was merely a distorted version of his own. It is true that some of Lashley's ideas about psychology are reflected in Hebb's book. Students are supposed to absorb ideas from their teachers, but Hebb's unique contribution of a learned neural representation of the concept could never have come from Lashley; it was quite contrary to everything Lashley believed or preached up to that time. Lashley has managed to convey the impression to Orbach, and perhaps others, that he had been thinking along those lines for years, but the evidence is to the contrary. In fact I think at least as good a case could be made that Lashley (after 1945) adopted Hebb's ideas without acknowledgement as can be made for the reverse.

14. As a postscript, let me suggest that even if Lashley did get less credit than he deserved for his contribution to Hebb's book it may be poetic justice, because perhaps he got more credit than he deserved for his own book, Brain Mechanisms and Intelligence (1929). In that work Lashley published a summary of his researches on the effects of cortical lesions in the rat. His maze learning data show that error scores are highly correlated with cortical lesion size, but not location. From these data Lashley formulated his Laws of Mass Action and Equipotentiality.

16. It has been clear for many years that these laws have only limited application at the primate level, a limitation usually attributed to the rat's simpler brain and the poor specificity of maze tests. These may not be the only factors. Alerted by a remark in Orbach's book, I had another look at this seventy year old classic. Cortical lesions for the research were almost all produced by thermocautery, whose effects are not easy to predict. Fortunately Lashley made careful measurements of the extent of damage, including unintended subcortical lesions. In the rat the hippocampus lies closely below a considerable part of the posterior cortex, so it is not surprising that it bore the brunt of the unintended damage. Of the 37 experimental subjects in the study, 21 also sustained hippocampal lesions. Again, it is not surprising that the frequency and extent of these lesions appears to be highly correlated with the size of the corresponding cortical lesions.

17. Lashley, not having the benefit of subsequent findings on the dire effect of hippocampal lesions on maze learning, chose to ignore this damage (along with damage to the thalamus, striatum, and other subcortical structures). Considering that blindness did not prevent normal learning of the mazes, it is entirely possible that the hippocampal lesions were a major cause of the maze deficits seen by Lashley. Perhaps Lashley just missed finding his elusive engram -- in the hippocampus.

REFERENCES

Hebb, D.O. 1949. The Organisation of Behaviour. New York: Wiley. pp. 1-335.

Hebb, D.O. 1980. D.O. Hebb. In: G. Lindzey (Ed.), A history of psychology in autobiography (Vol.7). San Francisco: Freeman. pp. 273-303.

Hilgard, E.R. & Marquis, D.G. 1940. Conditioning and learning. New York: Appleton-Century. pp. 1-429.

Lashley, K.S. 1924. Studies of cerebral function in learning. VI. The theory that synaptic resistance is reduced by the passage of a nerve impulse. Psychological Review, 31, pp. 369-375.

Lashley, K.S. 1929. Brain mechanisms and intelligence. Chicago: University of Chicago Press. pp. 1-186.

Lashley, K.S. 1931. Mass action in cerebral function. Science, 73, pp. 245-254.

Lashley, K.S. 1942. The problem of cerebral organisation in vision. In: H.Klver (Ed.), Visual Mechanisms. Biological Symposia, vol. 7. Lancaster, Pa.: Cattell Press. pp. 301-322.

Lorente de N, R. 1933, 1934. Studies on the structure of the cerebral cortex. J. Psychol. Neurol., Lpz. 45, 381-438; 46, pp. 113-177.

Lorente de N, R. 1938. Analysis of the activity of the chains of internuncial neurons. J. Neurophysiol., 1, pp. 207-244.

Milner, P.M. 1999. The autonomous brain. Mahwah, NJ: Lawrence Erlbaum Assoc., pp. 1-155.

Orbach, J. (1998) (Ed.) The Neuropsychological Theories of Lashley and Hebb. University Press of America.

Orbach, J. (1999) Precis of: The Neuropsychological Theories of Lashley and Hebb. PSYCOLOQUY 10(23). ftp://ftp.princeton.edu/pub/harnad/Psycoloquy/1999.volume.10/ psyc.99.10.029.lashley-hebb.1.orbach http://www.cogsci.soton.ac.uk/cgi/psyc/newpsy?10.029


Volume: 10 (next, prev) Issue: 053 (next, prev) Article: 4 (next prev first) Alternate versions: ASCII Summary
Topic:
Article: