Investigations of the cellular bases of memory.

Part 1:


Behaviourally, memory is a change in behaviour as a result of experience. Slightly more specifically, it is a change in response to a stimulus as a result of experience. This covers everything from habituation onwards - classical conditioning, instrumental learning and so on. (It doesn't normally include rapid and reversible adaptations to changes in the sensory qualities of stimuli or response demands.) We assume that these changes in response to stimuli are achieved through some physical changes in the organism which learns. Given our understanding that behaviour is primarily controlled through the action of the nervous system it is reasonable to expect that these physical changes occur in neurons or at least within the nervous system.

Ideally we would like to understand the changes which occur within and between cells when these memories are formed. We can ask two different questions about these changes:

  1. What conditions are necessary for a change to occur?
  2. What changes occur?
together with a linking question

Before discussing research which has attempted to answer these questions let us briefly think about the types of physical changes which can occur in neurons or the nervous system in order to get a feel for the area.

I am going to concentrate on four areas of research in my survey of work on the cellular basis of memory: Although much of this work uses simple invertebrate preparations as models, the processes being examined are so basic that they provide useful insights into similar processes in vertebrate, mammalian and even human memory formation.

Beliefs about the bases of memory.

In the middle ages memories were thought to be contained in the cerebrospinal fluid. The fluid itself was somehow altered by the process of learning. This is in contrast to Descartes' hydraulic theory of nervous action which held that the function of the nervous system relied on the transmission of fluid from and to the brain through nerves. Descartes' theorised that memories were recalled via a pumping action of the pineal gland on the CSF - "When the mind wills to recall something, this volition causes the little gland (the pineal), by inclining successively to different sides, to impel the animal spirits towards different parts of the brain, until they come upon that part where the traces are left of the thing it wishes to remember."

In the 60's there were speculations that models of information storage and retrieval derived from genetics might be relevant to psychological explanations of learning and memory. The most direct hypotheses one can derive from these speculations are that memory is encoded in the structure of RNA, DNA or related molecules.

The opinion most acceptable today is that memories are encoded in the pattern and relative strengths of the connections between neurons in the brain. The Hebb hypothesis is one of the earliest and the most influential models of memory of this type. Hebb suggested that strength of connection across a synapse between two neurons would increase whenever the two neurons were simultaneously active. Repeated co-ocurrences of stimuli or actions would have concomitant patterns of neural activity which, as a result of Hebbian synaptic modification, would develop reverberating patterns of common neural activity called cell-assemblies. We will not address the question of the existence of cell-assemblies, but we will consider whether Hebb's model of synaptic modification, along with the other hypotheses mention previously are borne out by experimental findings.

Biochemical Memory in flat-worms?

There has been no convincing experimental work carried out which supports the mediaeval theory of mnemonic CSF or a mnemonic function for the pineal gland. The first body of work I will consider the evidence for a molecular basis for memory in flat- worms.

The flat-worm or planarian is a very simple invertebrate, nevertheless, in 1955 Thompson and McConnell showed that planaria could be classically conditioned to avoid light by pairing a light CS with an electric shock US. It becomes clear just how simple an animal a planarian is when you discover that if one is cut in half while alive the two halves regenerate into two complete flatworms - the tail-half grows a new head and the head-half a new tail! McConnell's initial discovery about memory in flatworms was that once a flatworm had been conditioned to avoid light if you cut it in half and allow the halves to regenerate both of the resulting worms show evidence of knowing the light-shock association. McConnell interpreted this as evidence that memory in flatworms was not localised in the head but was, rather distributed throughout the animal. In 1962 McConnell performed an experiment which appeared to be even more dramatic demonstration of this. After training some planaria he ground them up an fed them to other planaria. These animals were quicker at learning the light-shock association than controls who were fed ground-up untrained worms.

Given the excitement at the time about the discovery of the chemical basis of genetic information encoding and transfer - the structure and role of DNA was discovered by Watson, Crick, Wilkins and Franklin in March of 1953 - a number of researcher began to investigate whether DNA or RNA were involved in what McConnell had hypothesised was the diffuse chemical encoding of memory in planaria. In 1961 Corning and John showed that the apparent memory of the tail of a bisected planarian could be disrupted by RNAase - an enzyme which destroys RNA, although there was still evidence for a retained association in the head-half even after regeneration in RNAase.

These and other similar results prompted a rise in interest in memory by biochemists. Many experiments were carried out on transfer of memory and on attempts to isolate the chemicals carrying these 'memories', not only in planaria, but in fish, mice and rats. There was, however, a great deal of controversy about these experiments - many scientists found that they could not replicate others' studies and methods were often criticised. For example, Frank, Stein and Rosen (1970) carried out an experiment where one group of mice were trained to associate the light side of a test box with shock, another group were stressed by being rolled around in a jar but were not placed in the test box and a third group were untrained controls with experience of the test box. When the brains or livers of these animals were removed, ground up and fed to other animals it was found that recipients who had been fed the brains of trained animals escaped from the light faster than those that had been fed the brains of untrained controls, however, animals fed the brains of donors stressed in the jar escaped faster still. The speediest escapes were made by animals which had been fed the livers of jar-stressed donors. Stein interpreted this as showing that 'transfer' was not memory specific, rather, apparent changes in behaviour or learning-rate could be attributed to stress hormones transferred between donor's and recipients. Eventually it became clear that RNA did play a part in memory, but it did not appear to code specific memories. It was, however claimed that a peptide associated with RNA during the process of extracting the RNA did encode specific memories. There was a great deal of confusion at the time about the role of these peptides. It became clear that these peptides were part of the chain leading to stress hormone production, and so the hypothesis of memory specific biochemicals died.


Reconstructing the history of worm-running (McConnell ran a journal called the Worm Runner's Digest for a while!) can be a bit tricky. It was very controversial so textbooks (other than those written by worm runners themselves) tend to just ignore it. Most accounts written at the time are highly polarised, either uncritical or completely dismissive. As a topic it is almost entirely ignored in modern textbooks, probably because nearly all the results were misinterpreted at the time. It is, however, salutary to see how these misinterpretations occurred. It should cause healthy skepticism of more modern claims made about memory which might fall into the same pitfalls (inappropriate memory test, inappropriate interpretation of manipulations etc.). I based my account largely on these sources:

McConnell, J.V. and Jacobson, A.L. (1974) Learning in invertebrates. In D.A. Dewsbury and D.A. Rethlingschafer (Eds.) Comparative psychology: A modern surbey. Tokyo:McGraw Hill Kogashuka.

Kimble, D.P. (1965) The anatomy of memory (Volume 1). Palo Alto CA: Science and Behavior Books.

Thompson, R.F. (1975?) Introduction to physiological psychology (1st Edition). New York:Harper and Row.

Blundell, J.E. (1975) Physiological Psychology. London: Methuen.