Neuron Glia Biology, 2005 Memory beyond the synapse
Based on studies of the molecular and cellular cascade occurring during memory consolidation for a one trial passive avoidance learning task in the young chick, I review the evidence that memory is encoded in permanent changes in synaptic connectivity in a specific brain region, the Hebb hypothesis, and the implication of these findings for the development of so-called 'smart drugs' - cgontiive enhancers in the treatment of Alzheimer's Disease. I conclude that despite the fact that such a cascade occurs, culminating in the synthesis of cell adhesion molecules that are involved in synaptic remodelling, synaptic events are not in themselves sufficient to account for the phenomena of memory. Both whole brain (neuromodulator) and whole body (hormonal) processes are engaged. Memories are labile, disarticulated and stored in a distributed manner. How the mind/brain recreates coherent memories from this pattern is an unsolved mystery.
Another, more general, lesson is that Hebbianism is not sufficient. Although it might describe the processes involved in the early stages of memory formation, it cannot account for the ways in which over periods that vary from hours, weeks and even months in some species, the putative memory traces are disassembled and redistributed. Nor can it account for the renewed lability of memory following a reminder, which indi-cates that each time a memory is recalled, it is reconstructed in a molecular as well as a psychological sense. Furthermore, it is important to recognise that in practice nearly all the molecular and cellular studies of memory study the processes that are involved in learning and consolidation, not in recall of the memory.
We have no idea how recall occurs, whether it is the neural processes engaged in scanning the putative memory store to recover the instruction not to peck at an attractive bead, or the name of the person we are talking to. Nor do we understand how chicks, and humans, derive a coherent image from such distributed cues, the, so-called, binding problem. And even if we do explain this process in neural terms, I contend that no amount of sophisticated technology will enable us to read the actual CONTENT of that memory from the properties of the ensemble of neurons and their synapses, in whatever degree of detail. In this sense, memories are found, not only beyond the synapse, but beyond the brain itself, in the complex web of interacting processes that in humans we call a mind.