Historically, several biomedical disciplines have played a part in developing our understanding of memory, with physiology playing a key role. Best known is the discovery of long-term potentiation (LTP) in the hippocampus by Terje Lømo in Norway (Lømo, Acta Physiologica Scand. 1966) and the first full report of the phenomenon by Bliss and Lømo (Journal of Physiology1973).  From the outset, LTP was found to have properties desirable of a memory mechanism – that it long outlasts the duration of the initiating stimulus, is pathway specific, and apparently associative in character. Another key discovery was that of place cells in the hippocampus by O’Keefe and Dostrovsky (Brain Research 1971), a finding followed in the years afterwards by those of other spatially tuned cells such as head-direction units (Taube et al, J. Neuroscience 1990) and grid cells (Hafting et al,Nature 2005). Collectively, these helped build the idea of the hippocampus being key to spatial memory. An entirely different learning system, based in the striatum, is instrumental in the learning of actions and habits as established in physiological, human functional imaging data and computational models; it deploys an error-correcting learning rule (Schultz et al, J Neuroscience 1992; Montague et al, J Neuroscience 1996).

It would, however, be wrong to suppose that the significance of these findings rests solely on physiological data. Neuroanatomy has also played a key role, dating back to Cajal’s Croonian Lecture to the Royal Society in 1894; likewise neuropsychology, as in Hebb’s conjectures about cell-assemblies in the brain and his proposal for a simple synaptic learning rule, as exemplified by LTP (Hebb, The Organisation of Behavior, 1949); pharmacology also weighed in with the discovery that glutamate is the major excitatory transmitter of the brain and that selective glutamate antagonists such as D-AP5 blocks the induction of LTP without affecting baseline glutamatergic transmission (Collingridge et al, Journal of Physiology, 1983). Behavioral studies have also contributed by rigorously testing the idea that activity-dependent synaptic transmission is necessary for the formation of episodic and spatial memory traces (Morris et a, Nature, 1986).

Contemporary studies using several of the remarkable technological innovations of recent years (e.g. optogenetics, calcium imaging in awake animals) are building on these foundations in intriguing ways. This lecture, with its requested historical backbone, aims to outline progress over the years and the challenges that remain in understanding memory as a fundamental feature of higher cognitive function.