The work on Aplysia was extremely encouraging, and I considered myself much more fortunate than I thought I had a right to be, but I realised in Aplysia, as wonderful as it was, we were studying implicit memory, a memory for motor and perceptual skills, which did not require conscious attention.

What Brenda Milner described, and what really appealed to the psychoanalytically-oriented part of myself was explicit memory storage, a conscious recall of information by people, places, and objects. And conscious recall, I mean, that relates to consciousness per se, is a very deep problem, and I clearly, you know, wanted to get involved in studying that.

Now I left the hippocampus in 1960, and I decided in 1990 – 30 years later – that I wanted to return to it, to explore explicit memory storage. And the reason I wanted to return to it was not only because the problem was so challenging, but because there had been three major advances in the study of hippocampus.

One is Bliss and Lomo had described their form of synaptic plasticity which they call long-term potentiation (LTP) which they produced by stimulating one of the pathways in the hippocampus at very high frequencies – 100Hz or more – and they produced a facilitation that lasts for very long periods of time. Now this was really quite interesting although it had some problems with it.

One problem was that that frequency was unusually high, and many people, myself included, thought that this was not the kind of frequency that's going to be used by learning processes, certainly nothing of that frequency… nothing comparable to what we had seen in Aplysia.

Number two, that its relationship to memory in the hippocampus was also uncertain. Now Richard Morris had worked out a behavioural task, and had shown that if you block LTP, which we knew was mediated by NMDA receptors, if you blocked the NMDA receptors that you interfered with memory storage. And that was encouraging, but it was very correlational and indirect because this drug affected lots of synaptic actions in the brain. So I think, you know, I felt one needed to understand LTP better.

But in addition, John O'Keefe had made the marvellous discovery that the hippocampus encodes space. That there are cells in the hippocampus that fire when a particular animal assumes a certain position in space. So there are lots of these place cells, and some fire when the animal is here, some fire when the animal is here, some fire when the animal is there, and if you look at the whole pattern, you can predict where the animal is by the firing pattern of the cells. And he showed me one day, when I was in London, what this is like. It was absolutely remarkable. He put the mouse on the table and it had multiple recording electrodes in, it was really quite spectacular. And that also made a lot of sense to me, because when Alden and I were trying to see what sensory inputs drive hippocampal pyramidal cells, and we tried light and we tried touch, we tried auditory stimuli; they weren't particularly effective. But place is a multi-sensory modality, and you could understand why we were not effective.

And the third thing is that Capecchi and Smithies had just described methods for knocking out genes in mice. So here was a method that allowed you to dissect LTP, to relate different components of LTP to memory storage in hippocampus, and to relate both of those things to place cell behaviour. So it opened up a whole new approach to the hippocampus that made the hippocampus very attractive.