A new-hatched chick - chicks are hatched precocially, so they’re up and running - and they need to distinguish very quickly between what’s food and what’s not food, and they do that by pecking at the food. And they are particularly attracted to small, bright objects the size of a pinhead or something of that sort [SR correction: size of a waterdrop]. If you offer a chick on the end of a, a stick - you have a little bright pin or a little bright chrome bead or something - the chick will peck at it. If you dip the chrome bead in a bitter tasting substance, the chick will peck once; it’s distasteful stuff, they will shake their heads and back away from it and they won’t peck at a bead of that colour subsequently. So they’ve learnt from that single, one trial task, that that’s distasteful, and they’ll go on pecking at beads of other colours or other shapes or other sizes and so on. So it’s a very simple and very powerful learning task. And, so we took over the whole task of using the passive avoidance learning into the Open University. By that time we had got proper custom-built laboratories, a proper custom-built animal house, and we could therefore hatch the chicks, let them hatch up under controlled conditions. We could then train them, we could train chicks in groups of twenty, thirty or forty at a time. And, you would run the experiments. There are two things you can do basically. You can train the chicks and look at biochemical changes which are going on in the brain, or you could, if you thought a particular biochemical process was necessary for memory formation, you could block that biochemical process and see whether the chick still remembered the task. So, one of the clear ideas that was around in memory, biochemical memory research at that time, was the involvement of protein synthesis, that long-term memory depended on the synthesis of proteins. So you could train the chicks, and you could inject into their brains an inhibitor of protein synthesis, and there are a number of drugs which do this. And you could then test whether the chick still retained memory for the task or not. And there are a whole variety of experiments you can do in that sort of way, looking at the involvement of neurotransmitters, like looking at the involvement of proteins, looking at the involvement of RNA, and so on, and focusing down on the region in the chick brain was involved in the training task. I was very interested in coupling the biochemical observations with seeing if, at the level of particularly electron microscopy, you could detect any difference in the numbers and the structure of synapses, the junction points between the cells. So, the collaborative experiments also involved looking at those brain regions using the electron microscope, or, in another set of experiments, looking at the development of dendritic spines and showing that the number of spines - that is the junction points between one nerve cell and another - changed as a result of training in the task. And showing that [these changes did take place] became a classical and very well-cited paper, the first author being a PhD student who had done the actual work of counting the spines, which is a hell of a tedious job in that sort of way. These days you could probably automate it. So, that programme, and variants of that programme, working with passive avoidance, ran right the way through the rest of my research career there.