Memory, Learning, and Higher Function: A Cellular View

The basis of learning appears to be a network of interconnected adaptive elements (such as those found in the brain) by means of which transforms between inputs and outputs are performed. By adaptive I mean that the element can change in some systematic manner and in so doing alter the transform between input and output. In living systems, transmission within the neural network involves cpded nerve impulses and other physical chemical processes that form reflections of sensory stimuli and incipient motor behavior. The properties of the transmission network become significant determinants of behavior and depend on the mechanisms of neuronal adaptation, the means by which the connectivities between different neurons are modified. Particular paths through the network become labeled with reference to specific inputs and outputs. The network then operates through labeled interconnections linking specific elements within the network and through the mechanisms that underlie each element's adaptation. The adap­ tive features are crucial to learning and imply some associated, underlying mnemonic process. The labeling is of consequence with regard to the resulting specificities of stimulus reception and motor performance that characterize adaptive behavior. Memory involves time-dependent information processing relying on en­ coding and retrieval as well as storage itself. In the brain, engrams can be defined as those elemental adaptive changes that take place when learning and memory storage occur. Persistent engrammatic modifications of neuronal structure commonly arise through the same associative mechanisms responsi­ ble for learned behavior [397, 486, 759, 1020].