This action is to provide partial support for the Third Conference on the Neurobiology of Learning and Memory. This conference is being organized by the Center for the Neurobiology of Learning and Memory and will offer presentations and discussion by leading investigators from around the world. The topic of the conference will be the relationship between the design of cortical and cortico- subcortical circuitries and the emergence of particular aspects or types of memory, including those involved in cognition. Three issues related to this topic will be discussed on successive days. First, the question of the forms of memory will be explored; speakers on the first day of the meeting will address the question of how many types of memory systems co-exist in brain as well as the neurobiological and psychological relationships existing between them. Descriptions of the substrates of memory, either in terms of cellular chemistries or neural networks, very often fail to specify what is meant by "memory". This practice has generated considerable an unnecessary confusion. Accordingly, the conference will begin with an extended analysis of the characteristics of a wide range of learning paradigms and preparations (from conditioning in invertebrates to episodic memory in humans) and seek to identify commonalities and differences. Nearly all learning theorists assume that most of the storage and processing of memory by humans and other big- brained mammals is accomplished by networks which run through the neocortex and connect it with specific subcortical regions such as the amygdala and hippocampus. The second day of the conference will be concerned with the organized of cortical circuitries, what we know of their plasticity, and how their specialized features might be utilized for memorial operations. Surprisingly enough, there have been very few attempts to link the neurobiology of neocortex with higher psychological functions such as integration and partitioning of memory; in many senses, the second session, which will be directed at this issue, will be breaking new ground. The third and last day of the meeting will explore cortical memory systems using computer models. A spectrum of network models will be discussed, ranging from those grounded in physiology to more purely theoretical constructions. In summary, the conference is designed to explore in sequence three related questions: 1) How many forms of memory are there? 2) How do these forms relate to the specialized circuitry that characterizes cortex? and 3) What types of computational operations must these circuits perform in order to produce memory?
