Abstract View

LEARNING THAT FOOD IS INEDIBLE IN FREELY-BEHAVING APLYSIA CALIFORNICA: A STATISTICAL MODEL V.C. Lin1; M.E. Fitzgerald3; M.S. Branicky3; H.J. Chiel1,2,4* 1. Dept of Biomedical Engineering, 2. Department of Biology, 3. Department of Electrical Engineering and Computer Science, 4. Department of Neurosciences, Case Western Reserve University, Cleveland, OH, USA How can the temporal structure of an animal's encounters with a reinforcing stimulus be characterized? How does this temporal structure change with learning? Can criteria based on these changes predict whether an individual animal has learned? We have studied these questions in the marine mollusk Aplysia californica. Previous studies have shown that Aplysia decrease their responsiveness to inedible food either when food is continuously presented (Susswein and Schwarz, Behav. Neur. Biol. 1983) or when they encounter it as they move freely (Chiel and Susswein, 1993). Statistical analysis of the time intervals between successive attempts to ingest inedible food (interbout intervals) indicated that these intervals can be described by two exponential distributions, with a long or a short mean time. This statistical model of the interbout intervals remained valid over time, but parameters changed as the animal repeatedly encountered inedible food: the mean time for long interbout intervals increased, and the probability of initiating a short interbout interval decreased. Parameter changes showed that statistically significant behavioral changes occurred over a single four-hour exposure to inedible food, and over successive days of exposure. By examining the duration of three successive interbout intervals, it was possible to construct a criterion to detect learning in individual animals during the learning process itself. This criterion may make it possible to examine neural correlates of learning in freely-behaving animals. Supported by: NSF DMI97-20309