AMO Special Seminar - September 7th
David Biron
Univ. of Chicago
Quantitative
descriptions of complex behavior
Abstract |
The definition of
behavior as the total motion of an animal hides great complexity. Precise
measurements, relevant perturbations, and theoretical modeling are key for
understanding the dynamics, function, and evolution of behavior. The first
part of the talk will describe feeding behavior in the roundworm C. elegans. Worms exhibit bursty
feeding dynamics, the function of which is unknown. Our central assumption is
that food intake serves a dual purpose: to gather information about the
environment as well as to ingest nutrients. We have developed methods for
prolonged and precise measurement of feeding and characterized the observed
dynamics. The resulting data drove a model which feasibly implements
trade-offs between speed versus accuracy and exploration versus exploitation
and predicts three regimes in responding to a dynamic environment. The second part of the talk will focus on
the most mysterious of all behaviors - sleep. The roles of sleep and the
reasons for its universality remain controversial. The worm is the simplest
model system in which such questions can be addressed and it shares many of
the relevant mechanisms with more complex organisms. In particular, worms are
an excellent model for the roles of sleep on the scale of cells and tissues.
In all animals examined, sleep deprivation results in cellular “hardware
malfunction”. On the organismal level this manifests as reduced performance
of the tired animal. We have found similar effects in worms and extended them
to a type of malfunction that was only hinted at previously. Interestingly,
distinct types of damage are prominent in different tissues of tired worms.
In addition to reducing performance of simple behavioral modalities, we found
that tiredness can increase correlations between them. Therefore, sleep
deprivation can be used simultaneously to examine the functions of sleep and
the organization of composite behavioral patterns. |