Welcome to the Katz Lab @ Brandeis University

Figure 1

Work on this project is ongoing. We are performing an experiment in which restrained rats use a lever-press to self-administer tastants through the intra-oral cannula (this general procedure is useful when adapting rats to restraint, and serves as preparatory training for many of the experiments in our research program). Rats are trained to wait 30 seconds between lever presses-they quickly learn that they can control delivery of reward, and in the process, we assert, come to make predictions about the specific fluid that will be delivered with a lever press. In this protocol, lever press resulted in the delivery of a particular tastant (e. g., NaCl) during a 5-8 minute period, after which the "reward" was switched to a tastant of different palatability (e. g., quinine HCl). As the session continued, a highly palatable stimulus (e. g., sucrose) replaced the quinine HCl, and then another unpalatable tastant (e. g., nicotine) replaced sucrose, and then the cycle was repeated.

The animals' behavior in this paradigm reflects the growing beliefs and expectations that accompany self-administration: Through the 5-8 minutes in which lever press is associated with delivery of a palatable tastant inter-response intervals decrease; they increase, meanwhile, during time periods in which the association is with delivery of an unpalatable stimulus. Furthermore, recordings in GC reflect this expectation effect. Basic analysis of within-trial firing rate changes (Pauluis & Baker 2000) demonstrate that individual neurons produce tastant responses that change in latency with successive presentations, beginning earlier and earlier in response to palatable tastants, and later and later in response to unpalatable tastants (Figure 1); the "response" to the last presentation of highly palatable tastants often precedes tastant delivery, whereas the response to the last presentation of an aversive tastant appears at a latency consistent with earlier estimates of the start of the Middle epoch (Katz, et al., 2001). Experience with highly palatable tastants changes processing from the naīve state such that tastant-specific firing "predicts" the actual chemosensory epoch, when the tastant is self-administered.

At an even longer time-scale than the trial-to-trial (i. e., at the level of repeated runs through blocks of each self-administered tastant), further remarkable changes can be observed. Within a single session, some neurons developed dynamic responses to new tastants-e. g., see the neuron described in Figure 2, which developed responses first to quinine and then to sucrose and NaCl. Embedded within this pattern are, again, the cross-trial changes in dynamics related to palatability.

Figure 2

Bibliography:

Katz DB, Simon SA, Nicolelis MA. 2001. Dynamic and multimodal responses of gustatory cortical neurons in awake rats. J Neurosci 21: 4478-89.
Pauluis Q, Baker SN. 2000. An accurate measure of the instantaneous discharge probability, with application to unitary joint-even analysis. Neural Comput 12: 647-69

Brandeis University	Brandeis University Dept. of Psychology	Volen Center for Complex Systems