Date of Award

Spring 1993

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

Robert G Mair


Twenty-eight male rats were pretrained on a go/no-go, continuous olfactory non-matching to sample (CONMTS) task, and randomly assigned to one of four treatment groups: radiofrequency lesions of either the medial wall or rhinal sulcal cortical projection sites of the mediodorsal thalamic nucleus (MDn); lesions of the lateral internal medullary lamina site (L-IML) of thalamus; or sham surgical control.

All three lesioned groups exhibited a significant impairment on initial retraining of CONMTS. However, the two cortical groups rapidly improved, and did not differ from the control animals for the remainder of CONMTS training. Alternatively, the L-IML deficit persisted throughout retraining. The cortical and control subjects exhibited susceptibility to proactive interference effects created by reducing the number of odor stimuli, and a moderate decay in performance was observed with increased intertrial intervals. No differential effect of these manipulations was observed in the L-IML group, apparently due to a floor effect.

Subsequent training on go/no-go odor discrimination tasks demonstrated that the CONMTS deficit, characterized by an increase in "false alarms", was not attributable to an inability to suppress responding to nonreinforced odors. All subjects, including the L-IML lesioned rats, learned the discrimination tasks at equivalent rates.

The present findings demonstrate that singular destruction of specific MDn projection sites in prefrontal cortex does not disrupt performance of an olfactory working memory task. The L-IML lesions produced deficits on working, but not reference, memory components of the paradigm. This effect of L-IML lesions is consistent with previous studies utilizing spatial stimuli, demonstrating that the amnestic effects seen following thalamic destruction are multimodal.