Neurosciences
Program

 Brady, A.M., Bailey, A.M., Calhoon, G.G., Logan, T.T., McGill, J.A., Ruiz, C.T., and Saul, R.D.  (2007, November).  Impaired executive function and learning set formation in the neonatal ventral hippocampal lesion model of schizophrenia.  Poster presented at the Society for Neuroscience Annual Meeting, San Diego, CA.

Abstract 

Patients with schizophrenia display deficits in higher cognitive functions, including set-shifting and hypothesis formation.  Here we tested set-shifting and hypothesis formation abilities in the neonatal ventral hippocampal lesion (NVHL) model of schizophrenia.  Rats received bilateral infusions of ibotenic acid (NVHL group) or artificial CSF (sham group) into the ventral hippocampus at postnatal day (PD) 7.  As adults (PD 56 or older), rats were assessed for either set-shifting (n=15) or hypothesis formation (n=7).  In addition, a third set of rats (n=11) were tested on the set-shifting task as adolescents (beginning at PD 28).  Set-shifting was tested using a T-maze task, in which animals first learned one of two response rules (direction-based or visual cue-based) and were then required to shift to the previously irrelevant rule.  Animals were trained on each rule (counterbalanced for order) until they correctly completed 10 consecutive trials plus a probe trial.  Hypothesis formation was assessed by monitoring the acquisition of a learning set during a series of olfactory discriminations.  Animals were given 30 odor-unique discrimination problems with 5 trials per problem.  In the set-shifting T-maze task, NVHL (n=8) and sham (n=7) animals tested as adults required a similar number of trials to learn the initial response rule.  However, NVHL animals needed significantly more trials to shift to the new response rule (p < .05).  In contrast, NVHL (n=7) and sham (n=4) animals tested as adolescents demonstrated similar levels of performance at each phase of testing.  Thus, NVHL animals displayed normal acquisition of a response rule, but were impaired in suppressing responding to a previously learned rule and shifting to a new rule.  Furthermore, this deficit did not emerge until after adulthood.  In the learning set task, NVHL animals tested as adults were significantly less accurate than sham animals over all trials (p < .05).  Furthermore, sham animals (n=4) demonstrated a performance level significantly higher than chance (50% correct) on Trials 2, 3, 4, and 5, suggesting the successful use of a learning set.  However, NVHL animals (n=3) did not perform better than chance until Trials 4 and 5, suggesting that they were not able to use a learning set, although they were able to solve olfactory discrimination problems.  Together, these findings suggest that NVHL animals are able to acquire simple response rules and perform basic discriminations, but are unable to use these skills to perform higher cognitive functions.  These results add to previous research suggesting that the NVHL model is a valid animal model for investigating cognitive impairments in schizophrenia.