Abstract

This dissertation represents efforts to characterize the ontogeny of complex task variants of eyeblink classical conditioning (EBC) and to adapt these tasks to a rat model of fetal alcohol spectrum disorders (FASD). The well established behavioral, neurobiological, and developmental properties of EBC make this paradigm attractive for comparative studies of developmental disorders involving cerebellar injury. Alcohol exposure at critical early developmental periods in both humans and rats produces substantial cerebellar damage and associated deficits in single-cue EBC. Relative to single-cue EBC, interstimulus interval (ISI) discrimination and temporal uncertainty tasks appear to place greater demands on cerebellar mechanisms required for the production of well-timed conditioned responding (CRs). In ISI discrimination, two distinct conditioned stimuli (CSs; tone and light) are reinforced with a periocular shock unconditioned stimulus (US) at two different CS-US intervals (280 and 880 ms). Temporal uncertainty is identical in design with the exception that the same CS is presented at both intervals, thus making the subject "uncertain" at CS onset as to when the US will occur.

Experiments 1a and 1b (Chapter 6) examined performance in ISI discrimination and temporal uncertainty over postnatal days (PD) 23, 30, 45, and 70. Previous studies have indicated that performance in these tasks undergoes a protracted developmental emergence beyond ages when single-cue EBC achieves adult-like levels. The findings of Experiment 1a replicated and extended our previous findings using ISI discrimination training across ontogeny. Specifically, post-weanling/juvenile (PD23, 30) rats displayed elevated and earlier-timed CRs to the long CS relative to older subjects (PD45, 70). These differences can be attributed to enhanced cross-modal processing between the short (e.g., tone) and long (e.g., light) CS in younger subjects, though comparison with age-matched subjects trained in temporal uncertainty suggests that this generalization is incomplete. Previous findings of enhanced double-peaked CRs in juveniles relative to post-weanlings trained in temporal uncertainty were not replicated in Experiment 1b.

ISI discrimination and temporal uncertainty were further examined in Experiments 2-5 following neonatal exposure to varying doses of alcohol (Chapters 7 and 8). It was predicted that the complexity of these tasks would reveal functional impairments following exposure to lower doses than those required to impair single-cue EBC. Specifically, neonatal exposure to 3 g/kg/day of alcohol over PD 4-9 was expected to produce early-timed, non-adaptive CRs while exposure to 4 and 5 g/kg/day was hypothesized to produce robust impairments in CR generation measures (percentage and peak amplitude). Contrary to predicted outcomes, robust CR generation impairments were observed in both adults (PD70; Experiment 2) and juveniles (PD30; Experiment 4) trained in ISI discrimination following neonatal exposure to all three alcohol doses (3, 4, and 5 g/kg/day). CR generation was also significantly disrupted in adults (Experiment 3) and juveniles (Experiment 5) trained in temporal uncertainty following neonatal exposure to 4 and 5 g/kg/day of alcohol. Impairments produced by exposure to 3 g/kg/day of alcohol were less consistent in temporal uncertainty training. CR timing (to the long CS) was affected by neonatal alcohol exposure only in adults trained in ISI discrimination (Experiment 2). Late onset and late peaking CRs were observed in alcohol-exposed subjects relative to controls in Experiment 2, contrary to predicted premature timing effects. Interestingly, alcohol-induced deficits appeared to be greater in adults trained in ISI discrimination relative to juveniles.

Experiments 1-5 taken together, demonstrate that ISI discrimination EBC undergoes robust developmental modification beyond the juvenile period and that application of this task (and to a lesser extent, temporal uncertainty) to a rat model of FASD is useful in identifying functional deficits at alcohol doses lower than those required to disrupt conditioning in single-cue tasks. Findings in this dissertation may represent early steps in using EBC to establish reliable early identification of children prenatally exposed to moderate alcohol levels.