Lever conditioned stimulus-directed autoshaping induced by saccharin-ethanol unconditioned stimulus solution: effects of ethanol concentration and trial spacing.

Two experiments were designed to evaluate whether brief access to a saccharin-ethanol solution would function as an effective unconditioned stimulus (US) in Pavlovian-autoshaping procedures. In these experiments, the insertion of a lever conditioned stimulus (CS) was followed by the brief presentation of a sipper tube containing saccharin-ethanol US solution. Experience with this Pavlovian-autoshaping procedure engendered lever CS-directed autoshaping conditioned responses (CRs) in all rats. In Experiment 1, the concentration of ethanol [0%, 2%, 4%, 6%, or 8% (vol./vol.)] in 0.1% saccharin was systematically increased within subjects across autoshaping sessions to evaluate the relation between a rat's drinking and lever pressing. In Experiment 2, the mean intertrial interval (ITI) duration (60, 90, 120 s) was systematically increased within subjects across autoshaping sessions to evaluate the effect of ITI duration on drinking and lever pressing. A pseudoconditioning control group received lever CS randomly with respect to the saccharin-ethanol US solution. In Experiment 1, lever-press autoshaping CRs developed in all rats, and the tendency of a rat to drink an ethanol concentration was predictive of the performance of lever-press autoshaping CRs. In Experiment 2, longer ITIs induced more lever CS-directed responding, and CS-US paired procedures yielded more lever CS-directed responding than that observed in CS-US random procedures. Saccharin-ethanol is an effective US in Pavlovian-autoshaping procedures, inducing more CS-directed responding than in pseudoconditioning controls receiving CS-US random procedures. More lever CS-directed responding was observed when there was more drinking of the saccharin-ethanol US solution (Experiment 1); when the CS and US were paired, rather than random (Experiment 2); and with longer mean ITI durations (Experiment 2). This pattern of results is consistent with the hypothesis that lever CS-directed responding reflects performance of Pavlovian-autoshaping CRs.