Lidocaine infusion in a critical region of cerebellum completely prevents learning of the conditioned eyeblink response.

New Zealand white rabbits were implanted with cannulas in the dorsal or ventral aspect of the anterior interpositus nucleus. Three days (and 6 days for some dorsal-infusion animals) of standard tone-airpuff training was given with continuous infusion (constant rate of 0.2 M-L/min) of lidocaine or saline. All animals were then given 3 days of training with no infusion. All lidocaine-infused animals exhibited no conditioned responses in the 3 or 6 days of infusion training. Dorsal-infusion animals learned in the subsequent 3 days of no infusion training as if naive, that is, they exhibited no savings. Animals with ventral cannula locations learned during infusion training, as shown in postinfusion training. These results strongly support the hypothesis that the essential memory trace for eyeblink conditioning is formed and stored in the cerebellum. Permanent lesions of neural structures can serve to identify the circuitry essential for a given form of learning and memory but cannot provide conclusive evidence for the localization of memory traces. The use of the reversible inactivation technique, on the other hand, can provide such evidence. Thus, if naive animals are trained during inactivation of a region of the brain essential for a type of learning, expression of the learned response will be prevented during inactivation training. If learning occurs during inactivation, as evidenced in postinactivation training, then the site of memory trace formation must be before the region of inactivation in the essential circuitry. If no learning occurs during inactivation training (i.e., if animals learn as though naive in postinfusion training), then the region of inactivation must be the site of memory formation or a mandatory afferent projecting ultimately to the site of memory formation. To use reversible inactivation, it is first necessary to identify the brain circuitry essential for a given form of learning and memory. This has been largely achieved for one form of basic associative memory in animal studies: aversive classical conditioning of discrete behavioral responses, such as eyeblink and limb flexion (Lavond, Kim, & Thompson, 1993; McCormick, Clark, Lavond, & Thompson, 1982; McCormick & Thompson, 1984; Rosenfield & Moore, 1983; Thompson, 1986; Yeo,