Toward a cellular analysis of intracranial self-stimulation: contributions of collision studies

Since the discovery of brain stimulation reward by Olds and Milner, researchers have struggled to identify the underlying neural circuitry. This goal has proved surprisingly elusive. For example, the identity of the directly-activated neurons ("first stage" neurons) responsible for the rewarding effect of stimulating the numerous brain sites that support self-stimulation remains largely or entirely unknown. It was to address this problem that the collision test was adapted for use in experiments on intracranial self-stimulation. By estimating the trajectory, conduction velocity and axonal diameter of the first stage neurons, it was hoped that their identification would be facilitated. Indeed, the choice of candidate pathways has been tightly constrained by collision data. For example, such data suggest that the circuitry underlying brain stimulation reward includes myelinated fibers directly linking self-stimulation sites in the lateral hypothalamus and ventral tegmental area; the conduction velocity of these fibers has been estimated at 1-8 meters/sec. Additional collision data suggest direct axonal links between self-stimulation sites in the preoptic area and lateral hypothalamus, as well as between sites in the ventral tegmental area and periaqueductal gray matter. Although collision data constrain the choice of candidate pathways, they cannot prove that a given population of neurons is part of the first stage. No matter how closely the anatomical and physiological characteristics of a given population match the properties inferred from collision data, the possibility remains that the population in question plays no role in reward but happens to resemble neurons that do. This ambiguity can be reduced by assessing how collision effects are altered by lesions of the candidate pathway. Coupled with data from single-unit recording experiments, inferences drawn from such lesion-induced changes provide a powerful means of linking an identified population of neurons to the rewarding effect of electrical brain stimulation.