Sign- versus goal-tracking: effects of conditioned-stimulus-to-unconditioned-stimulus distance.

Three pigeons were exposed sequentially across experimental phases to five different distances between the conditioned stimulus and the site of the unconditioned stimulus in a sign-/goal-tracking procedure. A computer-controlled tracking system provided a continuous record of the bird's position by continuously monitoring the location of the bird's head in three-dimensional space. It was found that birds sign-tracked (i.e., approached the conditioned stimulus) when the conditioned stimulus was closest to the site of the unconditioned stimulus, goal-tracked (i.e., approached the site of the unconditioned stimulus in the presence of the conditioned stimulus) when the conditioned stimulus was farthest from the site of the unconditioned stimulus, and engaged in both sign- and goal-tracking (or something intermediate) at intermediate conditioned-stimulus-to-unconditioned-stimulus distances. When both sign- and goal-tracking occurred, the former tended to occur in the first half and the latter in the second half of the interval in which the conditioned stimulus was present. The results suggest (a) whether sign- or goal-tracking (or both) occurs is a function of the distance of the conditioned stimulus from the site of the unconditioned stimulus, (b) the fact that pigeons but not rats have been found to sign-track consistently throughout the duration of the conditioned stimulus may be due to quantitatively rather than qualitatively different effects of conditioned-stimulus-to-unconditioned-stimulus distance across species (i.e., a "short" conditioned-stimulus-to-unconditioned-stimulus distance for a pigeon may be a "long" one for a rat), and (c) sign- and goal-tracking may be competing behavioral tendencies that can (e.g., at intermediate conditioned-stimulus-to-unconditioned-stimulus distances) cancel each other out. The findings lend support to theories that specify an interaction between phylogenetic and reinforcement variables in determining whether sign- or goal-tracking will occur in any given experimental preparation.