Neuronal organization underlying visually elicited prey orienting in the frog--III. Evidence for the existence of an uncrossed descending tectofugal pathway.

A complete transverse hemisection of the neuraxis just caudal to the optic tectum in the frog, Rana pipiens, results in a failure to orient toward stimuli in one visual hemifield [Kostyk and Grobstein (1986) Neuroscience 21, 41-55]. This finding indicates that each tectal lobe gives rise to a crossed descending pathway adequate to cause turns in a direction contralateral to that tectal lobe, and suggests that each may also give rise to an uncrossed descending pathway adequate to cause turns in the ipsilateral direction. To determine whether there is in fact such an uncrossed pathway, we have studied the orienting behavior of frogs after lesions which interrupt crossed pathways. Two groups of animals were studied. In one group we made midline lesions of the ansulate commissure, through which run the major crossed descending projections from both tectal lobes. In the other group, we combined a complete transverse hemisection with removal of the tectal lobe on the same side of the brain, leaving intact only an uncrossed pathway from one tectal lobe. A persistence of orienting turns was observed in both groups of animals. In both, the direction of the turns was that expected on the assumption that an uncrossed pathway would cause ipsilateral turns. We conclude that such a pathway exists. While both groups of animals turned in the expected directions, they did so for stimuli at unexpected locations. Increasingly eccentric stimulus locations to one side of the mid-sagittal plane were associated with increasing amplitude turns to the other. The observation suggests that tectal regions mapping areas of visual space to one side of the mid-sagittal plane are capable of triggering turns not only in that direction but in the opposite direction as well. In the case of ansulate commissure section, mirrored orienting responses were observed for tactile stimuli as well. These and other behavioral anomalies described in the preceding papers [Kostyk and Grobstein (1986) Neuroscience 21, 41-55 and 57-82] suggest that between the topographic retinotectal projection and the premotor circuitry for orienting there may exist an intermediate processing step, one in which stimulus location is represented in a generalized spatial coordinate frame.