A visual lamina in the medulla oblongata of the frog, Rana pipiens.

We have discovered a lamina of visually responsive units in the medulla oblongata of the frog. It spans the entire medial aspect of the rostrocaudal length of the medulla and extends dorsoventrally from the cell-dense dorsal zone into the cell-sparse ventral zone. Most visual units within this lamina have large receptive fields, with the majority extending bilaterally in the frontal visual field. Most of these neurons are binocular, have no apparent directional preference, respond equally well to stimuli of a variety of shapes and sizes, and exhibit strong habituation. More medial locations in the visual lamina represent ipsilateral visual space while more lateral locations within the lamina represent contralateral visual space. Many units in the caudal aspect of the visual lamina are bimodal, responding to both visual and somatosensory stimuli. HRP tracing reveals inputs to the lamina from many primary and secondary visual areas in the midbrain and diencephalon. There is no area-by-area segregation of the projections to the visual lamina. For example, most parts of the tectum project across the visual lamina. The only spatial order in the visual lamina is that at more medial sites there tends to be more input from contralateral tectum; and at more lateral sites there tends to be more input from ipsilateral tectum. There is bilateral input to the visual lamina from tectum, tegmentum, posterior nucleus of the thalamus, posterior tuberculum, and ventromedial thalamic nucleus. There is ipsilateral input to the visual lamina from torus semicircularis, pretectum, nucleus of Bellonci, and ventrolateral thalamic nucleus. There is contralateral input to the visual lamina from basal optic complex. Collectively, these results show the presence of visual influences in regions of the medulla that likely represent an important step in sensorimotor transformation.

Connections of Anterior Thalamic Visual Centers in the Leopard Frog, Rana pipiens.

The amphibian retina projects to two discrete regions of neuropil in the anterior thalamus: the neuropil of Bellonci and the corpus geniculatum. These retinorecipient areas are encompassed within a larger zone of surrounding neuropil we call the NCZ (for neuropil of Bellonci/corpus geniculatum zone). The NCZ is characterized electrophysiologically by a distinctive tonic oscillatory response to blue light; it appears to be a visual module involved in processing the stationary visual environment. Using horseradish peroxidase (HRP), we mapped the connections of the NCZ. Retrogradely labeled cell bodies are found in: (1) the contralateral anterior thalamus; (2) both retinas; and (3) the posterior medial dorsal thalamus (PMDT). Anterogradely labeled fibers are found in: (1) the contralateral anterior thalamus; (2) the ipsilateral PMDT; (3) the ipsilateral neuropil lateral to the posterior tuberculum in the ventrolateral posterior thalamus; and (4) the ipsilateral anterior medulla. There are no direct connections between the NCZ and the telencephalon, the tectum, or the suprachiasmatic nucleus. Applying HRP to the PMDT, we found that its inputs are limited to the contralateral and ipsilateral NCZ and the contralateral PMDT. Thus, PMDT appears to be a satellite of the NCZ. Blue light elicits tonic oscillatory electrical responses in the PMDT quite similar to the responses to blue light in the NCZ. We discuss how the leopard frog NCZ and the mammalian ventral lateral geniculate nucleus share anatomical and physiological properties.

Representation of the visual field in the anterior thalamus of the leopard frog, Rana pipiens.

We used physiological and anatomical methods to elucidate how the visual field is represented in the part of the dorsal anterior thalamus of the leopard frog that receives direct retinal projections. We recorded extracellularly while presenting visual stimuli, and characterized a physiologically defined region that encompasses the retinal projections as well as an extended zone beyond them. We probed the area systematically to determine if the zone is organized in a visuotopic map: we found that it is not. We found that units in this region respond only to stimuli in the contralateral half of the visual field, which is similar to what is seen in the dorsal lateral geniculate nucleus in mammals. When we backfilled retinal ganglion cells from application of HRP to the anterior thalamus, we found labeled cells only in those parts of the retina corresponding to the contralateral hemifield, confirming our physiological observations.

Light and shadow: visual recognition of the stationary environment by leopard frogs.

We determined how leopard frogs respond to non-moving aspects of the environment. We have discovered that these frogs are attracted to dark, stationary, opaque objects. This attraction depends on the relative reflectance of the object, i.e., the darker the block, the more attractive it is, and the attraction is found under both bright and dim ambient light levels. Larger blocks are more attractive than smaller blocks, but frogs are still attracted to blocks much smaller than themselves. Previous studies have shown that frogs are also attracted to sources of light. Using a choice experiment, we show that the probability a frog will choose a dark object versus a light source depends on the intensity of the light source relative to the intensity of the ambient light. The frog only moves toward a light source when it is at least 20 times brighter than the brightest object in the environment. These findings help to clarify the frog's "phototactic" nature.