Investigating the development of the visual motion pathway at the network level

Interactions between visual areas clearly play an important role in the processing of visual information in adult animals. At the same time, we know little about how these interactions shape development. This question is especially important for the visual motion pathway, as anatomical data from primates show a very early development of higher motion area MT. Here, we use the ferret’s motion pathway – in particular V1 and area PSS, an MT homolog – to investigate this issue. In a first set of experiments, we quantify the contribution of either area on the other at different ages by combining simultaneous extracellular recordings in V1 and PSS with reversible inactivation of either area by cortical cooling. These data so far show that PSS in visually naïve animals, but not in adults, retains responsiveness and tuning despite V1 inactivation. In a second set of experiments, we then directly test how development of one area depends on input from the other. To this end, we use an established training paradigm, in which visually naïve animals are exposed to drifting gratings for 8 hr, in conjunction with cooling. Normally, this paradigm induces rapid maturation of direction selectivity in V1 and PSS. Our preliminary data show that if V1 is cooled during the training, PSS still develops direction selectivity, but V1 does not. Finally, using retrograde tracer injections we demonstrate that the lateral geniculate nucleus in the thalamus sends a much stronger projection to PSS in visually naïve than in mature animals. All these findings contradict the basic assumption of a feedforward, sequential development that propagates forward from V1, and highlight the need to study development at the network level. This research is essential for understanding the etiology of perceptual symptoms associated with disorders of development or abnormal visual experience in early life.