Nonlinear careers through academia are increasingly common, but funding agencies and search committees penalize these paths. Why do scientists stray from the beaten path, how do they contribute to science, and how do we level the playing field?

The starburst amacrine cell in the mouse retina presents an opportunity to examine the precise role of sensory input location on neuronal computations. Using visual receptive field mapping, glutamate uncaging, two-photon Ca(2+) imaging, and genetic labeling of putative synapses, we identify a unique arrangement of excitatory inputs and neurotransmitter release sites on starburst amacrine cell dendrites: the excitatory input distribution is skewed away from the release sites. By comparing computational simulations with Ca(2+) transients recorded near release sites, we show that this anatomical arrangement of inputs and outputs supports a dendritic mechanism for computing motion direction. Direction-selective Ca(2+) transients persist in the presence of a GABA-A receptor antagonist, though the directional tuning is reduced. These results indicate a synergistic interaction between dendritic and circuit mechanisms for generating direction selectivity in the starburst amacrine cell.

Direction-selective ganglion cells (DSGCs) are tuned to motion in one direction. Starburst amacrine cells (SACs) are thought to mediate this direction selectivity through precise anatomical wiring to DSGCs. Nevertheless, we previously found that visual adaptation can reverse DSGCs's directional tuning, overcoming the circuit anatomy. Here we explore the role of SACs in the generation and adaptation of direction selectivity. First, using pharmacogenetics and two-photon calcium imaging, we validate that SACs are necessary for direction selectivity. Next, we demonstrate that exposure to an adaptive stimulus dramatically alters SACs' synaptic inputs. Specifically, after visual adaptation, On-SACs lose their excitatory input during light onset but gain an excitatory input during light offset. Our data suggest that visual stimulation alters the interactions between rod- and cone-mediated inputs that converge on the terminals of On-cone BCs. These results demonstrate how the sensory environment can modify computations performed by anatomically defined neuronal circuits.