Stimulus competition and visuospatial selection: Neural circuit and computational mechanisms

PROJECT SUMMARY Animals routinely operate in complex environments rich in sensory stimuli. They handle this informational complexity by having their behavior guided by the most salient (and more generally, highest priority) stimulus in the environment. The neural circuit mechanisms underlying competitive selection of the highest priority stimulus across space remain poorly understood. Recent evidence from behaving monkeys has revealed that the intermediate and deep layers of the superior colliculus (SCid), a major sensorimotor hub in the vertebrate midbrain, are required for normal competitive stimulus selection. In parallel, our work in the barn owl optic tectum (OTid, avian homolog of SCid) has revealed that OTid neurons signal the highest priority stimulus categorically, which can account for SCid’s critical role in spatial selection behavior. Such categorical signaling requires a specialized, donut-like pattern of spatial inhibition onto OTid from an inhibitory nucleus called Imc in the nearby midbrain isthmic complex. Separately, feedback from a cholinergic isthmic nucleus, Ipc, is known to amplify OTid firing rates. Despite these insights, several fundamental questions about the functional logic of this isthmo-tectal (OT-Imc-Ipc) network for competitive stimulus selection remain open, and here, we address three. First, we aim to elucidate how the preferential neural signaling of the most salient stimulus is implemented in OTid. Specifically, our hypotheses are (1a) that OTid signals the strongest among competing stimuli with a combination of enhanced gamma synchrony, reduced trial-trial variability, greater resistance to adaptation, and reduced noise correlations (rather than just with elevated firing rates), and (1b) that cholinergic Ipc serves as a multiplexed mechanism for regulating these diverse OTid coding features. Second, we aim to elucidate a neuromodulatory mechanism for the control of categorization. Our hypotheses (based on our model predictions) are (2a) that (hitherto uncharacterized) spatially-specific cholinergic input onto Imc neurons causes multiplicative modulation of Imc activity, and (2b) that it serves as a mechanism to enhance the degree of categorical signaling of the most salient stimulus by OTid. Third, we aim to uncover the source and broader functional logic of this cholinergic input. Our hypotheses are (3a) that Ipc is the dominant source of cholinergic input onto Imc neurons, and (3b) that in parallel, Ipc, which influences activity across the layers of the OT, selectively spares the OT layer that provides input to it (layer 10); revealing a precisely organized circuit design that minimizes runaway feedback excitation in this network. We will test these hypotheses using in vivo electrophysiology and drug iontophoresis during visual stimulus presentation in awake, head-fixed barn owls, together with computational modeling. Preliminary data from the three aims support our hypotheses. Results from the proposed work have the power to reveal fundamental neural circuit mechanisms involving multiple neurotransmitter systems for executing the sophisticated computations that underlie stimulus competition and spatial selection across space.