My long-term career goal is to run my own laboratory studying neural circuit mechanisms and computation principles for processing visual information. I expect to make my own contribution by answering the questions that bridge neurophysiology, neural computation, and animal behavior. I plan to apply versatile approaches, such as in vivo whole-cell patch-clamp technique, in vivo two-photon calcium imaging, optogenetic and chemogenetic perturbation, and computational modeling. I will dissect the neural circuits from synaptic mechanism to population coding, and from computation to function. Vision is the most important sense. A better understanding of the neural mechanisms underlying visual processing will help us uncover the principles of brain computation and shed light to vision-related neurologic disorders. The retina sends visual information to two principal brain centers: the superior colliculus (SC) and the visual cortex. Although the SC has long been implicated in mediating visually guided behaviors, previous studies on neural circuit mechanisms underlying visual processing have largely neglected the SC and focused on the retina and the primary visual cortex (V1). The superficial layer of SC receives direct inputs from ~30 distinct subtypes of retinal ganglion cells (RGCs), and each of these parallel channels carries specific visual features. In addition, the SC receives top-down projections from layer 5 of V1. How are the retinal features processed in the SC and modulated by cortical feedback in awake animals? We will answer this question at both population level and synaptic level using mice as a model with three distinctive aims.
In Aim 1, we will reveal the neuronal population coding for representing visual features in the SC of behaving mice. We will first investigate how different visual features are represented at the population level by imaging calcium activity in the SC of behaving mice with the cortex left intact. We will then probe the roles of genetically labeled RGCs and collicular neurons in this representation. Last, we will uncover how the representation is modulated by different behavioral states.
In Aim 2, we will reveal the roles of inhibition in the representation of visual features in SC. We will first image population responses of inhibitory neurons. Then, we will examine how the representation in excitatory neurons is affected by optogenetically modulating inhibitory neurons. Last, we will explore how inhibition contributes to the synaptic integration by directly recording synaptic excitation and inhibition using in vivo whole-cell voltage-clamp technique.
In Aim 3, we will investigate the relative contributions of retinal and cortical excitation to the representation of visual features in awake SC at the synaptic level. We will first probe how the retinal excitation contributes to the feature representation by combining optogenetic silencing of both the SC and the visual cortex with in vivo whole-cell voltage-clamp recording. Then we will reveal the cortical contributions by directly recording synaptic excitatory currents with and without cortical inputs in the same collicular neuron.
This study will reveal neural circuit mechanisms underlying visual features representation in superior colliculus of behaving mice at multiple scales: from population coding to synaptic computation. The results will offer new mechanistic insight into visual processing beyond the retina. Ultimately, this will lead to improvements in diagnosing and treating vision-related neurological disorders.