Retinotopic-to-egocentric transformation in visuoparietal cortex

The National Eye Institute awarded the University of California, Santa Barbara $117,840 on March 8, 2026, under the Vision Research program (CFDA 93.867) to support research on retinotopic-to-egocentric transformation in visuoparietal cortex.

The award funds a Project Grant investigating how the brain transforms visual object representations from a retinotopic reference frame (based on retinal photoreceptor topography) to an egocentric reference frame (mapped relative to the animal's head or body). The recipient will use two-photon calcium imaging to record cortical neuron activity in mice as they locomote within a chamber containing a salient visual object. By measuring eye movements and object location relative to the animal, the research will develop analysis methods to distinguish retinotopic and egocentric object vector tuning across visuoparietal cortex spanning primary visual cortex (V1) and posterior parietal cortex (PPC). The first aim maps the distribution of reference frame coding between V1 and PPC to identify which neural populations compute this transformation. The second aim records functional calcium activity of dendritic spines to determine how retinotopic position and eye position signals integrate within individual neurons showing joint coding for multiple behavioral variables.

Performance occurs in Santa Barbara, California. The period of performance runs through October 31, 2028.

Abstract

To perceive and interact with a complex and changing environment, animals use visual inputs to build an internal model of their local space. Early visual regions one or two synapses from the retina represent object locations in a retinotopic reference frame preserving the relative topography of retinal photoreceptors. Significant cortical processing is required to transform these object representations to an egocentric reference frame found in association cortex, in which object locations are mapped relative to the animal's head or body. The neural circuitry underlying this critical transformation is not well understood, especially under conditions of active visual sampling. Computational models have proposed that egocentric object location can be calculated by integrating retinotopic location and gaze orientation in a multi-layer artificial neural network. This model makes testable predictions about the neuron response types that would be expected in the intermediate steps of reference frame transformation. In this proposal, I will use two-photon calcium imaging to record the activity of large populations of cortical neurons as mice locomote within a chamber containing a salient visual object. By measuring eye movements and the location of the object relative to the animal, I will develop an analysis approach for distinguishing retinotopic and egocentric object vector tuning in large populations of cortical neurons spanning visuoparietal cortex. I will first map the distribution of reference frame coding between primary visual cortex (V1) and posterior parietal cortex (PPC) to determine which neural populations compute this transformation (aim 1). To determine how retinotopic position and eye position signals are integrated within single neurons which show joint coding for multiple behavioral variables, I will record the functional calcium activity of dendritic spines in animals with sparse calcium indicator expression (aim 2). Together, these results will reveal the biological implementations underlying reference frame transformation in the mammalian cortex. In understanding this canonical computation, we will better understand how visual deficits contribute to deficits in spatial perception and navigation.