Coordinated mental simulation in the hippocampus and primary visual cortex during active behavior

University Of California, San Francisco, San Francisco CA

Investigators

Abstract

PROJECT SUMMARY / ABSTRACT Our sensory experience goes beyond what is immediately in front of us, as evident when we recall memories or imagine future scenarios. These “mental simulations”, operationally defined as neural representation of states or stimuli that do not refer to actual present experience, are supported by a distributed circuit in the mammalian brain. Dysfunctions of this network could lead to psychiatric illnesses that affect sensory perception, like schizophrenia. Previous work shows that two key loci in this circuit are the hippocampus (HPC), a cognitive area that encodes space, and the primary visual cortex (V1), a sensory area that processes visual input. The goal of this proposal is to investigate how these regions interact to generate coordinated mental simulations during active behavior. To achieve this goal, this proposal combines innovative tools like flexible neural probes to obtain stable chronic recordings in these regions in freely moving rats and applies advanced state-space algorithms to read out a common cognitive variable from them with a high temporal resolution. Preliminary data shows that the animal’s position decoded from V1 moves ahead or behind the actual position by as much as ~ 15 cm. Furthermore, this is correlated with position decoded from HPC, indicating coordinated mental simulation across these regions. Aim 1 (K99) will test the generality of this coordination by changing external factors such as visual features and geometry of the environment. Aim 2 (K99/R00) will study the feedforward mechanism underlying coordinated mental simulations by examining the head and eye movements as well as comparing receptive field properties and laminar location of V1 neurons to their spatial tuning. Aim 3 (K99 pilot/R00) will identify the feedback mechanism underlying coordinated mental simulations by conducting the experiments in the dark to isolate the top-down influence and optogenetically silencing HPC to block its feedback to V1. This work will build on the candidate’s extensive expertise in visual neuroscience, electrophysiology, and quantitative data analysis by providing training in four key scientific areas, under primary mentor Loren Frank: (1) gaining a deeper understanding of the decoding algorithm with the guidance of advisor Uri Eden; (2) refining techniques for applying flexible neural probe with the guidance of advisor Chong Xie; (3) training in freely moving visual neurophysiology; and (4) learning optogenetic manipulation techniques, both with the guidance of co-mentor Massimo Scanziani. The training plan will build professional skills in mentorship, lab management and scientific communication to propel the transition to independence. UCSF offers a collaborative and supportive environment to pursue cutting-edge research in systems neuroscience and launch an independent career. The proposed work will provide key insights into the neural basis of mental simulations and build a foundation for the candidate’s long-term goal: to understand how the brain forms an internal model of the world from sensory information and uses it to guide behavior.