Our Research
Towards a neural basis for sensory-guided behavior in health and disease.
Sensory-guided behavior is ubiquitous in everyday life, whenever sensory information is used to guide the decisions we make and the actions we take. Consider the sequence of behavioral outcomes that occur when the sound of sirens become present during our morning commute; we first hear the auditory stimulus, before additional neural computations take place that likely elicit an emotional response, alongside a motor action as we engage in a perceptual decision regarding whether, and where, to pull over. This simple example illustrates the complexity of this neurobiological problem; auditory signals need to be routed to downstream regions to elicit an appropriate behavioral response. This transformation of sensory information into such a behavioral response is necessarily a brain-wide endeavor.
Our research program is dedicated to understanding how the flow of sensory information through brain-wide neural networks gives rise to purposeful behavior in models of health and disease. Using a variety of cutting edge experimental and theoretical systems neuroscience tools to monitor, manipulate, and model the neural circuits of awake, behaving mice, we investigate brain-wide circuits in a number of different contexts:
Perceptual decision-making, learning, and state-dependent dynamics. Mice learn to make decisions based on the identity of auditory stimuli. We seek to identify which neural circuits are involved, what auditory information is propagated brain-wide, what neural computations take place across learning, and how this information is modulated through top-down interactions from higher-order areas. We also seek to identify how a fluctuating internal state impacts upon the ability to make sensory-guided decisions.
Sensory-guided foraging. Mice are tasked with using auditory cues to forage for hidden targets (resources or prey). These freely-moving, naturalistic, behaviors are combined with chronically implanted technologies that facilitate the recording of massive neural populations so that we can identify the neural computations that take place as mice navigate complex environments.
Acoustic communication. Multiple mice are allowed to to interact with one another in a naturalistic setting. By combining neurophysiological approaches with video/sound tracking, we can accurately record and assign sound to space (to accurately track ongoing bouts of vocalizations) and identify which neural circuits contribute to social behaviors that rely on acoustic communication (courtship, for example), as well as probing how communication deficits manifest in disease.
Brain-wide propagation of auditory hallucinations. Auditory hallucinations feature prominently in many psychiatric disorders. We utilize targeted noise-exposures to induce auditory hallucinations in mice, prior to engagement auditory-guided behaviors. By longitudinally monitoring distinct neural populations, we aim to track cognitive deficits across learning, correlate these cognitive deficits with neural activity, and use cell-type specific manipulation strategies to alter the brain-wide propagation of hallucinatory signals to reverse these cognitive deficits.
Our Approaches
PHYSIOLOGY: OBSERVATION AND PERTURBATION OF NEURAL CIRCUITRY DURING SENSORY-GUIDED BEHAVIOR
Sensory-guided behavior: We develop and utilize both head-fixed and freely-moving sensory-guided behaviors where mice are tasked with utilizing sensory information to drive a behavioral outcome.
Internal state monitoring: We routinely monitor face/body movements and changes in pupil diameter, to use as internal state proxies that can be combined with neurophysiological and behavioral measurements.
Vocalization tracking: We track vocalizations of mice as they navigate their environment and communicate with one another.
Calcium imaging: We monitor large-scale network activity of genetically identified cell-types in the auditory cortex during sensory-guided behaviors using calcium imaging (two-photon microscopy/miniscopes), and then perturb these networks using optogenetics (spatial light modulation) or chemogenetics. These techniques allow us to characterize large neural populations during sensory-guided behavior, and to understand how local networks of different cell-types interact with themselves and each other.
Fiber photometry: We utilize fiber photometry to simultaneously monitor the population calcium dynamics of neurons in the auditory cortex as well as in their downstream targets. This allows us to investigate how sensory signals are propagated brain-wide to coordinate sensory-guided behavior.
Electrophysiology: We monitor and perturb neural activity of distinct auditory cortical cell-types alongside their downstream projection targets using chronic electrophysiology and optogenetics. This again allows us to investigate how sensory signals are propagated brain-wide to coordinate sensory-guided behavior, but opens the door to more targeted manipulation strategies, and optimal temporal precision.
COMPUTATIONAL NEUROSCIENCE & MACHINE LEARNING: STATISTICAL ANALYSIS OF NEUROPHYSIOLOGICAL DATA
Advances in experimental techniques allow for the simultaneous observation of thousands of neurons. Such high-dimensional data sets can be difficult to analyze and interpret. We develop and utilize techniques from machine learning to build statistical models that can characterize neural population responses, extract structure from high-dimensional neural data, and quantify neural population dynamics.
ANATOMY: MAPPING BRAIN-WIDE CONNECTIVITY PATTERNS
We use a variety of viral tracing strategies alongside whole-brain anatomical reconstruction to characterize the input/output organization of genetically identified cell-types in the auditory cortex.
