Research

Our lab studies how the brain recognizes and remembers odors. The brain’s olfactory system, particularly the piriform cortex, helps identify different odors. We conducted electrophysiological recordings of single units over several weeks to study the stability of odor responses in the mouse piriform cortex. While the piriform cortex could distinguish between odors at any given time, the odor responses changed over days to weeks. Even training the brain or exposing it to the same smell daily didn’t fully stop this change. This drifting response raises interesting questions about how the brain processes odors and whether this “instability” might also be seen in other parts of the brain. Our work helps us understand the complex ways in which the brain interprets the world around us.

Learning is understood to constitute experience-dependent synaptic plasticity, but the difficulty of measuring synaptic connectivity in vivo has limited our understanding of this process. My lab employs methodology I have developed to overcome this challenge, with the aim of discovering plasticity rules7 that govern learning. We ask how changes in synaptic connectivity—captured in a living, learning brain—depend on neuronal activity and experience, in the context of an intact circuit, with knowledge of the animal’s stimulus history and behavior.

To study this plasticity, I developed an approach to reliably (<5% error rate) infer synaptic connectivity in vivo from spiking. This is possible because Neuropixels 2.0 probes isolate sufficient numbers of pairs (~106/animal) to infer connectivity in the ultra-sparse piriform network (~0.1% connected), where common input is minimized.

A deep convolutional network was trained to identify pairs of piriform neurons whose correlograms had known features of monosynaptic connectivity: a sharp asymmetric peak at short latency with a fast rise and slower decay. Validation of this approach using ground truth data25 and a spiking neural network, both with known connectivity, showed that inferred synapse are >95% likely to correspond to true connections with ~50% of all connections recovered.                                                                                                                            

We developed a behavioral paradigm to study volitional olfactory investigation in mice over several months. We placed odor ports in the wall of a standard cage that administer a neutral odorant stimulus when a mouse pokes its nose inside. Even though animals were fed and watered ad libitum, and sampling from the port elicited no outcome other than the delivery of an odor, mice readily sampled these stimuli hundreds of times per day.

This self-paced olfactory investigation persisted for weeks with only modest habituation following the first day of exposure to a given set of odorants. If an unexpected odorant stimulus was administered at the port, the sampling rate increased transiently (in the first 20 min) by an order of magnitude and remained higher than baseline throughout the subsequent day, indicating learned implicit knowledge. Thus, this system may be used to study naturalistic olfactory learning over extended time scales outside of conventional task structures.

Videos for odor sampling.

We’ve developed a novel behavioral approach called the Virtual Burrow Assay to study decision-making in mice. Mice are placed in a situation similar to being at the entrance of their burrow, deciding whether to go out or stay inside. This setup allows us to measure the mice’s reactions, like approach or avoidance, in just milliseconds. The best part is that it works with standard methods for recording brain activity, and the mice don’t need any training. With this assay, we can observe how mice get used to new things, tell different cues apart, and respond to things they find scary or unpleasant. This helps us understand how the brain processes these behaviors.

The Virtual Burrow Assay consists of a tube enclosure (virtual burrow), constrained to slide back and forth along the anterior-posterior axis of the body of a head-fixed mouse. When placed inside the virtual burrow, mice invariably pull the tube up around themselves as far as possible and remain within the enclosure until a linear actuator pulls the virtual burrow back. After initial resistance (< 30 sec), mice voluntarily maintain this exposed position ('egress'). The assay measures the position of the burrow on a millisecond timescale and detects the precise timing of retreat into the burrow ('ingress'). A laser displacement sensor measures burrow position and a force sensor measures the force generated by the animal when pulling via the tether against a linear actuator.