What features of odorants and their neuronal representations are relevant for olfactory percepts?

The relationship between perceived odor quality, the underlying spatial-temporal patterns of activity in the brain and the physical space of odors remains elusive. To determine what features in the glomerular input activity patterns are captured by the brain, we use patterned illumination in mice engaged in discrimination tasks. Odors are replaced with tightly controlled and flexible patterns of light. Employing these artificial optogenetic stimuli informs on the detection thresholds and resolution limits of the system. By combining odors with precise optogenetic manipulations, we test our predictions against olfactory behaviors.

Reformatting and decoding odor information across the brain

How is information relating to odor intensity, identity, position and valence, all entangled at the level of glomerular inputs disambiguated in the brain, ultimately leading to specific, context dependent actions? Using patterned illumination, optical imaging and recording techniques, we study the integration rules between olfactory bulb and its targets, and observe how information is routed and made use of. To probe the function of specific circuit elements in the bulb and the downstream cortical areas, we use targeted cell ablation and optogenetic strategies. Further, we capitalize on innovative DNA barcoding strategies for neuronal labeling to obtain connectivity statistics and determine the functional diversity of bulbar and piriform cortex outputs.

Top-down cortical feedback

A flurry of olfactory cortical areas send feedback to the sensory periphery, vastly outnumbering the feedforward input from the nose. Yet, to date, the behavioral significance and neuronal correlates of these feedback signals are largely unknown. We study what roles these distinct feedback projections play in mice that actively explore changing olfactory scenes. While testing computational scenarios related to predictive coding and invariant perception, we keep our eyes open for the unexpected.

Selective attention

Sensory environments are complex and rich in information. Processing all the inputs – relevant and irrelevant – would vastly overwhelm our sensory capacity and overall remain quite inefficient. Animals instead employ different mechanisms to selectively process information only from the source of interest and ignore irrelevant information. This process, loosely termed selective attention, though experienced in our daily lives (i.e. cocktail party problem), remains poorly understood. The rodent olfactory system offers a suitable model for understanding selective attention: rodents display excellent behavioral acuity for tracking down intermittent odor plumes, despite the presence of several distractor odor sources in the environment.

3D Optical monitoring and manipulation of neuronal circuits

Understanding the neuronal underpinnings of behaviors requires closed loop monitoring large populations of neurons, and simultaneous perturbation of specific subsets. Patterned illumination enables generation of flexible spatial-temporal photo-stimulation profiles, which together with multiphoton imaging and optogenetic strategies provide a robust framework to achieve this goal. Towards this end, we implement closed loop digital micro-mirror device (DMD)-based and phase modulation (digital holography) approaches.