Visual Behavior

We have used large-scale physiological recordings in behaving animals to characterize how sensation and behavior are encoded in activity across the visual cortex and how these representations are influenced by behavior state, expectation, and experience during an image change detection task. Just like the Visual Coding project, the Visual Behavior project leverages the Allen Brain Observatories to collect highly standardized datasets.

This dataset can be used to address questions such as: How do different types of neurons in the brain encode sensory and behavioral information? Are these representations flexible depending on motivation or expectation? How are interactions across cortical areas and depths modified by experience?

Visual Change Detection Task

The Visual Behavior Optical Physiology and Visual Behavior Neuropixels projects are built upon a change detection behavioral task. Briefly, in this go/no-go task, mice are shown a continuous series of briefly presented visual images and they earn water rewards by correctly reporting when the identity of the image changes (diagrammed below). Five percent of images are omitted, allowing for analysis of expectation signals.

2-Photon Imaging

Using single- and multi-plane 2-photon calcium imaging in the visual cortex of transgenic mice expressing the calcium indicator GCaMP6f in populations of excitatory and inhibitory neurons, we have recorded neural activity during performance of a visual change detection task from 50,482 neurons in 704 in vivo imaging sessions. A key aspect of the experimental design is the repeated imaging of the same populations of neurons across multiple days, allowing analysis of single cell changes across behavioral and sensory conditions, including task engagement and stimulus novelty.

Neuropixels Recordings

We used Neuropixels probes to measure spiking activity of neuronal populations distributed across multiple visual cortical regions, in addition to subcortical structures such as the thalamus, hippocampus and midbrain, while mice performed the change detection task. Overall, this dataset includes ~200,000 recorded neurons (units) from 153 experimental sessions. The simultaneous recording of activity across multiple visual areas permits analysis of inter-regional interactions and signal flow during visually guided behavior. In addition, each experimental session includes a passive stimulus replay block that allows investigation of task-dependent modulation sensory and behavioral coding.