Imaging whole-brain activity to understand behavior

Abstract

The brain evolved to produce behaviors that help an animal inhabit the natural world. During natural behaviors, the brain is engaged in many levels of activity from the detection of sensory inputs to decision-making to motor planning and execution. To date, most brain studies have focused on small numbers of neurons that interact in limited circuits. This allows analyzing individual computations or steps of neural processing. During behavior, however, brain activity must integrate multiple circuits in different brain regions. The activities of different brain regions are not isolated, but may be contingent on one another. Coordinated and concurrent activity within and across brain areas is organized by (1) sensory information from the environment, (2) the animal's internal behavioral state, and (3) recurrent networks of synaptic and non-synaptic connectivity. Whole-brain recording with cellular resolution provides a new opportunity to dissect the neural basis of behavior, but whole-brain activity is also mutually contingent on behavior itself. This is especially true for natural behaviors like navigation, mating, or hunting, which require dynamic interaction between the animal, its environment, and other animals. In such behaviors, the sensory experience of an unrestrained animal is actively shaped by its movements and decisions. Many of the signaling and feedback pathways that an animal uses to guide behavior only occur in freely moving animals. Recent technological advances have enabled whole-brain recording in small behaving animals including nematodes, flies, and zebrafish. These whole-brain experiments capture neural activity with cellular resolution spanning sensory, decision-making, and motor circuits, and thereby demand new theoretical approaches that integrate brain dynamics with behavioral dynamics. Here, we review the experimental and theoretical methods that are being employed to understand animal behavior and whole-brain activity, and the opportunities for physics to contribute to this emerging field of systems neuroscience.

Figure 1:. Recording from the brains of behaving animals.

(A) A C. elegans worm crawls freely on a motorized stage. A low-magnification 10x objective captures the animal’s entire body to record posture and behavior, while a high magnification 40x objective records calcium activity from the animal’s brain. Real-time feedback keeps the animal in the objectives’ field of view. Adapted from Nguyen et al., 2016. (B) A larval zebrafish swims in a thin, water-filled nchamber. A high-speed, low-magnification optical setup tracks the animal’s motion, while a high-resolution light-field setup records whole-brain calcium activity.Real-time feedback in all three dimensions keeps the animal’s brain in the field of view. Adapted from Cong et al., 2017. (C) An adult fruit fly is tethered and placed on an air-cushioned ball. A high-resolution objective allows for two-photon excitation and recording of calcium activity from the animal’s head. The fly is free to walk in any direction on the ball, with low resolution cameras recording the animal’s posture and behavior. Visual and auditory stimuli are presented to the animal while it is on the ball. Adapted from Seelig et al., 2010.

Figure 2:. Samples of pan-neuronal recordings in behaving animals.

(A) A sample image of the brain of C. elegans (left), labeled with pan-neuronal cytosolic GCaMP6s and nuclear-localized Tag-RFP. Normalized activity traces (right) of 84 neurons in a freely crawling worm. Adapted from Venkatachalam et al., 2015. (B) Top and side views of the brain of a larval zebrafish (left), labeled with GCaMP6f. Activity of segmented neurons in the brain (right) during fictive swim behavior. Adapted from Mu et al., 2019. (C) On the left, a schematic of volumetric imaging of the brain of an adult Drosophila being presented with auditory stimuli. The brain is labeled with GCaMP6s and tdTomato. To the right are responses from recorded regions of interest (ROIs) to auditory stimuli. Adapted from Pacheco et al., 2021. AL note: we may want to reach out to the authors so we can plot higher-resolution neuron trace plots in a consistent color scheme

Figure 3:. Computational methods for neural and behavioral analysis.

Top Left: The first challenge is to develop statistical methods to extract biological signals of interest from raw data. For example, extracting the times of action potentials (i.e. “spikes”) from extracelluar voltage recordings, demixing and deconvolving calcium fluorescence traces, or tracking body parts in videos. Top Right: Computational models for exploratory analysis aim to reveal simplifying structure in high dimensional signals, like repeated sequences of spikes, low dimensional trajectories of neural activity, or clusters of stereotyped behaviors. Bottom left: Top-down analyses hypothesize an algorithm and circuit implementation to solve a computational problem, like tracking heading given visual inputs and proprioceptive feedback. The model makes predictions about neural activity that can be tested against measured data. Bottom right: Rather than hand-tuning an algorithm and circuit, task-based modeling learns a circuit to solve a particular computation by minimizing a loss function. This relatively new approach offers an indirect way of making testable predictions of neural activity.