Fully autonomous mouse behavioral and optogenetic experiments in home-cage

Abstract

SummaryGoal-directed behaviors involve distributed brain networks. The small size of the mouse brain makes it amenable to manipulations of neural activity dispersed across brain areas, but existing optogenetic methods serially test a few brain regions at a time, which slows comprehensive mapping of distributed networks. Laborious operant conditioning training required for most experimental paradigms exacerbates this bottleneck. We present an autonomous workflow to survey the involvement of brain regions at scale during operant behaviors in mice. Naïve mice living in a home-cage system learned voluntary head-fixation (>1 hour/day) and performed difficult decision-making tasks, including contingency reversals, for 2 months without human supervision. We incorporated an optogenetic approach to manipulate activity in deep brain regions through intact skull during home-cage behavior. To demonstrate the utility of this approach, we tested dozens of mice in parallel unsupervised optogenetic experiments, revealing multiple regions in cortex, striatum, and superior colliculus involved in tactile decision-making.

A miniature kinematic coupling device for mouse head fixation

“…Other existing head-fixation systems made by laboratories (Andermann et al, 2010, Guo et al, 2014, Hao et al, 2021 or by companies (Neurotar, personal correspondence) achieve micrometer-scale precision or larger. These head-fixation systems implement headplates with non-kinematic (such as threaded holes and screws) or quasi-kinematic methods to achieve such precision, and thus they still suffer from micrometer-scale z-axis fluctuations, which may introduce false positive (calcium independent) transients.…”

“…In addition, a promising next step is incorporation of these headplates into a system for mouse voluntary head restraint. Recently several groups have reported systems to train mice in voluntary head restraint (Aoki et al, 2017, Murphy et al 2020, Hao et al, 2021. In these systems mice show robust self-initiation fixations over months and can perform decision making tasks during fixation (Murphy et al 2020, Hao et al, 2021.…”

“…Recently several groups have reported systems to train mice in voluntary head restraint (Aoki et al, 2017, Murphy et al 2020, Hao et al, 2021. In these systems mice show robust self-initiation fixations over months and can perform decision making tasks during fixation (Murphy et al 2020, Hao et al, 2021. Such systems allow precise presentation of sensory stimuli and provide the stability needed for optogenetic stimulation and mesoscale optical imaging using widefield microscopes.…”

“…Reports that rats could be trained to self head-fix using computer-controlled training systems (Kampff et al, 2010) inspired a resurgence of technological advancements for voluntary head restraint. Since then, several systems that involve voluntary head fixation in mice have been developed (Murphy et al, 2016, Aoki et al, 2017, Murphy et al 2020, Hao et al, 2021, however to our knowledge none have demonstrated micron-scale repositioning accuracy necessary for cellular resolution imaging.…”

MesoNet allows automated scaling and segmentation of mouse mesoscale cortical maps using machine learning

“…Automatic registration and segmentation of brain imaging data can greatly improve the speed and precision of data analysis and do not require an expert anatomist. This is particularly crucial when using high-throughput neuroimaging approaches, such as automated mesoscale mouse imaging [15][16][17][18] , where the amount of data generated greatly exceeds the capacity of manual segmentation.…”