ReachingBot: An automated and scalable benchtop device for highly parallel Single Pellet Reach-and-Grasp training and assessment in mice.

Background: The single pellet reaching and grasp (SPRG) task is a behavioural assay widely used to study motor learning, control and recovery after nervous system injury in animals. The manual training and assessment of the SPRG is labour intensive and time consuming and has led to the development of multiple devices which automate the SPRG task.

New Method: Here, using robotics, computer vision, and machine learning analysis of videos, we describe a device that can be left unattended, presents pellets to mice, and, using two supervised learning algorithms, classifies the outcome of each trial with an accuracy of greater than 94% without the use of graphical processing units (GPUs). Our devices can also be operated using our cross-platform Graphical User Interface (GUI).

Results: We show that these devices train and assess mice in parallel. 21 out of 30 mice retrieved > 40% of pellets successfully following the training period. Following ischaemic stroke; some mice showed large persistent deficits whilst others showed only transient deficits. This highlights the heterogeneity in reaching outcomes following stroke.

Comparison With Existing Method(s): Current state-of-the-art desktop methods either still require supervision, manual classification of trial outcome, or expensive locally-installed hardware such as graphical processing units (GPUs).

Conclusions: ReachingBots successfully automated SPRG training and assessment and revealed the heterogeneity in reaching outcomes following stroke. We conjecture that reach-and-grasp is represented in motor cortex bilaterally but with greater asymmetry in some mice than in others.