Proceedings of The Physiological Society

Europhysiology 2018 (London, UK) (2018) Proc Physiol Soc 41, C075

Oral Communications

Analysis of motor learning in head fixed mice during voltage imaging

J. Jimenez-Martin1, D. Potapov1, R. M. Empson1

1. Physiology, University of Otago, Dunedin, Otago, New Zealand.


Acquisition of complex movement sequences is termed motor skill learning1. The motor cortex provides a substrate for adaptive alterations during the achievement of these skills2. Voltage imaging techniques enable monitoring of electrical activity from a large population of Layer 2/3 motor cortex neurons in awake behaving animals3 and correlates neural activity with behaviour during learning4. Here we aim to standardize a motor learning behavioural test (lever pulling task) for head fixed conditions in order to determine changes in Layer 2/3 connectivity during motor learning. We developed a novel self-initiated lever-pull task and monitored the mouse cortex in vivo using wide field voltage imaging. Twelve Rasgrf 2-2AdCre;Camk2a-ttA;Ai78 (TITL-VSFP-B) mice were used, aged 3 to 4 months. Mice were anaesthetised with ketamine (150mg/Kg) and domitor (1.5mg/Kg), the skull was thinned and a head holder frame was attached to the skull using dental cement. After 15 days of recovery, mice were water restricted (body weight was maintained >85% of the initial) for 5 days prior to training. Head-restrained mice were trained to perform a lever pull task, a four μl water drop was provided as a reward when the lever was pulled beyond 3o positional threshold. Mice preformed 16 trials (10 seconds each) per day over 3 days. Different parameters were recorded using a custom made interface, variables are reported as means ± S.E.M., compared by Friedman test and Dunn's multiple comparison test. Four mice obtained less than one reward per trial on day 3 of training and were excluded. Mice obtained more water rewards on day 3 of training (294±28) compared with day 2 (177±39) and day 1 (189±43) (p<0.05). Out of the 16 trials mice performed, water rewards were obtained in 69.53±9.98%, 70.31±9.13% and 80.47±5.83% of the trials for days 1, 2 and 3, respectively; significant differences were not detected. Mice learned to associate pulling the lever with a water reward, as the total pulling time per trial on day 3 (5.42±0.73s) was higher than day 2 (3.82±1.01s) (p<0.05) and day 1 (2.93±0.63s) (p<0.01). Conversely, the total push time (no reward) per trial decreased on day 3 (1.46±0.56s) compared with day 2 (2.45±1.009s) (p≥0.05) and day 1 (2.93±0.63s) (p<0.05). Mice also increased the amount of time the lever crossed the reward threshold per trial on day 3 (3.03±0.73s) compared with day 2 (1.85±0.81s) (p≥0.05) and day 1 (1.01±0.35s) (p<0.01). Mice improved their movement accuracy by decreasing the pulling range and keeping the lever close to the reward threshold position. Thus, mice successfully learned to pull a lever and refine motor movements to obtain greater amounts of water over 3 days of training. Our results indicate that this lever pulling test measures motor skill learning in mice that can be integrated with mesoscopic voltage imaging to generate connectivity maps during motor learning.

Where applicable, experiments conform with Society ethical requirements