Behavioral Training Procedures for Head-fixed Virtual Reality in Mice

This research explores detailed methodologies for training mice in virtual reality environments, specifically focusing on linear track navigation and complex Y-maze tasks. It aims to address gaps in virtual reality training protocols and assess the ability of mice to perform complex decision-making behaviors within virtual reality. The protocol addresses gaps in detailed virtual reality training methodologies for mice, including the integration of surgical procedures, fluid restriction, and system setup. It also provides a framework for implementing complex behavioral tasks, improving experimental consistency, and facilitating the adoption of virtual reality for nuanced behavioral and cognitive studies in mice.

This VR protocol allows for precise control over experimental conditions and minimizes movement-related noise, enabling detailed behavioral analysis and neurophysiological recordings in head fixed mice. It also supports progressive complexity in tasks, enhancing the study of intricate behaviors and cognitive processes that other methods might not effectively capture.

[Narrator] To begin with, establish baseline weights for the mice one week from the day of the head bar implantation surgery. On days one, two, and three, provide the mice with 15, 10, and five milliliters of water per 100 grams of body mass respectively. Start habituating the mice to the spherical treadmill while also introducing them to fluid regulation to associate the lick tube with the reward through correctly-timed physiological motivation. On the first day, handle the mice for five minutes after weighing them. Gently grasp the head bar implant while they are in their cage to familiarize them with this manipulation. Introduce the mice to the area where the virtual reality system is housed to allow them to anticipate the spatial environment in which experimental trials will occur. On the second day, handle the mice again for five minutes. Affix the head bar to the holder and allow the mice to familiarize themselves with the spherical treadmill for 5 to 20 minutes, either on an infinitely repeating track or without the software program activated. On the third day, after handling the mice for five minutes, securely attach them to the air-cushioned spherical treadmill and introduce them to liquid rewards through the reward tube. Before positioning the mouse, extend the centered reward tube with a small droplet of reward at the tip. Elevate the reward tube 5 to 15 millimeters above the spherical treadmill so that licking the spout requires a natural forward-facing posture of the head. To head fix the mouse, place the mouse on the handler's dominant side of the spherical treadmill. Using the handler's dominant hand, pull the mouse by its head bar towards the head fixing platform. Then place the head bar in the slot meant for fixing and using the handler's non-dominant hand, click the head bar into place. Now line the midsagittal plane of the mouse with the center of the spherical treadmill. Ensure that the mouse's hind paws are no more than 11 centimeters from the apex of the spherical treadmill, and that the head is behind the apex. Confirm that all four paws are touching the treadmill and that the abdomen can touch the treadmill when the mouse is at rest. Also, ensure that any side preference observed in the mouse is not due to asymmetry in how the animal is mounted on the ball. To guide the mouse towards the extended lick tube, use the kiss-it method where the mouse is gently maneuvered until its mouth almost touches the tip of the spout. When the mice receive rewards, set the duration of the extended reward tube to one second. Once the mice have acclimated to the experimental paradigm, perform a daily 30-minute session where the mice move along a linear virtual corridor starting with a one meter length. Upon reaching the end of the corridor and receiving the sugar droplet reward, teleport the mice back to the starting point. Document daily recordings of timestamped data concerning reward retrieval and the distance covered by the mice on the spherical treadmill for further analysis. In the Y-maze paradigm, ensure the mice navigate towards a choice point where two arms extend out at a 45-degree angle in either direction, forming a Y shape. Deactivate rotation from the starting point of the maze until reaching the choice point, having two arms of varying visual features. Then activate the rotation within the decision zone to allow the mouse to pivot toward its desired direction. Upon entry into the arm leading to the reward zone, deactivate the rotation once again. Train the mice to navigate towards the black arm to obtain a sugar reward, with each trial concluding with the mice being teleported back to the starting location. Randomly shuffle the reward location between the left and right sides to ensure that mice associate the reward with visual cues rather than a specific side. The number of rewards acquired by the mice increased progressively with learning across the different track lengths in the VR task, with faster speeds observed as track length increased. A noticeable increase in mean velocity was observed as the track length increased. The Y-maze showed above chance performance over days. However, only a subset successfully progressed to the more difficult discrimination stages.