Abstract
Objective Brain-machine interfaces (BMIs) are promising candidates for restoring the lost motor system functions. Center-out reaching task is a commonly used BMI control paradigm in humans and monkeys. In this work, our goal was to develop a behavioral paradigm which enables rats to control a neuroprosthesis in a center-out reaching task applied in one-dimensional space.
Approach The experimental setup mainly consisted of a behavioral cage and a robotic workspace outside the cage. Two distant targets were located on the left and right sides of the central starting position of the robot endpoint. An online transform algorithm was used to convert the activity of a pair of recorded primary motor cortex units into two robotic actions. An increase in the activity of one of the units directed the robot endpoint toward left while an increase in the other moved it toward right. The task difficulty level which was proportional to the distance between the selected target and the initial position of the robot endpoint at the beginning of trials was increased gradually as the rat adapts with the transform.
Main Results All three rats involved in the study were capable of achieving randomly selected targets with at least 78% accuracy in the highest task difficulty level, in center-out reaching task. A total of 9 out of 16 pairs of units examined were eligible for training in center-out reaching task. Two out of three rats were capable of reversal learning where the mapping between the activity of the unit pairs and the robotic actions were reversed.
Significance The present behavioral paradigm and experimental setup may be used to study the neural mechanisms involved in neuroprosthetic control. Using the present approach the performance of BMI decoders may also be assessed for one-dimensional center-out reaching task.
Competing Interest Statement
The authors have declared no competing interest.