Fig. 7: Illustration demonstrating how pseudo-linear summation explains the changes of population tuning for a finger and its dependence on the strength of neural representation.

Points indicate neural activity corresponding to different gestures, with gray (black) indicating the gestures with flexion (extension) of a particular finger (finger 1). The red line joins a pair of gestures when finger 1 moves alone (i.e., other fingers are at rest), and indicates the population tuning vector during isolated movement. The contribution of the movement of other fingers is indicated by displacement along the black arrow. The displaced population tuning vector under linear summation is indicated in orange. This results in preserved population tuning magnitude and direction (red and orange lines are parallel and of the same length). However, the magnitude of neural activity in biological neural circuits is limited (indicated by the black circle). Hence, the experimentally observed neural activity differs from the prediction of linear summation, and the observed population tuning vector (blue) shows a change in magnitude and direction compared to the neural tuning vector for the finger movement in isolation (red). When the relative contribution from other fingers is large (such as for the index and middle fingers), the changes in population tuning magnitude and direction are greater.