Fig. 8: Changes in the prediction accuracy of the initial pole movement in individual PMdc and M1 neurons during TTR training sessions.

a Top, representative images of the same FOV in sessions 11 (early) and 22 (late) of marmoset 4. Bottom, expanded images of the boxed regions shown in the top image. b Trial-averaged activity of the arrowed neurons in (a) for pull (red) and push (blue) trials. Bottom, averaged traces of Y-axial pole trajectory. Data are presented as mean ± SEM. c The cvR² of all pursued PMdc and M1 neurons for pull and push trials in early and late sessions (n = 359 pursued neurons in PMdc and 336 pursued neurons in M1). Black horizontal bars indicate the pursued neurons that were movement-unrelated. d Proportions of the increase, decrease, unchanged, and movement-unrelated neurons in PMdc and M1 for pull and push trials. Numbers in parentheses represent the numbers of neurons. The cvR² of the pursued neurons for pull and push trials in each imaging field of PMdc (e) and M1 (f) in early and late sessions. Data are presented as mean ± SEM. *P < 0.05, Wilcoxon signed-rank test, two-sided. For pull trials, the cvR² of pursued neurons that were pull-related in the early sessions and the cvR² of those that were pull-related in the late sessions were used, while for push trials, the cvR² of pursued neurons that were push-related in the early sessions and the cvR² of those that were push-related in the late sessions were used. The numbers of neurons that were used and the imaging field # (Supplementary Fig. 3b and Supplementary Table 1) are also shown. The cvR² of the population decoders in PMdc (g) and M1 (h). Each number indicates the imaging field #. The numbers in parentheses represent the numbers of neurons used for the decoder. The number of neurons used for the decoder was the minimum number of neurons across the early and late sessions for each imaging field. *P < 0.05, permutation test, two-sided. Source data are provided as the Source Data file.