Figure 4
Disentangling cortical signals contribution to steering decoding. (A) Steering decoding correlations using EEG data with progressively longer anticipation, with the LLR window set to 1.5 s. Results indicate decoding correlations when using the EEG before and after removing motion-related components (EEG and EEGden respectively). Error-bars indicate the SD across participants. The grey shaded area indicates the baseline prediction correlation (95th percentile of the null distribution). The plot on the bottom reports the effect-size for EEGden > baseline. (B) The steering signal was prediction by removing selectively each potential non-brain signal. The LLR window was set at 1.5 s. Error-bars indicate the SD across participants. The grey shaded area indicates the baseline prediction correlation (95th percentile of the null distribution). The plot on the bottom reports the effect-size for EEG > EEGNO-SIGNAL for each signal of interest.
