Mu and Beta Power Effects

1. Introduction

This repository belongs to the codes for analysis and generating plots for:

Kamali S, Baroni F, Varona P. Mu and beta power effects of fast response trait double dissociate during precue and movement execution in the sensorimotor cortex. bioRxiv. 2024:2024-11. doi: 10.1101/2024.11.11.621252

Using any of these codes or materials requires the authors’ permission. To refer to the results, please cite the work as listed above. All the codes are in Matlab.

---

2. Dataset

The dataset used in this study belongs to a stereotypical finger-pinching task and is publicly available at: doi:10.1093/gigascience/gix034. Figure 1 in the paper depicts the execution steps and timeline.

---

3. Preprocessing Pipeline

The codes to perform the preprocessing pipeline, shown in Figure 2 are:

• Preprocessing shared with EEG and EMG: preprocessing.m
• Import channel locations: channel_locator.m
• Function to enter events field: event_struct.m
• Find the event where hand change from right to left happens: detect_hand_change.m
• Compute time-frequency decomposition of EMG single trials: EMG_ersp_singletrl.m
• Detect EMG onset latency: emg_onset_detector.m
• Select brain and eye dipoles to include in the study: dipfit_criteria.m

---

4. Clustering ICs

To generate the STUDY for group-level analysis and perform clustering on brain dipoles and get the results in Figures 3 and 4, use the following codes:

• Create STUDY: study_generator.m
• Import the STUDY into EEGlab GUI. Set the parameters for the ERSP, ERP, and spectrum fields manually.
• Perform the pre-clustering calculations by adjusting the weights as recommended in the paper’s section 2.5 and select the ICs as recommended.
• Get the subjects and ICs’ indexes from the study: cluster_info_gen.m

---

5. Classification Based on Latency

To form the trait and state fast and slow groups based on latencies and generate Figure 5 and Tables 1 and 2, use the following codes:

• Find the fast and slow subject and trials: find_fast_slow_subs.m
• Make a plot of the mean latencies for all the subjects and the violin plot for the latencies of the single trials both for trait and state: emg_onset_data_fast_slow_subtrls.m
• Find the fast and slow subjects and trials to extract the time-frequency of single trials for each brain area: fast_slow_subtrials_extract_for_t_test_analysis.m
• Do the same for the EMG data: fast_slow_subtrl_EMG_extract_for_ttest_analysis.m

---

6. Statistical Tests

To perform cluster-based permutation t-tests and FDR correction and generate results presented in Figures 6 to 8, use the following codes:

• Function to perform cluster-based permutation t-test: permutation_test_on_clusters.m
• Cluster-based test and FDR correction for EEG data over brain areas: cluster_based_permutation_test_fast_slow_subjects_trials.m
• Cluster-based test and FDR correction for EMG data: cluster_based_permutation_test_fast_slow_subjects_trials_EMG.m
• Function to plot the IQR shades: shade.m