Appendix


A. Auditory Oddball Paradigm

The Auditory Oddball Paradigm (AOD) consisted of detecting an infrequent target sound within a series of frequent regular sounds. The stimuli were presented via headphones, and participants were asked to respond as quickly as possible by pushing a mouse button with their right index finger. The standard stimulus was a center-panned 500 Hz tone, the target stimuli were left or right-panned location deviants, respectively. Targets occurred with a probability of 0.20, 0.10 for each location; the standard stimuli occurred with a probability of 0.80. Stimulus duration was 75 ms and the inter-stimulus interval was 700 ms, with no immediate repetitions of targets. In the example data, the epochs contain a sequence of standard – target – standard sounds, to dissociate between obligatory stimulus related and target related components, respectively. All stimuli were presented at approximately 65 decibels above threshold. Response times ranged from 250-500 ms, the overall accuracy was larger than 98%. Participants - 32 healthy right-handed undergraduates were recruited, and took part in the experiment after providing a written statement of informed consent. Participants were sitting in a comfortable recliner in an electro-magnetically shielded and sound-attenuated testing chamber (Rainford EMC Systems, Wigan, UK). Participants were fitted with 63 Ag/AgCl scalp electrodes mounted in an elastic cap (EasiCap, Falk Minow Services, Breitenbrunn, Germany) according to the 10-20 System: FP1, FP2, F7, F3, Fz, F4, F8, T7, C3, Cz, C4, T8, P7, P3, Pz, P4, P8, O1, OZ, O2, with additional intermediate sites at Fpz, AF7, AF3, AFz, AF4, AF8, FT7, FC5, FC3, FC1, FCz, FC2, FC4, FC6, FT8, C5, C1, C2, C6, TP7, CP5, CP3, CP1, CPz, CP2, CP4, CP6, TP8, P5, P1, P3, P6, PO7, PO3, POz, PO4, PO8, including mastoid sites TP9 and TP10. All channels were referenced to the nose, with a ground electrode on the cheek, and impedances were kept below 10kΩ. Vertical eye movement (EOG) was acquired from a bipolar derivation between Fp1 and an additional electrode placed below the left eye, horizontal EOG was acquired from a F7-F8 bipolar montage. EEGs were recorded continuously at 500 Hz sampling frequency with a band-pass from 0.01-250 Hz with BrainAmp DC amplifiers (BrainProducts, Munich, Germany). EEGs were down-sampled to 250 Hz, filtered with a zero-phase Butterworth filter from 1-45 Hz (24 db per octave), and re-referenced to common average reference. The data were then segmented from -800 to 1200 ms around target stimuli and baseline activity was defined from -300 to 0 ms around stimulus onset, and trials with amplitudes exceeding ±150µV on any of the channels were excluded from further analysis. Concatenated single sweeps around target onset, including the previous and the following standards were subjected to single subject independent component analysis (ICA, Infomax algorithm), implemented in EEGLAB ([2]) running in MATLAB. Components with topographies attributable to eye movement and frontotemporal muscle activity were identified and removed from the data ([4]). For each dataset, we kept 20 components. Missing trials were padded with the mean from surrounding trials. Hereafter, single-trials were additionally denoised with a wavelet filter ([5]). Wavelet coefficients used for the reconstruction of the single trials were selected on the basis of the grand-averaged response, and were the same for all electrode sites and participants. Hereafter, group ICA was computed, estimating 20 PC/ICs.

B. Defaults

EEG defaults are in 'icatb_defaults.m' file. In GIFT manual and walk-through we explained necessary variables required for analysis and display. We explain below EEG specific variables used in EEGIFT:

EEG_RMBASE
Vector used for removing baseline from EEG signal during smoothing trials. The default value is [-100 0].
EEG_IMAGE_VALUES
Image values used for displaying stacked EEG signal. The options available are 'Positive and Negative', 'Positive', 'Absolute' and 'Negative'. The default value is 'Positive and Negative'.
EEG_CONVERT_Z
Option is provided to convert stacked EEG image to Z-scores during display. The default value is 'Yes'.
EEG_THRESHOLD_VALUE
Threshold used for displaying stacked EEG signal. The default value is '1.0'.
EEG_IMAGES_PER_FIGURE
The number of components per figure that can be displayed. The options available are '1', '4' and '9'. The default value is '4'.
EEG_TOPOPLOT_COLORMAP
Color map used for topography plot. The default value is 'jet(64)'.

C. Output Files