Low amplitude depths and high frequencies SSVEp stimuli for reactive BCI

We have conducted two distinct study to systematically design more comfortable SSVEP stimuli, exploring both reduction of amplitude depth and the use of high frequency stimuli.
The participants did not report any of the exclusion criteria (neurological antecedents, usage of psychoactive medication). The study was approved by the ethics committee of the University of Toulouse (CER approval number  2020-334) and was carried in accordance with the declaration of Helsinki. Participants gave informed written consent prior to the experiment.
Experiment took place in ISAE-Supaero, Toulouse, FRANCE.

# EEG data acquisition

EEG data was recorded from 32 electrodes fitted in an elastic cap according to the 10-20 international system and connected to a LiveAmp amplifier (Brain Products, Munich, Germany). The ground electrode was placed at the Fpz electrode location and all electrodes were referenced to FCz electrode. The electrode impedance were brought below 20kOhm prior recording. The signal was acquired at a rate of 500Hz with a digital band-pass filter ranging from 0.1 to 250Hz. At the onset of every stimulus presentation, an event trigger was generated by the stimulus presentation program and synchronized to the EEG data stream via Lab Streaming Layer(LSL).
Data are in .set/.fdt format from EEGLab.

# High frequencies vs low frequencies

The twelve helathy subjects (mean age=27, SD=6, 8 male and 4 female) underwent a single session that was cut in two separated parts. Firstly, we have presented sequentially and individually stimuli with twelve, low, frequencies from 8 to 30Hz with a range of 2Hz and then another, high, frequencies from 32 to 60 with a 2Hz range (48, 50, 52Hz were excluded because of power line noise). Half of the subjects started the session with low frequencies and the other half with high ones. Each stimulation was preceded by a red circling, visual cue, for 0.5s, it then lasted for 3s, followed by an inter-trial of 0.5s with a self-paced break (space bar press to continue)  between blocks. A block consists of the presentation of the twelve corresponding frequencies of the condition. Each block was repeated 15 times.
Naming of the files follow the convention: P{subject number}_high for high frequencies or P{subject number}_low for the low frequencies.
The data includes all the triggers and the corresponding label. Use for instance

mne.io.read_raw_eeglab

# Amplitude reduction

We have studied amplitude depth reduction for both high and low frequency stimuli. Amplitude depth corresponds to the contrast, i.e. the difference between the two antagonist states of the stimuli. In our case we have reduced the maximum amplitude reachable and kept the minimum the same. For instance, full amplitude depth stimulus oscillates between white (minimum) and black (maximum). With reduced amplitude it oscillates between white (minimum) and gray (maximum that vary).
Twelve healthy individuals (mean age=27, SD=5, 8 male and 4 female) took part in this experiment. We have selected 12, 14, 16 and 18Hz as for the low frequencies condition and 32, 34, 36 and 38Hz frequencies. Each stimulation was preceded by a red circling, visual cue, for 0.5s, it then lasted for 3s, followed by an inter-trial of 0.5s with a self-paced break (space bar press to continue)  between blocks. A block consists of the presentation of the four corresponding frequencies of the condition. Each block was repeated 15 times.
The amplitude depth was manipulated across six levels (100, 80, 60, 50, 40, 30).
The total duration of this experiment was twice as long as the first one and therefore was split in two sessions. The two sessions were separated by a maximum of one week.

Fore more details on the data please contact simon.ladouce@isae-supaero.fr or ludovic.darmet@isae-supaero.fr.