Synamp

During the experimental session, children sat comfortably in an electrically shielded dimly lit room in front of a 24″ monitor that displayed the stimuli. We recorded EEG (band pass DC-200 Hz, sampling rate 1000 Hz) from 30 scalp positions referenced to earlobes with the ground at FzA, in accordance with the 10/20 system with the amplifier SynAmps, NeuroScan, (Compumedics, El Paso, TX, USA) (Figure 1B). Electrode impedances were kept under 5 kOhm. We also recorded the electrooculogram (same bandpass and sampling rate as for EEG) to exclude trials contaminated with eye movements for further analysis. Notch filter was set at 60 Hz. Behavioral data under Navon’s paradigm were recorded in parallel with the electrophysiological data. Data were stored and analyzed offline.

As explained in Horwitz et al. (2017a) (link), we used a soft elastic cap with 64 surface Ag/AgCl electrodes (Quick-Cap, Compumedics), placed the electrodes according to the international 10–20 system, and connected the cap to a bioamplifier (SynAmps, Compumedics). We used Neuroscan (Curry7.4 NeuroScan, Compumedics) for signals recording and data processing. The ground electrode is incorporated over the midline frontal region of the cap and the reference was placed between the Cz and Pz-electrodes by the manufacturer. In the data analysis, the signals were re-referenced to bimastoid electrodes (M1, M2).
We kept electrode impedance below 5 kΩ. Furthermore, we registered horizontal and vertical electro-oculographic (EOG) data with two bipolar channels. Moreover, we placed two EKG-electrodes in the heart axis on the chest and two electrodes in submental positions to record EMG-artifacts. We sampled the EEG measurements at 2 kHz with an analog, antialiasing RC low-pass filter at 800 Hz and digitally high-pass and low-pass bandpass filtered the EEG-data offline with a Hann function filter at 0.5 and 250 Hz with a tapering window at 10%. We did not apply any notch filters.

EEG acquisition and analyses followed the same procedures in E1 and E2. EEG data were collected using SynAmps amplifiers and Neuroscan software (Compumedics). Electrode placement followed the international 10-10 system, using a 61-electrode montage. Data were referenced to the left mastoid at acquisition. Measurements of vertical and horizontal EOG were included for off-line independent component analysis (ICA) correction. Data were digitized at 1,000 Hz.
EEG data were analyzed in MATLAB, using the FieldTrip toolbox (Oostenveld et al., 2011 (link)). Data were rereferenced to the average of both mastoids and downsampled to 250 Hz. ICA was performed to identify components associated with blinking or lateral eye movements, and bad components were removed from the data. After ICA correction, trials with exceptional variance were identified based on visual inspection (ft_rejectvisual with the “summary” method) and excluded. Noisy trials were identified without knowledge of the relevant conditions to which particular trials belonged. We additionally excluded trials in which the decision time (the time between cue onset and report onset) was below 200 or above 2,000 ms. In E1, on average 1,027 ± 25 trials (86 ± 2%) were retained for analysis. In E2, on average 700 ± 11 trials (87 ± 1%) were retained for analysis.

We placed 19 disk-type Ag/AgCl electrodes on the scalp of each participant, in accordance with the international 10/20 system37 . Reference and ground electrodes were placed on the left ear and forehead, respectively. Vertical and horizontal electrooculograms were also recorded. Signals were amplified and band limited using differential amplifiers and band-pass filters (SynAmps, NeuroScan, Charlotte, USA). Gains were 60 dB and cut-off frequencies were 0.1 and 100 Hz. The input impedance of the amplifiers was 10 MΩ. Signals were digitized using a 16-bit analogue-to-digital converter at 500 Hz and stored on a hard disk.

Electroencephalogram signals were recorded from 64 scalp sites according to the International 10–20 system with Ag/AgCl electrodes and AC Amplifier (Synamps, Neuroscan Inc.). The vertical and horizontal electrooculogram (VEOG and HEOG) sites were used to monitor the horizontal eye movements and eyeblinks, and the electrode impedance was kept below 5 kΩ during recoding. EEG signals were acquired by referenced to the left mastoid online and then were recomputed offline against to the right mastoid.