Eight Categories of Chinese Nouns: an EEG-based semantic-decision experiment
Description
The EEG experiment conducted used a semantic-decision task. The words presented to participants were nouns (written in simplified Chinese characters) belonging to eight categories: four containing nouns referring to living entities and four containing nouns referring to nonliving entities. Each category included 30 nouns, making for a total of 240. The nonliving stimuli were chosen from the following categories: clothing, part of a house, tools, and vehicles ; the living stimuli were chosen from fruits/vegetables, animals, body parts, and people. The participants had to state whether the words presented referred to biological or non-biological entities. The EEG signals were recorded by a Neuroscan Quik-cap device connected to 64 electrodes whose impedances were kept under 10 kohms. The electrodes were positioned according to the 10/20 system. The sampling frequency was 1000 Hz. The EEG data were pre-analyzed with Fieldtrip-matlab software. Segmentation windows started 250 ms before and ended 700 ms after the appearance of the word. Baseline correction using a [-250-0ms] window before stimulus onset and various filters was applied. The Fieldtrip ft_rejectvisual function was used for segment rejection and bad-electrode detection. Bad electrodes were repaired using triangulation on the cap surface and spline functions (Fieldtrip ft_channelrepair). EEG channels were re-referenced offline to the average reference, including all electrodes except the mastoid electrodes (M1, M2), CB1, CB2, and the electro-oculograms (EOGs)