README Description: The research data includes electroencephalographic (EEG) recordings from 35 young adults (67 recordings total), collected between May and June, 2014. The EEG data was recorded in response to auditory-visual speech items (below). The data was recorded in the .bdf format. This type of data can be converted into .dat format using software tools such as MATLAB and can be analysed using freely available EEG analyses toolbox 'fieldtrip'. DOI: 10.4225/35/54bf146fa4012 The data here consists of (1) Main study data (including EEG data as well as timing and behavioural data from the recordings) and (2) Pilot data. (1) Main Study data includes: 67 .bdf Biosemi EEG files (2 per participant) 67 .mat EEG log files (2 per participant) 70 .mat Behavioural identification response files (2 per participant) (2) Pilot Data includes: 1 CSV file consisting of behavioural responses for 11 participants (Identification Experiment) 1 CSV file consisting of behavioural responses for 11 participants (Incongruence Experiment) Notes on the EEG Dataset: The EEG data provided here is raw and contains following codes, which will help the reader to reanalyse according to the procedure mentioned in the paper. The codes are onset triggers, which appeared in the data at the same time when the auditory stimuli were presented to the participant. CODE CONDITION 1 Ba Visual Only 2 Da Visual Only 3 Ga Visual Only 4 Fa Visual Only 5 Ba Auditory Only 6 Da Auditory Only 7 Ga Auditory Only 8 Fa Auditory Only 9 Ba Congruent AV 10 Da Congruent AV 11 Ga Congruent AV 12 Fa Congruent AV 13 Ba Incongruent AV 14 Da Incongruent AV 15 Ga Incongruent AV 16 Fa Incongruent AV 1 17 Fa Incongruent AV 2 18 Fa Incongruent AV 3 Corresponding to each EEG Dataset is a .mat file that contains the timing and behavioural responses during the EEG recording. Behavioural responses were required during the experiment for every Fa stimulus. A response is indicated in the .mat file in the d.trial_data structure (header codes are in d.trial_header). The filename is coded as 'SUBJNUMBER_CONTEXT', where REL = 'Reliable', UNREL = 'Unreliable'. Prepared by Tim Paris The MARCS Institute paris.tj@gmail.com January 2015