As part of the Bremer Initiative to Foster Early Childhood Development (BRISE), EEG data were collected from approximately 150 infants aged seven to eleven months. A passive two stimulus auditory oddball paradigm was conducted, consisting of standard (466 Hz) and deviant (554 Hz) piano tone stimuli. Such an oddball paradigm typically elicits P3 event related potentials (ERPs). Here, specifically the P3a subcomponent is expected, since the deviants are not declared as targets and thus not responded to. Since these potentials are quite robust, previous research has used them on a single trial level to make predictions about the occurrence of stimuli. The goal of this study is to adapt previous machine learning classifiers to the raw infant EEG data at hand. The purpose is to learn about 1) the features used by the classifier 2) how generalizable the algorithms are and where limitations exist 3) human ERP development in general.
Vortragsdetails
Deviant Tone Detection in lnfants by P3a ERP Classification
In der Regel sind die Vorträge Teil von Lehrveranstaltungsreihen der Universität Bremen und nicht frei zugänglich. Bei Interesse wird um Rücksprache mit dem Sekretariat unter sek-ric(at)dfki.de gebeten.
