Gaze-independent ERP-BCIs: augmenting performance through location-congruent bimodal stimuli.

Marieke Thurlings, Anne-Marie Brouwer, Jan B.F. Erp, van, Peter Werkhoven

    Research output: Contribution to journalArticleAcademicpeer-review

    Abstract

    Gaze-independent event-related potential (ERP) based brain-computer interfaces (BCIs)
    yield relatively low BCI performance and traditionally employ unimodal stimuli. Bimodal
    ERP-BCIs may increase BCI performance due to multisensory integration or summation
    in the brain. An additional advantage of bimodal BCIs may be that the user can
    choose which modality or modalities to attend to. We studied bimodal, visual-tactile,
    gaze-independent BCIs and investigated whether or not ERP components’ tAUCs and
    subsequent classification accuracies are increased for (1) bimodal vs. unimodal stimuli;
    (2) location-congruent vs. location-incongruent bimodal stimuli; and (3) attending to both
    modalities vs. to either one modality. We observed an enhanced bimodal (compared to
    unimodal) P300 tAUC, which appeared to be positively affected by location-congruency
    (p = 0.056) and resulted in higher classification accuracies. Attending either to one
    or to both modalities of the bimodal location-congruent stimuli resulted in differences
    between ERP components, but not in classification performance. We conclude that
    location-congruent bimodal stimuli improve ERP-BCIs, and offer the user the possibility
    to switch the attended modality without losing performance.
    Original languageEnglish
    Article number143
    Number of pages14
    JournalFrontiers in Neuroscience
    Volume8
    DOIs
    Publication statusPublished - 2014

    Keywords

    • BCI
    • ERP
    • gaze-independent
    • bimodal
    • tactile
    • multisensory
    • location-congruency
    • selective attention

    Fingerprint

    Dive into the research topics of 'Gaze-independent ERP-BCIs: augmenting performance through location-congruent bimodal stimuli.'. Together they form a unique fingerprint.

    Cite this