Chapter

Brain-Computer Interface Research

Part of the series SpringerBriefs in Electrical and Computer Engineering pp 61-72

Date:

An Affective BCI Using Multiple ERP Components Associated to Facial Emotion Processing

  • Qibin ZhaoAffiliated withLaboratory for Advanced Brain Signal Processing, Brain Science Institute, RIKEN Email author 
  • , Yu ZhangAffiliated withLaboratory for Advanced Brain Signal Processing, Brain Science Institute, RIKEN
  • , Akinari OnishiAffiliated withLaboratory for Advanced Brain Signal Processing, Brain Science Institute, RIKEN
  • , Andrzej CichockiAffiliated withLaboratory for Advanced Brain Signal Processing, Brain Science Institute, RIKEN

* Final gross prices may vary according to local VAT.

Get Access

Abstract

P300-based brain computer interfaces (BCIs) have successfully demonstrated that attention to an oddball stimulus can enhance the P300 component of the event-related potential (ERP) phase-locked to the event. However, it was unclear whether the more sophisticated face-evoked potentials can also be modulated by related mental tasks under the oddball paradigm. This study investigated ERP responses to image stimuli of objects, neutral faces, and emotional faces when subjects perform attention, face recognition and discrimination of emotional facial expressions respectively under the oddball paradigm. The results revealed the significant difference between target and non-target ERPs for each mental task. In addition, significant differences among the three mental tasks were observed for vertex-positive potential (VPP) over the fronto-central sites, late positive potential (LPP)/P3b over the centro-parietal sites and N250 over the occipito-temporal sites. These findings indicate that a novel affective BCI paradigm can be developed based on detection of multiple ERP components reflecting human face encoding and emotion processing. The high classification performance for single-trial emotional face-related ERPs demonstrated the effectiveness of the affective BCI.