C. Stephanidis and M. Antona (Eds.): UAHCI/HCII 2013, Part I, LNCS 8009, pp. 672–681, 2013.
© Springer-Verlag Berlin Heidelberg 2013
A Collaborative Brain-Computer Interface
for Accelerating Human Decision Making
Peng Yuan
1
, Yijun Wang
2
, Xiaorong Gao
1
, Tzyy-Ping Jung
2
, and Shangkai Gao
1
1
Department of Biomedical Engineering, School of Medicine,
Tsinghua University, Beijing, China
2
Swartz Center for Computational Neuroscience, Institute for Neural Computation,
University of California, San Diego, San Diego, USA
yuanp09@mails.tsinghua.edu.cn, {yijun,jung}@sccn.ucsd.edu,
{gxr-dea,gsk-dea}@tsinghua.edu.cn
Abstract. Recently, collective intelligence has been introduced to brain-
computer interface (BCI) research, leading to the emergence of collaborative
BCI. This study presents an online collaborative BCI for improving individuals’
decision making in a visual Go/NoGo task. Six groups of six people
participated in the experiment comprising both offline and online sessions. The
offline results suggested that the collaborative BCI has the potential to improve
individuals’ decisions in various decision-making situations. The online tests
showed that using Electroencephalogram (EEG) within the first 360 ms after
the stimulus onset, which was 50 ms earlier than the mean behavioral response
time (RT) (409±85 ms), the collaborative BCI reached a mean classification
accuracy of 78.0±2.6% across all groups. It was 12.9% higher than the average
individual accuracy (65.1±8.1%, p<10
-4
). This study suggested that a
collaborative BCI could accelerate human decision making with reliable
prediction accuracy in real time.
Keywords: brain-computer interface (BCI), group decision making,
Electroencephalogram (EEG), collaborative BCI.
1 Introduction
In human-performance studies, a team of individuals usually outperforms individuals
especially when performance requires diverse skills, judgments, and experiences
under time constraints [1]. Two heads are better than one, known as collective
intelligence, the mechanism and neural basis of which has recently attracted growing
attention of researchers in psychology and neuroscience [2, 3].
Recently, the collective intelligence has been introduced to the brain-computer
interface (BCI) research field. For Instance, the concept of collaborative BCI has been
proposed in [4] and [5]. Through offline demonstrations of collaborative BCIs, these
studies suggest that a collaborative BCI, which integrated neural information from
multiple individuals, can outperform a single-brain BCI. More recently, we
implemented the first online collaborative BCI in a visual target detection task [6].