Classifying Self-Driven Mental Tasks from Whole-Brain Activity Patterns
Multivariate pattern analysis (MVPA) has been successfully employed to advance our understanding of where and how information regarding different cognitive-affective mental states is represented in the human brain, bringing new insights into how these states come to fruition, and providing a promising complement to the mass-univariate approach. In this talk, I will mostly present results from two studies where we examined the possibility of determining which mental task (out of two) a participant was performing at a given point in time from a single whole-brain functional magnetic resonance imaging scan. Participants were asked to count down numbers or recall negative/positive autobiographical episodes of their personal lives, for 32 seconds at a time, during which they could engage in the execution of those tasks in a self-paced, self-driven manner. Results indicated that it is possible to accurately classify mental tasks from whole-brain activity patterns recorded in a time interval as short as 2 seconds. In the second study, we applied the same technique to retrieve answers to simple questions using two mental tasks as proxies for ‘yes’ and ‘no’. Results indicated that binary answers can be retrieved from whole-brain activity patterns, suggesting that MVPA may provide an alternative way to establish basic communication with unresponsive patients when other techniques are not successful. I would also like to talk about other on-going projects, and look for further opportunities to expand the collaboration between our organisations.