87
signaling only that a touch has occurred. Decoding performance in MST∕STP was also
poor for foot touches, consistent with a role in eye-hand coordination. While eye-foot
coordination is important in some tasks, such as directing a ball with the foot, these
tasks may be subserved by other brain areas.
Building on the previous literature (Haynes and Rees 2006; Kamitani and Tong
2005; Kriegeskorte, et al. 2006; LaConte, et al. 2005; Norman, et al. 2006), we
performed MVPA on individual trials presented in a rapid event-related somatosensory
stimulation design. The ability to classify single trials has several important advantages.
First, it results in a large testing and training set, important for good classification
performance (Mitchel, et al. 2004). Second, it allows for real-time designs that provide
feedback to the subject or make adjustments in the task difficulty (LaConte, et al. 2007).
Third, it is a necessity for correlating behavior and information content on a trial-by-trial
basis (Wagner, et al. 1998). In the somatosensory modality, event-related designs are
particularly critical because there is a great amount of adaptation both peripherally and
centrally (Deuchert, et al. 2002). Taken together, these studies illustrate how MVPA
allows a closer investigation of the function Ofdifferent cortical areas by examining their
information content, above and beyond simple fMRI activation maps of single voxels
responding to a sensory stimulus (Kriegeskorte and Bandettini 2007).