From Head to Toe: Efficient Somatosensory Mapping with Fast Stimulation and Multivariate Pattern Analysis

Background Somatosensory evoked potentials (SEPs) measured with electroencephalography (EEG) are widely used to study cortical responses to touch but most research has limited the focus on few body parts, typically a finger, and applied time-consuming testing protocols. Multivariate pattern analysis (MVPA) provides a complementary approach that may increase sensitivity and allow faster stimulation, yet its relationship to classical SEP analysis in somatosensory research remains largely unexplored.

Fifteen participants received vibrotactile stimulation on the finger, hand, cheek, and foot while EEG was recorded. We compared a traditional “slow” stimulation protocol (800-1200 ms inter-stimulus intervals) with a “fast” protocol (300-500 ms). We compared temporal and topographical aspects between SEP and MVPA.

Both stimulation protocols produced highly similar SEP components (P100, N140, P200), topographies, and classification results, while the fast protocol reduced testing time by about 60%. SEPs revealed systematic body-part differences, with earlier components for cheek stimulation and delayed responses for the foot. Multivariate classification distinguished body parts with accuracies up to ∼50-55% (chance: 25%), peaking around 100 ms after stimulus onset. Classifier weight maps closely matched SEP topographies over centroparietal electrodes, indicating that classification relied on physiologically meaningful somatosensory signals. Classification accuracy peaked around 100 ms after stimulus onset, coinciding with the SEP P100 component, but declined gradually thereafter, suggesting that early somatosensory responses contain particularly informative multivariate patterns that generalize over time.

Faster stimulation protocols substantially increase efficiency without compromising interpretability. Combining classical SEP analysis with multivariate classification provides complementary insights and offers a powerful framework for mapping somatosensory representations across the body. Competing Interest Statement The authors have declared no competing interest.