We introduce the Hidden Process Model (HPM), a probabilistic model for multivariate time series data intended to model complex, poorly understood, overlapping and linearly additive processes. HPMs are motivated by our interest in modeling cognitive processes given brain image data. We define HPMs, present inference and learning algorithms, study their characteristics using synthetic data, and demonstrate their use for tracking human cognitive processes using fMRI data.

13 pages

*Computer Science Department, Carnegie Mellon University
**Center for the Neural Basis of Cognition, Carnegie Mellon University

This page maintained by reports@cs.cmu.edu