New Directions on
Decoding Mental States from fMRI Data

NIPS 06 workshops, December 8, Whistler, Canada


Time Title Speaker
7:30-7:50 Introduction and overview of FMRI concepts and terminology John-Dylan Haynes
7:50-8:10 Overview of decoding of mental states and processes Tom Mitchell
8:10-8:40 Implications of decoding for theories of neural representation James Haxby
8:40 Coffee break
8:50-9:20 Sampling of neural topographies by voxels/Comparing neural and perceptual information John-Dylan Haynes
9:20-9:40 Challenges and limitations in interpreting learnt classifiers Francisco Pereira
9:40-10:30 Discussion:
Which novel conclusions about cognitive processing does fMRI-decoding allow us to draw?
How can decoding be used to understand the flow of information through the human brain?
Time Title Speaker
3:30-3:50 Exploring human object-vision with hi-res fMRI and information-based analysis Nikolaus Kriegeskorte
3:50-4:10 Spatiotemporal classification Janaina Mourao-Miranda
4:10-5:00 Submitted talks
Abstract Authors Institution
Enhancing functional magnetic resonance imaging with supervised learning Stephen LaConte Georgia Tech and Emory University
What Mental States? Exploring How Dimensionality Reduction Might Contribute to the Refinement of Cognitive Models Kenneth Whang Bryn Mawr College
Generative Models for Decoding Real-Valued Natural Experience in FMRI Greg Stephens, David Blei Princeton University
Classifying single trial fMRI: What can machine learning learn? Paul Mazaika, Golijeh Golarai Stanford University
5:00 Coffee break
5:10-6:20 Submitted talks
Abstract Authors Institution
Hidden Process Models:Decoding Overlapping Cognitive States with Unknown Timing Rebecca Hutchinson, Tom Mitchell Carnegie Mellon University
Deconvolution of component map time courses for task related activation discovery Rudolph Mappus, Charles Isbell Georgia Tech
Unsupervised fMRI Analysis David Hardoon, Janaina Mourao-Miranda, Michael Brammer, John Shawe-Taylor University College London/Institute of Psychiatry
Hierarchical Gaussian Naive Bayes Classifier for Multiple-Subject fMRI Data Indrayana Rustandi Carnegie Mellon University
Feature Induction Using Boosting and Logistic Regression on fMRI Images Melissa Carroll, Miroslav Dudik, Robert Schapire, Kenneth Norman Princeton University
fMRI-based decoding of the modified default-mode network in mild cognitive impairment Fabian Theis et al. Bernstein Center for Computational Neuroscience/MPI for Dynamics and Self-Organization, Gottingen
Automated fMRI Feature Abstraction using Neural Network Clustering Techniques (full paper) Radu Stefan Niculescu, Tom Mitchell Siemens Medical Solutions/Carnegie Mellon University
6:20 Coffee break
6:30-7:30 Discussion:
Can we use classifiers as "confirmatory" tools allowing the testing of competing hypotheses about structure in the data?
Is there a continuum between that and their "exploratory" use to find structure?