Invited Symposium

The Role of Functional Neuroimaging in Advancing Psychological Science

Time and Location
Saturday May 24, 2008, 9:30 AM - 11:20 AM
Chicago 10


This symposium will address the utility and future of functional neural imaging, and in particular functional magnetic resonance imaging, in psychology. The use of these techniques has expanded rapidly, perhaps faster than the attention to the construct validity of what we believe we are measuring. Have these techniques advanced psychological science or have they simply produced maps of an illusory neural landscape? These issues, as well as the progress and promise in the field, are addressed.

Martha Farah

Martha Farah (Co-Chair)
University of Pennsylvania

John Allen (Co-Chair)
University of Arizona




Greg Miller

Greg Miller
University of Illinois at Urbana-Champaign
What Does Spatial and Temporal Localization in Neuroimaging Localize?
Among psychophysiological methods such as fMRI, EEG, and MEG, the best electromagnetic imaging procedures match the spatial resolution of standard fMRI, and the temporal resolution of MR-based methods continues to improve as well. Optical imaging shows great promise as a means of bridging electromagnetic and hemodynamic methods, and multimodal imaging now has considerable momentum. Unfortunately, progress in localization methods has fostered naive reductionism about the relationship between psychological and biological events (even though we do not have a clue about causal mechanisms in either direction). Beyond the spatial and temporal limits of localization, the logical limits of localization warrant careful consideration, with implications for how we design and interpret basic and clinical research.

Stefan Debener

Stefan Debener
MRC Institute of Hearing Research, Royal South Hants Hospital
Towards Single Trial Analysis in Cognitive Brain Research
The vast majority of electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI) investigations of cognitive functions focus on trial-averaged data, disregarding to some extent the fact that behavior takes place on a single-trial level. Moreover, many cognitive processes typically fluctuate over time (e.g., novelty processing, attention, cognitive control), thus hampering the study of their neural correlates. Therefore, a single-trial oriented analysis of EEG and fMRI data has been proposed. While this approach faces the problem of spatio-temporally mixed signals, recent advances in using independent component analysis and multimodal integration of EEG and fMRI suggest that this problem can be successfully addressed. Single-trial analysis provides a new dimension of studying the brain-behavior relationship and is likely to play a major role in the future of cognitive neuroimaging.

Steven Small

Steven Small
The University of Chicago
Developing a Neurobiological Language for Theory Development in Psychology
fMRI can be used to elaborate theories speaking to the neural mechanisms underlying human behavior. This will require critical changes in the design and analysis of fMRI studies, such that (1) behaviors of interest be studied without ancillary tasks (e.g., decisions), (2) stimuli be presented in as natural a way as possible, (3) results be expressed as networks of activity, and (4) data analysis take into account the distributed nature of regional representations. This lecture will focus on several novel methods, including peak and valley analysis of continuous time series and exhaustive connectivity modeling of activation data.

Matthew Botvinick

Matthew Botvinick
Princeton University
Pattern-Analytic fMRI
In certain quarters of cognitive psychology, it is a strongly held belief that human information processing operates upon representations that are distributed in nature. Computational models employing distributed representations have provided powerful accounts of human behavior in domains ranging from to perceptual decision-making to semantic cognition. And direct evidence for distributed representations has come from a variety of single-unit recording studies. Until recently, functional neuroimaging research has failed to make contact with this important dimension of psychological thought. However, over the past few years, several labs have begun to apply multivariate methods to the analysis of functional neuroimaging data, revealing that neuroimaging may be capable of testing for and characterizing distributed representations at the neural level. I’ll review a series of key studies, and describe current work applying pattern-analytic fMRI to the study of language.

Go back

APS Journals