Deconvolving BOLD activation in event-related designs for multivoxel pattern classification analyses
Jeanette A Mumford, Benjamin O Turner, F Gregory Ashby, Russell A Poldrack, Jeanette A Mumford, Benjamin O Turner, F Gregory Ashby, Russell A Poldrack
Abstract
Use of multivoxel pattern analysis (MVPA) to predict the cognitive state of a subject during task performance has become a popular focus of fMRI studies. The input to these analyses consists of activation patterns corresponding to different tasks or stimulus types. These activation patterns are fairly straightforward to calculate for blocked trials or slow event-related designs, but for rapid event-related designs the evoked BOLD signal for adjacent trials will overlap in time, complicating the identification of signal unique to specific trials. Rapid event-related designs are often preferred because they allow for more stimuli to be presented and subjects tend to be more focused on the task, and thus it would be beneficial to be able to use these types of designs in MVPA analyses. The present work compares 8 different models for estimating trial-by-trial activation patterns for a range of rapid event-related designs varying by interstimulus interval and signal-to-noise ratio. The most effective approach obtains each trial's estimate through a general linear model including a regressor for that trial as well as another regressor for all other trials. Through the analysis of both simulated and real data we have found that this model shows some improvement over the standard approaches for obtaining activation patterns. The resulting trial-by-trial estimates are more representative of the true activation magnitudes, leading to a boost in classification accuracy in fast event-related designs with higher signal-to-noise. This provides the potential for fMRI studies that allow simultaneous optimization of both univariate and MVPA approaches.
Copyright © 2011 Elsevier Inc. All rights reserved.
Figures
References
- Carlson TA, Schrater P, He S. Patterns of activity in the categorical representations of objects. J Cogn Neurosci. 2003;15:704–717.
- Chen X, Pereira F, Lee W, Strother S, Mitchell T. Exploring predictive and reproducible modeling with the single-subject FIAC dataset. Hum Brain Mapp. 2006;27:452–461.
- Cox DD, Savoy RL. Functional magnetic resonance imaging (fMRI) ”brain reading”: detecting and classifying distributed patterns of fMRI activity in human visual cortex. Neuroimage. 2003;19:261–270.
- De Martino F, Valente G, Staeren N, Ashburner J, Goebel R, Formisano E. Combining multivariate voxel selection and support vector machines for mapping and classification of fMRI spatial patterns. Neuroimage. 2008;43:44–58.
- Hastie T, Tibshirani R, Friedman J. The elements of statistical learning. Springer; 2001.
- Haxby JV, Gobbini MI, Furey ML, Ishai A, Schouten JL, Pietrini P. Distributed and overlapping representations of faces and objects in ventral temporal cortex. Science. 2001;293:2425–2430.
- Hoerl A, Kennard R. Ridge regression: applications to nonorthogonal problems. Technometrics. 1970;12:69–82.
- Kamitani Y, Tong F. Decoding seen and attended motion directions from activity in the human visual cortex. Curr. Biol. 2006;16:1096–1102.
- Kriegeskorte N, Mur M, Bandettini P. Representational similarity analysis - connecting the branches of systems neuroscience. Front Syst Neurosci. 2008;2:4.
- LaConte S, Strother S, Cherkassky V, Anderson J, Hu X. Support vector machines for temporal classification of block design fMRI data. Neuroimage. 2005;26:317–329.
- Misaki M, Kim Y, Bandettini PA, Kriegeskorte N. Comparison of multivariate classifiers and response normalizations for pattern-information fMRI. Neuroimage. 2010;53:103–118.
- Mourao-Miranda J, Reynaud E, McGlone F, Calvert G, Brammer M. The impact of temporal compression and space selection on SVM analysis of single-subject and multi-subject fMRI data. Neuroimage. 2006;33:1055–1065.
- Mur M, Bandettini PA, Kriegeskorte N. Revealing representational content with pattern-information fMRI–an introductory guide. Soc Cogn Affect Neurosci. 2009;4:101–109.
- Pereira F, Mitchell T, Botvinick M. Machine learning classifiers and fMRI: a tutorial overview. Neuroimage. 2009;45:199–209.
- Polyn SM, Natu VS, Cohen JD, Norman KA. Category-specific cortical activity precedes retrieval during memory search. Science. 2005;310:1963–1966.
- Rissman J, Gazzaley A, D’Esposito M. Measuring functional connectivity during distinct stages of a cognitive task. Neuroimage. 2004;23:752–763.
- Scholkopf B, Smola A. Learning with kernels. MIT press; 2002.
- Turner B. A comparison of methods for the use of pattern classification on rapid event-related fMRI data. Poster session presented at the Annual Meeting of the Society for Neuroscience; San Diego, CA. 2010.
- Woolrich MW, Ripley BD, Brady M, Smith SM. Temporal autocorrelation in univariate linear modeling of FMRI data. Neuroimage. 2001;14:1370–1386.
- Xue G, Dong Q, Chen C, Lu Z, Mumford JA, Poldrack RA. Greater neural pattern similarity across repetitions is associated with better memory. Science. 2010;330:97–101.
Source: PubMed