Pattern classification of EEG signals reveals perceptual and attentional states

PLOS ONE, Dec 2019

Pattern classification techniques have been widely used to differentiate neural activity associated with different perceptual, attentional, or other cognitive states, often using fMRI, but more recently with EEG as well. Although these methods have identified EEG patterns (i.e., scalp topographies of EEG signals occurring at certain latencies) that decode perceptual and attentional states on a trial-by-trial basis, they have yet to be applied to the spatial scope of attention toward global or local features of the display. Here, we initially used pattern classification to replicate and extend the findings that perceptual states could be reliably decoded from EEG. We found that visual perceptual states, including stimulus location and object category, could be decoded with high accuracy peaking between 125–250 ms, and that the discriminative spatiotemporal patterns mirrored and extended our (and other well-established) ERP results. Next, we used pattern classification to investigate whether spatiotemporal EEG signals could reliably predict attentional states, and particularly, the scope of attention. The EEG data were reliably differentiated for local versus global attention on a trial-by-trial basis, emerging as a specific spatiotemporal activation pattern over posterior electrode sites during the 250–750 ms interval after stimulus onset. In sum, we demonstrate that multivariate pattern analysis of EEG, which reveals unique spatiotemporal patterns of neural activity distinguishing between behavioral states, is a sensitive tool for characterizing the neural correlates of perception and attention.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0176349&type=printable

Pattern classification of EEG signals reveals perceptual and attentional states

April Pattern classification of EEG signals reveals perceptual and attentional states Alexandra List 0 1 2 Monica D. Rosenberg 0 2 Aleksandra Sherman 0 2 Michael Esterman 0 2 0 Funding: National Institutes of Health R01 EY018197-02S1 (supported AL); National Science Foundation Graduate Research Fellowship (to MR); Veterans Affairs Clinical Science Research and Development Career Development Award 1IK2CX000706-01A2 (to ME). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript 1 Department of Psychology and Neuroscience Program, Hamilton College, Clinton, New York, United States of America, 2 Department of Psychology, Yale University , New Haven , Connecticut, United States of America, 3 Department of Cognitive Science, Occidental College , Los Angeles , California, United States of America, 4 Research Service, VA Boston Healthcare System, Boston, Massachusetts, United States of America, 5 Department of Psychiatry, Boston University School of Medicine , Boston, Massachusetts , United States of America 2 Editor: Lawrence M Ward, University of British Columbia , CANADA Pattern classification techniques have been widely used to differentiate neural activity associated with different perceptual, attentional, or other cognitive states, often using fMRI, but more recently with EEG as well. Although these methods have identified EEG patterns (i.e., scalp topographies of EEG signals occurring at certain latencies) that decode perceptual and attentional states on a trial-by-trial basis, they have yet to be applied to the spatial scope of attention toward global or local features of the display. Here, we initially used pattern classification to replicate and extend the findings that perceptual states could be reliably decoded from EEG. We found that visual perceptual states, including stimulus location and object category, could be decoded with high accuracy peaking between 125±250 ms, and that the discriminative spatiotemporal patterns mirrored and extended our (and other wellestablished) ERP results. Next, we used pattern classification to investigate whether spatiotemporal EEG signals could reliably predict attentional states, and particularly, the scope of attention. The EEG data were reliably differentiated for local versus global attention on a trial-by-trial basis, emerging as a specific spatiotemporal activation pattern over posterior electrode sites during the 250±750 ms interval after stimulus onset. In sum, we demonstrate that multivariate pattern analysis of EEG, which reveals unique spatiotemporal patterns of neural activity distinguishing between behavioral states, is a sensitive tool for characterizing the neural correlates of perception and attention. - Data Availability Statement: Data from the paper are available through the Open Science Framework at osf.io/nnke6. Introduction Over the last decade, multivariate pattern-classification analyses of fMRI BOLD signals have emerged as a fruitful approach for using neural activity to decode various behavioral states including perceiving, attending to, and imagining features, objects, and scenes (for reviews, see [1±4]). Recently, pattern-classification analyses have also been applied to electroencephalography (EEG) signals (e.g., [5±16]). This application to EEG has extended the standard event-related potential (ERP) analyses in which a critical electrode (or a cluster of electrodes) is selected within a specific scalp region (based on data inspection and/or prior results), and the trial-averaged stimulus-evoked EEG signals (i.e., ERPs) from the selected electrode(s) are compared between conditions. Instead, as applied here, multivariate classification techniques can reveal, in an agnostic data-driven manner, topographic weightings of EEG signals that maximally distinguish specific perceptual, attentional, or behavioral states within a given time interval. Thus, pattern-classification analyses offer greater sensitivity than standard ERP analyses by simultaneously integrating information across electrodes. Because pattern-classification analyses identify EEG correlates with high sensitivity, they are typically evaluated by how well they predict the corresponding perceptual, attentional, or behavioral states on a trial-by-trial basis (rather than how well trial-averaged signals from selected electrodes differentiate experimental conditions, as in standard ERP analyses). Cross-validated predictive measures, like the ones we use here, are also less susceptible to false positives than analyses traditionally applied to ERPs, because inaccurate models will not generalize to the held-out data. The first aim of the current study is to replicate and extend prior EEG applications of pattern-classification analyses toward decoding perceptual states. Although prior studies have applied similar analyses toward classifying object category (e.g., faces versus cars), they have done so in the (...truncated)


This is a preview of a remote PDF: https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0176349&type=printable

Alexandra List, Monica D. Rosenberg, Aleksandra Sherman, Michael Esterman. Pattern classification of EEG signals reveals perceptual and attentional states, PLOS ONE, 2017, Volume 12, Issue 4, DOI: 10.1371/journal.pone.0176349