The timing of exploratory decision-making revealed by single-trial topographic EEGanalyses

Abstract

Decision-making in an uncertain environment is driven by two major needs: exploring the environment to gather information or exploiting acquired knowledge to maximize reward. The neural processes underlying exploratory decision-making have been mainly studied by means of functional magnetic resonance imaging, overlooking any information about the time when decisions are made. Here, we carried out an electroencephalography (EEG) experiment, in order to detect the time when the brain generators responsible for these decisions have been sufficiently activated to lead to the next decision. Our analyses, based on a classification scheme, extract time-unlocked voltage topographies during reward presentation and use them to predict the type of decisions made on the subsequent trial. Classification accuracy, measured as the area under the Receiver Operator's Characteristic curve was on average 0.65 across 7 subjects. Classification accuracy reached a plateau for each of the subjects after ~510ms on average. We speculate that decisions were already made before this critical period, as confirmed by a positive correlation with reaction times across subjects. On an individual subject basis, distributed source estimations were performed on the extracted topographies to statistically evaluate the neural correlates of decision-making. For trials leading to exploration, there was significantly higher activity in dorsolateral prefrontal cortex and the right supramarginal gyrus; areas responsible for modulating behavior under risk and deduction. No area was more active during exploitation. We show for the first time the temporal evolution of differential patterns of brain activation in an exploratory decision-making task on a single-trial basis.3

Quantifying the time for accurate EEG decoding of single value-based decisions

“…In a similar way to our previous study (Tzovara et al, 2012a), we examined whether the two time-periods revealed by the TANOVA also carry predictive power of single decisions. To test this, we restricted the single-trial decoding of accept vs. reject decisions only to the periods identified from the average ERP analysis.…”

“…The single-trial topographic EEG analysis was carried out at the single-subject level and was based on a multivariate EEG decoding approach, whose suitability for decoding subjects' decisions has been shown in a previous study (Tzovara et al, 2012a). In this previous study, we evaluated the time of decisions at the neural level by taking advantage of the EEG activity locked to the reward presentation.…”

“…This range of parameters was chosen in an empirical way, based on the range that provided satisfying results on previous studies using a similar approach (Tzovara et al, 2012a). On a second step we selected the optimal pair of models by maximizing decoding performance across the 10 test datasets, in a cross validation procedure.…”

“…This method can therefore deal with trials of different lengths. A simpler version of this approach has been previously used for decoding decisions in an exploration/exploitation task based on a fixed-length trial (Tzovara et al, 2012a). The present study aims also at taking those previous findings one step further by detecting the time when single decisions are made at a neural level and by coupling behavioral models' parameters with single-trial neural modulations.…”

“…Previous studies based on EEG or magnetoencephalography (MEG) have typically investigated neural activity at an across-trials level (Cohen and Ranganath, 2007;Frank et al, 2005;Talmi et al, 2012;Tzovara et al, 2012a;Yeung et al, 2005) and provided some insight on the timing of decision-making. These studies have shown that the neural correlates of outcome evaluation typically appear as modulations of EEG components, as early as ∼200-300 ms after the presentation of the outcome of subjects' decisions (Cohen and Ranganath, 2007;Yeung et al, 2005).…”