Effects of parametrical and trial-to-trial variation in prior probability processing revealed by simultaneous electroencephalogram/functional magnetic resonance imaging

Abstract

Prior knowledge of the probabilities concerning decision alternatives facilitates the selection of more likely alternatives to the disadvantage of others. The neural basis of prior probability (PP) integration into the decision-making process and associated preparatory processes is, however, still essentially unknown. Furthermore, trial-to-trial fluctuations in PP processing have not been considered thus far. In a previous study, we found that the amplitude of the contingent negative variation (CNV) in a precueing task is sensitive to PP information (Scheibe et al., 2009). We investigated brain regions with a parametric relationship between neural activity and PP and those regions involved in PP processing on a trial-to-trial basis in simultaneously recorded electroencephalographic (EEG) and functional magnetic resonance imaging (fMRI) data. Conventional fMRI analysis focusing on the information content of the probability precue revealed increasing activation of the posterior medial frontal cortex with increasing PP, supporting its putative role in updating action values. EEG-informed fMRI analysis relating single-trial CNV amplitudes to the hemodynamic signal addressed trial-to-trial fluctuations in PP processing. We identified a set of regions mainly consisting of frontal, parietal, and striatal regions that represents unspecific response preparation on a trial-to-trial basis. A subset of these regions, namely, the dorsolateral prefrontal cortex, the inferior frontal gyrus, and the inferior parietal lobule, showed activations that exclusively represented the contributions of PP to the trial-to-trial fluctuations of the CNV.

Figure 1.

Task and behavioral data. A, Number comparison task. After a color change of the fixation cross, S1 was presented either left or right of the fixation cross. After a 2000 ms foreperiod, S2 was added to the display on the other side. The appearance of S2 demanded a button press on the side of the larger number. S1 served as a precue for the PP of the side of the larger number. RT was parametrically modulated by PP. B, RT data and vincentized cumulative probability distributions. Mean RT and SEs for the conditions PP 0.75-invalid, PP 0.5, PP 0.75-valid, and PP 1.0 (top). Vincentized cumulative probability distributions of RT for the conditions PP 0.75-invalid, PP 0.5, PP 0.75-valid, and PP 1.0 (bottom). The RT effect of PP is clearly present throughout the RT distribution.

Figure 2.

Mean CNV amplitude and single trials. A, Mean CNV amplitudes. Time course (top) and amplitude (bottom) of the CNV at electrode position Cz for the conditions PP 0.75, PP 0.5, and PP 1.0. The CNV increased parametrically with increasing PP. B, Single-trial data. Single-trial data of one exemplary participant. Trials are depicted separately for conditions PP 0.75, PP 0.5, and PP 1.0 and sorted by RT (sigmoid black line). All conditions show a considerable trial-to-trial variability in the CNV amplitude.

Figure 3.

fMRI results of the conjunction analysis. A, Conjunction analysis. The conjunction analysis (model 1) designed to identify voxels parametrically modulated by PP revealed activation in the pMFC. The percentage signal changes for each condition are depicted on the right. Neural activation (shown at p < 0.005, uncorrected) is projected on an MNI template (Colin). B, Contrast PP 1.0 > 0.5. The contrast PP 1.0 > 0.5 activated regions in the pMFC, MFG, IFG, superior parietal lobule (SPL), cingulate gyrus, and midbrain structures. Activation clusters (shown at p < 0.001, uncorrected) are projected on an MNI template (Colin).

Figure 4.

fMRI results of the single-trial analysis. A, The BOLD signal covaried with the single-trial CNV amplitude in a set of regions including prefrontal, parietal, and striatal regions (model 2). B, This shows differentially activated regions in which BOLD signal covaried with CNV amplitude (model 2, blue), differentially activated regions in which BOLD signal covaried with CNV amplitude residuals (model 3, red), and the overlap between the two models (green). The middle frontal gyrus (i.e., DLPFC), the IFG, and the IPL represent the contribution of PP to the trial-to-trial fluctuations of the CNV amplitude. Activation clusters (shown at p < 0.005, uncorrected) are projected on an MNI template (Colin). IPS, Inferior parietal sulcus; L, left; R, right; S, sulcus.