A Model-Based Approach to Trial-By-Trial P300 Amplitude Fluctuations

Abstract

It has long been recognized that the amplitude of the P300 component of event-related brain potentials is sensitive to the degree to which eliciting stimuli are surprising to the observers (Donchin, 1981). While Squires et al. (1976) showed and modeled dependencies of P300 amplitudes from observed stimuli on various time scales, Mars et al. (2008) proposed a computational model keeping track of stimulus probabilities on a long-term time scale. We suggest here a computational model which integrates prior information with short-term, long-term, and alternation-based experiential influences on P300 amplitude fluctuations. To evaluate the new model, we measured trial-by-trial P300 amplitude fluctuations in a simple two-choice response time task, and tested the computational models of trial-by-trial P300 amplitudes using Bayesian model evaluation. The results reveal that the new digital filtering (DIF) model provides a superior account of the trial-by-trial P300 amplitudes when compared to both Squires et al.’s (1976) model, and Mars et al.’s (2008) model. We show that the P300-generating system can be described as two parallel first-order infinite impulse response (IIR) low-pass filters and an additional fourth-order finite impulse response (FIR) high-pass filter. Implications of the acquired data are discussed with regard to the neurobiological distinction between short-term, long-term, and working memory as well as from the point of view of predictive coding models and Bayesian learning theories of cortical function.

P300 amplitude variations, prior probabilities, and likelihoods: A Bayesian ERP study

“…Specifically, s 1 stimuli in uncertain likelihood conditions evoked more prominent P3b responses compared to s 1 stimuli in certain likelihood conditions. This finding is in accord with previous research on the P3b ERP component, which revealed that P3b amplitude variations are related to the uncertainty or surprise of its eliciting events (Donchin, 1981;Kolossa et al, 2013;Sutton, Braren, Zubin, & John, 1965).…”

“…Therefore, our findings are novel because there are no previous studies that demonstrate ERP correlates of instructed Bayesian parameters. This may help to understand why our P3b negative results with regard to the unexpectedness of the evidence seem to be in disagreement with the existing P3b literature (see Kolossa et al, 2013;Kolossa et al, 2015). Most P3b studies show that P3b amplitude variations are sensitive to the unexpectedness of the evidence or surprise, but none of these studies has examined instructed unexpectedness or surprise against instructed likelihoods.…”

“…Their data suggest that early processing stages (somatosensory cortex) may implement short-term Bayesian perceptual learning, intermediate processing stages (right inferior frontal cortex) index active stimulus engagement, and late processing (cingulate cortex) may reflect learning-induced updating of top-down attentional control mechanisms (as expressed in scalprecorded P3b responses). Kolossa, Fingscheidt, Wessel, and Kopp (2013) modeled the surprise-sensitivity of trial-by-trial P3b amplitude variations as originating from the operation of multiple digital filters in the brain (i.e., two parallel first-order infinite impulse response low-pass filters and an additional fourth-order finite impulse response high-pass filter). Kolossa, Kopp, and Fingscheidt (2015) replicated the surprise-sensitivity of P3b amplitude variations and showed that the anteriorly distributed P3a encodes Bayesian updating, a finding that was also reported by Bennett, Murawski, and Bode (2015).…”

“…4a; see also MacKay, 2003) such as surprise (Kolossa et al, 2013), entropy (Kopp & Lange, 2013), and redundancy (Barlow, 2001) as response functions for modeling the measured ERP responses (Friston, 2005). Surprise (I) equals a logarithmic measure of the probability p(x) that a binary random variable X takes on the value x.…”

Zur Psychophysiologie exekutiver Funktionen

“…Die zeitgenössische Art der Datenanalyse beruht weitgehend auf der Methode des statistischen Nullhypothesen testens, die eine theorieferne Vorgehensweise begünstigt. Dem gegenüber ermöglichen die komputationalen Neurowissen schaften, erhobene Daten gegen präzise Vorhersagen zu testen, die sich aus formalisierten Theorien herleiten lassen [55,106]. In ihrer Gesamtheit ebnen diese Neuerungen den Weg für bislang für unmöglich gehaltene Fortschritte der Psychophysiologie exekutiver Funktionen.…”