Boosts in brain signal variability track liberal shifts in decision bias

Abstract

Adopting particular decision biases allows organisms to tailor their choices to environmental demands. For example, a liberal response strategy pays off when target detection is crucial, whereas a conservative strategy is optimal for avoiding false alarms. Using conventional time-frequency analysis of human electroencephalographic (EEG) activity, we previously showed that bias setting entails adjustment of evidence accumulation in sensory regions (Kloosterman et al., 2019), but the presumed prefrontal signature of a conservative-to-liberal bias shift has remained elusive. Here, we show that a liberal bias shift is reflected in a more unconstrained neural regime (boosted entropy) in frontal regions that is suited to the detection of unpredictable events. Overall EEG variation, spectral power and event-related potentials could not explain this relationship, highlighting that moment-to-moment neural variability uniquely tracks bias shifts. Neural variability modulation through prefrontal cortex appears instrumental for permitting an organism to adapt its biases to environmental demands.

A silent disco: Differential effects of beat-based and pattern-based temporal expectations on persistent entrainment of low-frequency neural oscillations

“…By calculating entropy for patterns of different lengths (“multiscale”), the contributions of slower and faster timescales in the signal can be assessed. However, the mapping between entropy timescales and spectral frequencies is not absolute, especially since entropy is not per se related to a signal being oscillatory in nature (Kloosterman, Kosciessa, Lindenberger, Fahrenfort, & Garrett, 2020; Kosciessa et al, 2020). The advantage of using MSE is that it does not require filtering of the data, and it does not assume stationarity (e.g., it can pick up on regularities that are asymmetrical, or that do not have a fixed amplitude or period).…”

“…A section of samples is considered a match if it resembles the template pattern enough to fall within a set boundary, which is defined as r x SD (the similarity bound). The number of pattern matches is counted. Subsequently, the same procedure is followed for patterns of m + 1 samples long. For the total counts of pattern matches throughout the time series, sample entropy is then calculated as the logarithm of the ratio between pattern matches of length m and pattern matches of length m + 1. Thus, sample entropy reflects the proportion of patterns in the time series that stays similar when an extra sample is added to the pattern. Here, we used m = 2 and r = 0.5, as was done previously for EEG data (Kloosterman, Kosciessa, Lindenberger, Fahrenfort, & Garrett, 2020).…”

“…To compute MSE, we used the mMSE toolbox, a plugin to the Fieldtrip toolbox (Kloosterman et al, 2020). To compute entropy, epochs were concatenated.…”

Dynamic regulation of neural variability during working memory reflects dopamine, functional integration, and decision-making

“…Such models separate non-decision (e.g., motor) from decision-related components (e.g., rate of evidence accumulation, or drift rate ) and can also estimate the extent to which participants tend towards certain choice alternatives. Recent work suggests that those who can modulate evidence accumulation with increasing cognitive demand 30 and adjust their decision criteria when required also express greater EEG-based variability 31 . However, spatially specific (especially striato-thalamic) signatures of how neural variability reflects evidence accumulation and decision criteria under cognitive load remain unknown.…”

Decision-Making in the Human-Machine Interface

“…This process can of course be willfully biased by the decision-maker, for example by adopting a liberal or conservative response criterion under differing task demands. For example, Kloosterman et al (2019) found that liberal response criteria are associated with a suppression of alpha band activity, relative to conservative criteria, which appears to systematically bias the direction of evidence accumulation toward a “target present” response in a go/no-go task (see also: Kloosterman et al, 2020 ). But given the logic of the DDM, influencing neural activity through explicit task demands is only one method of introducing systematic bias into the decision-making process, and other mechanisms may be external to the cognitive dynamics of evidence accumulation entirely.…”