Individual differences in error tolerance in humans: Neurophysiological evidences

Abstract

When interacting in error-prone environments, humans display different tolerances to changing their decisions when faced with erroneous feedback information. Here, we investigated whether these individual differences in error tolerance (ET) were reflected in neurophysiological mechanisms indexing specific motivational states related to feedback monitoring. To explore differences in ET, we examined the performance of 80 participants in a probabilistic reversal-learning task. We then compared eventrelated brain responses (ERPs) of two extreme groups of participants (High ET and Low ET), which showed radical differences in their propensity to maintain newly learned rules after receiving spurious negative feedback. We observed that High ET participants showed reduced anticipatory activity prior to the presentation of incoming feedback, informing them of the correctness of their performance. This was evidenced by measuring the amplitude of the stimulus-preceding negativity (SPN), an ERP component indexing attention and motivational engagement of incoming informative feedback. Postfeedback processing ERP components (the so-called Feedback-Related Negativity and the P300) also showed reduced amplitude in this group (High ET). The general decreased responsiveness of the High ET group to external feedback suggests a higher proneness to favor internal(rule)-based strategies, reducing attention to external cues and the consequent impact of negative evaluations on decision making. We believe that the present findings support the existence of specific cognitive and motivational processes underlying individual differences on error-tolerance among humans, contributing to the ongoing research focused on understanding the mental processes behind human fallibility in error-prone scenarios.Keywords Error tolerance . Decision-making . Action-monitoring . ERPs . Individual differences Humans have different tolerance to negative feedback about their performance and changing their ongoing behaviors based on external environmental cues. As these individual differences are likely to influence the way their decisions are acted upon, an important challenge for cognitive psychologists and neuroscientists is not only to understand the nature of the brain mechanisms and inherent cognitive and motivational processes underlying these differences but also to explain how these differences are related to (in)adequate decisional processes carried out by human agents in their interaction with real-world and error-prone environments.The study of human error processing has advanced enormously during the last 2 decades, mostly due to the discovery of specific event-related brain components (ERPs) and neural networks that respond selectively to error monitoring, error correction, and compensation processes (Carter et al., 1998;

Theta oscillations integrate functionally segregated sub-regions of the medial prefrontal cortex

“…A potential problem of this procedure would be differences in task performance between sessions (e.g., because participant used different strategies in the two sessions). However, in the present study, behavioral measures presented a very high reliability, suggesting that the strategy used by the participants in the two sessions was very similar, and supporting the idea that individual differences in the reversal learning task are highly consistent and might reflect a trait characteristic, as proposed in different studies (Jocham et al, 2009, den Oude et al, 2013Padrao et al, 2015. Additionally, the use of two separate sessions allows us to adapt each session to the different technical requirements of each neuroimaging modality.…”

“…Note that participants may not fully predict the occurrence of a reversal as the stimuli were the same for all the blocks (see Figure 1) and the length of each block varied randomly from 10 to 16 trials. Similar approaches have been extensively used in the literature (Cools et al, 2002, Fellows and Farah, 2003, Cools et al, 2007, Jocham et al, 2009, Philiastides et al, 2010Chase et al, 2011, Padrao et al, 2015. However, to facilitate the reversal, during the first five trials following the contingency reversal, a selection of the previously correct stimulus resulted in punishment.…”