Unveiling consistency in flexibility: the role of reward and cognitive control in moral decisions

Abstract

Moral decisions are multifaceted with two essential aspects, flexibility and consistency. However, the interaction between these two and the underlying mechanisms is rarely studied. Here, we combined mouse-tracking and functional magnetic resonance imaging (fMRI) together in a value-based moral decision task, which allows us to quantify accumulative history responses as self-consistency. Using a multi-attribute time-dependent drift-diffusion model (tDDM), we disentangled the role of consistency and self-interest, highlighting the dominant role of cognitive-control-related regions. The drift rate of self-consistency was directly associated with the brain activity responsible for cognitive control, while the relationship between the reward and the activity of the related brain regions was mediated by the mouse-tracking index area under the curve(AUC). Dorsolateral prefrontal cortex was revealed as a hub connecting prACC and ventral striatum, whose functional connectivity was correlated with consistency drift rate and reward drift rate respectively. Furthermore, decision flexibility was quantified by choice entropy, which links to mouse tracking indices and the activity of cognitive - control related regions. Together, our study uncovers the interplay between self-consistency and reward in behavior, and highlights the key role of cognitive control in modulating these two attributes, thereby deepening our understanding of consistency and flexibility in moral decisions.

Unethical amnesia brain: Memory and metacognitive distortion induced by dishonesty

“…These results suggest that the interaction between DLPFC, a key component in cognition control and dishonesty 50 , and lTPJ which involves guilt processing 83,84 and self-image processing 50 , plays a crucial role in predicting the extent of metacognition change following dishonest behavior. Consistent with previous studies, moral versus immoral decisions are driven by both rewards and self-image 33,[85][86][87] . Dishonesty involves cognitive-control related brain regions like DLPFC and anterior cingulate cortex (ACC).…”

“…However, the current reward only affected future consistency in the next run, so current reward and current consistency could be viewed as two independent variables. Previous study 33 showed that both consistency and reward affected the dishonesty responses, we plotted the dishonesty probability as the function of reward and consistency, which was displayed in Figure S4D. This suggested that the participants considered both reward and consistency when they chose to lie or not in the IST.…”

“…On the item level, four types of transition probability for responses were conducted to calculate the choice entropy (for details, see supplemental materials). Then, we averaged the choice entropy across items as an assessment of the randomness 33 . First, the results showed that there was a significant main effect of condition on choice entropy (F 2,75 = 34.35, p < 0.001; η 2 g = 0.00096; Figure S4C).…”

“…Cognitive control plays a key role in (dis)honesty to override the moral default 50,52 and keep self-consistency 33 . Our main task in fMRI scanner employed rewards to motivate individuals to lie, so we first selected ROIs involving cognitive control and reward processing.…”

“…To manipulate moral conflict setting, reward and consistency, are two critical components in dishonesty 33 . Most motivated dishonesty studies are attributable to reward or self-interest 34 , and neuroimaging studies imply the involvement of the reward brain region in dishonesty [35][36][37][38][39] .…”