State-dependent interoception, value-based decision-making, and introspection

National Institute Of Mental Health

Investigators

Linked publications & trials

Abstract

In the last fiscal year, Dr. Lopez-Guzman continued the recently established Unit on Computational Decision Neuroscience and expanded data collection. She has focused on staff recruitment, expansion of protocol objectives, and preliminary results. Protocol MH000978 (NCT05666726) is a large study aimed at understanding how stress, pain, and emotion induce acute (momentary) changes in decision-making (valuation) and metacognition (confidence). Participants complete decision-making and metacognition tasks after inducing a perturbation of their state (a negative affective state). So far, we have conducted over 350 in-person visits and 93 participants have completed our 3-visit protocol. Preliminary results suggest there is great interindividual variability in the change in risky decision-making and impulsivity in response to acute inductions of stress (n = 32) or incidental negative emotions (n = 49). Individual difference factors such as emotional metacognition seem to explain this variability. This suggests that contrary to what has been described in the literature, risky decision-making is affected by acute negative affect and stress but individuals with higher emotional metacognition may be more resilient and display less maladaptive decision-making when acutely stressed. Our computational modeling has allowed us to isolate that the target of acute stress is uncertainty valuation. Acute stress increases tolerance of uncertainty (ambiguity) more strongly for individuals with low emotional metacognition. Interestingly, we have also found that acute stress affects metacognition, by biasing participants’ confidence and reducing their local metacognition that their choice reflects their true preferences. This work in healthy volunteers is relevant for our future studies in clinical populations with depression, anxiety, and substance who may be highly susceptible to acute stress. We are beginning to prepare a manuscript describing this work (Haeffner C, Kao C-H, Trudeau J, Trowbridge, Raymond C, Blay-Tofey M, Lopez-Guzman (2025) In Preparation). In addition to these negative affective state induction experiments, we have also explored valuation and risky decision-making in the context of approach-avoidance conflict in healthy volunteers and youth with anxiety (in collaboration with Dr. Daniel Pine). We have found that trait anxiety and current mood correlate with specific computational parameters of valuation that correspond to the valuation of aversive outcomes. This has resulted in one submitted article (Feldman SB, Dayan O, Somerville Y, Kao CH, Cohen S, Glickman O, Ruiz SG, Newsome P, Linke JO, Shechner T, Pine DS, Lopez-Guzman S, Abend R. Behavior in motivational conflicts is determined by magnitude of potential outcomes and relates to anxiety levels. Res Sq [Preprint]. 2025) and one in preparation (Kao CH, Pine DS, Abend R, Lopez-Guzman S, 2025 In preparation). As these insights have clarified that stress can bias decision-making by modifying valuation, we have also expanded this study to investigate the role of stress in shaping the subjective experience derived from progressing towards a goal. Previous work has suggested that the hedonic experience of progress can be computationally modeled as the delta distance between two states (beginning and end) relative toa goal in a multidimensional state-space. We are testing the hypothesis that introducing an irrelevant acute stressor can disrupt this subjective representation of progress by distorting the state-space. To do this, we present healthy volunteers with a distressing video and a control video and compare their reported pleasure from playing an unrelated maze task where we can compute objective state-space progress. Our preliminary results confirm our hypothesis and indicate that the stress condition weakens the relationship between objective progress and hedonic experience. Metacognitive awareness protected against this effect. Given the behavioral results of our experiment, we are moving forward with a neuroimaging version of this study to investigate the neural mechanisms of this disruption which may be relevant for understanding the nature of consummatory anhedonia. Taken together, our results also point to the importance of metacognition, specifically emotional awareness and control, as a factor that may provide resilience to stress-induced behavioral changes and be potential treatment targets for emotional disorders.