The midbrain dopaminergic system is known to be critical for learning and updating reward expectations. However, how dopaminergic signaling contributes to pessimistic biases in reward expectation—both transient, trial-to-trial shifts and more stable, trait-like tendencies—remains unclear. The authors set out to determine whether distinct dopaminergic signatures accompany state-level versus trait-like pessimism in mice.
To address this question, the investigators used a Pavlovian ambiguous-cue task while recording mesolimbic dopaminergic activity. The design allowed the authors to examine responses to clearly predictive cues as well as cues that were ambiguous with respect to forthcoming reward. By analyzing trial-to-trial variability and differences across animals with stable biases in cue interpretation, the study separated short-term or state pessimism from longer-term or trait-like pessimism.
On a trial-by-trial basis, the authors report that episodes of state pessimism—moments when mice interpreted an ambiguous cue pessimistically—were associated with decreased dopaminergic activity during presentation of the ambiguous cues. Following these low-activity ambiguous-cue trials, the pattern of behavioral and neural responses shifted in a compensatory manner: there was a self-correcting positive shift in the balance of responses to actual reward versus reward omission. In other words, transient reductions in dopaminergic cue responses preceded adjustments that favored stronger signaling for reward outcomes relative to omission on subsequent trials.
Mice classified as having a pessimistically skewed interpretation of cues across sessions—referred to as trait-like pessimism—also exhibited reduced dopaminergic activity during ambiguous cues. Crucially, however, the dopaminergic signature associated with trait-like pessimism differed from that seen with transient state pessimism when outcomes were examined. Trait-like pessimistic mice showed a net negative shift in outcome signaling driven by relatively larger dopaminergic responses to reward omission compared with reward receipt. That is, the balance of outcome-related dopaminergic activity was shifted toward stronger omission signaling, producing a sustained negative bias in reward expectation.
The observed dissociation indicates that state and trait-like pessimism are supported by distinct dopaminergic mechanisms. State pessimism appears linked to transient reductions in cue-evoked dopamine with subsequent self-correcting adjustments in outcome signaling, whereas trait-like pessimism is associated with a persistent reweighting of outcome responses in which omission signals are amplified relative to reward signals. These differences suggest separate circuit or plasticity processes maintain short-term fluctuations versus long-term pessimistic biases.
Based on their data, the authors propose a model in which trait-like pessimism is maintained by asymmetrically weighted dopaminergic outcome signals: larger or more influential responses to reward omission relative to reward receipt produce a sustained negative bias in expectation. By contrast, state-level pessimism reflects momentary reductions in cue-evoked signaling that are followed by corrective shifts favoring reward signaling on subsequent trials.
These results are reported in a bioRxiv preprint that has not been certified by peer review. Details such as specific recording methods, sample sizes, statistical values, and full experimental controls are reported in the source preprint PDF and supplementary material; readers should consult the original manuscript for those technical particulars. Because this is an early report, the findings should be interpreted as provisional pending peer review and replication.
The authors disclosed funding sources that include an NIH grant and institutional and foundation support. The authors declared no competing interests. The full preprint contains additional author and funding information and links to supplementary material.
This summary is derived from the bioRxiv preprint titled “Dopaminergic signatures of state and trait-like pessimism in mice” by Srijani Biswas and Steven Shabel (doi: 10.64898/2026.07.30.741818). The article is a preprint and has not undergone peer review.