Objective
Suicidal thoughts and behaviors (STBs) partially stem from abnormalities in multiple neural systems. As a crucial process that modulates emotion, cognition, and behavioral decision-making, reward processing encompasses three primary stages—reward valuation, reward responsiveness, and reward learning—whose dysfunctions may shed light on the potential neural mechanisms underlying STBs. Event-related potentials (ERPs), a neurophysiological measure with high temporal resolution, capture key features of cognitive and emotional processing in individuals with STBs. This review aims to summarize the performance of individuals with STBs on ERP components implicated in reward processing and discuss their clinical significance.
Methods
Through a literature search focusing on STBs and reward processing, ERP components were categorized into three stages: the reward anticipation stage (including cue-P2, cue-P3, CNV, and SPN), the reward consumption stage (including FRN/RewP, FB-P3, and LPP), and the reward learning stage (FRN/RewP). We systematically reviewed the aberrant ERP findings in individuals with STBs, analyzed how these findings relate to emotion regulation, cognitive function, and behavioral decision-making during reward processing, and explored the potential of these components as risk markers of STBs.
Results & Discussion
(1). Reward Valuation In individuals with STBs, the primary ERP components at the reward valuation stage include cue-P2, cue-P3, CNV, and SPN. Research indicates significant differences in cue-P2 and cue-P3 amplitudes among groups with varying levels of suicide risk during the Monetary Incentive Delay (MID) task. Specifically, individuals with suicidal ideation (SI) show elevated cue-P2 amplitudes when confronted with punishment, reflecting heightened vigilance to negative cues. By contrast, individuals with suicide attempts (SA) demonstrate significantly decreased cue-P3 amplitudes in response to rewards, suggesting reduced reward-prediction sensitivity and deficits in motivational regulation.Research on CNV and SPN is relatively limited. Early studies found that those who had attempted suicide exhibit significantly reduced CNV amplitude, hinting at insufficient cortical resource mobilization and cognitive preparatory capacity. However, other work using the MID task did not find significant group differences in CNV between SA and non-SA participants, suggesting that CNV may be less sensitive to detecting SA or is influenced by other cognitive processes. Further investigation is required to clarify its role.
(2). Reward Responsiveness The ERP components at the reward responsiveness stage, including FRN/RewP, FB-P3, and LPP, also exhibit notable differences in individuals with SI and SA. In SI individuals, the RewP amplitude in response to positive feedback is diminished, whereas their response to negative feedback is enhanced, indicating impaired reactions to positive outcomes and heightened sensitivity to negative outcomes—possibly reflecting aberrant emotion regulation. In SA individuals, lower ΔFN and higher FRN amplitudes suggest increased sensitivity to negative feedback and a proclivity toward impulsive decision-making.Variations in FB-P3 may also differentiate SI and SA. Under negative feedback conditions, SA individuals exhibit higher FB-P3 amplitudes compared to reward and neutral conditions; however, no significant differences were observed between the SI group and healthy controls (HC). These findings underscore the neural characteristics of SI and SA in feedback processing and emotion regulation. Furthermore, SA individuals show markedly enhanced LPP amplitude under punishment conditions, indicating an excessive response to threatening stimuli, which may be related to pain-avoidance processes.
(3). Reward Learning While the ERP indices of the reward learning stage partially overlap with those of the reward responsiveness stage, the distinct abnormalities are more evident through specific behavioral tasks. In the Iowa Gambling Task (IGT), SA individuals display significantly reduced ΔFN and increased FRN amplitudes, reflecting an overemphasis on punishment signals and an impulsive decision-making tendency that neglects longer-term rewards. In the Ultimatum Game, SA individuals exhibit notably enhanced FRN amplitudes in response to unfair offers, implying a heightened concern for fairness accompanied by diminished consideration of punishment costs.In studies involving major depressive samples, extended FRN latency during the Balloon Analogue Risk Task (BART) correlates with increased suicidal ideation, indicating decreased processing efficiency for negative feedback and poorer adaptation to errors or failures. This also suggests a potential predictive value of FRN for suicidal ideation. Conversely, individuals with SI exhibit diminished ΔRewP in guessing tasks, pointing to reduced responsiveness to positive reward feedback. Among adolescents with active SI, simpler guessing tasks reveal heightened RewP in response to positive feedback and attenuated responses to negative feedback, suggesting an emotionally biased reward-processing pattern that may exacerbate suicidal risk behavior.
Conclusions
Individuals with STBs manifest distinct ERP abnormalities across different stages of reward processing. These include the cue-P2, cue-P3, CNV, and SPN components during reward anticipation; the FRN/RewP, FB-P3, and LPP during reward responsiveness; and FRN/RewP during reward learning. Such neurophysiological alterations underscore potential neural mechanisms of STBs within the context of reward processing. They may serve as neurobiological markers for predicting STBs, thereby illuminating novel directions for risk assessment and intervention strategies. Nevertheless, most existing findings derive from cross-sectional studies, underscoring the need for further replication and more in-depth research to establish causal and longitudinal insights.

suicide;reward;event-related potential