Low Reward and Heightened Threat Neural Activity are Both Associated with Heightened Anhedonia: A Transdiagnostic Perspective

Technical Summary: Low Reward and Heightened Threat Neural Activity are Both Associated with Heightened Anhedonia: A Transdiagnostic Perspective

Problem Statement
Traditional psychiatric neuroscience often investigates categorical diagnoses and brain systems in isolation. This approach is limited by high rates of comorbidity and within-diagnosis heterogeneity, failing to determine whether neural relationships are unique to specific symptom dimensions or common across disorders. Furthermore, the specificity of effects remains unknown when reward and threat systems are not evaluated simultaneously within the same individuals. The study addresses these gaps by examining associations between reward and threat neural activity and empirically derived symptom dimensions (General Distress, Fears, and Anhedonia-Apprehension) from the tri-level model of depression and anxiety.

Methodology

• Participants: The study utilized a sample of 339 participants aged 18–19 years, recruited from Los Angeles and Chicago as part of the "Brain, Motivation and Personality Development" (BrainMAPD) study. Recruitment was stratified based on self-reported trait Neuroticism and Reward Sensitivity to maximize variance. While 272 participants completed the fMRI protocol, the analysis retained all 339 participants using full-information maximum likelihood (FIML) to handle missing data.
• Experimental Tasks:
  • Monetary Incentive Delay (MID) Task: Assessed reward neural activity during anticipation and receipt of monetary rewards/losses. Regions of Interest (ROIs) included the ventral striatum, pallidum, putamen, and orbitofrontal cortex (OFC).
  • Pavlovian Fear Learning Task: Assessed threat neural activity across acquisition, extinction, and extinction recall phases. ROIs included the amygdala, bed nucleus of the stria terminalis (BNST), anterior insula, ventromedial prefrontal cortex (vmPFC), dorsal anterior cingulate cortex (dACC), and hippocampus.
• Symptom Measurement: Symptoms were measured using a multi-method tri-level dimensional model comprising General Distress (shared negative affect), Fears (anxiety-specific), and Anhedonia-Apprehension (depression-specific low positive affect). Data were derived from self-reports and interviewer-rated interviews (SCID-5) across multiple time points, utilizing latent intercepts from growth curve models to account for measurement error.
• Statistical Approach: Structural Equation Modeling (SEM) in Mplus was employed. The study used a latent construct approach to manage multiple comparisons, modeling reward and threat variables as latent factors where data justified. Regressions tested associations between symptom dimensions and neural variables, including tests for unique variance in specific ROIs and specificity across symptom dimensions and neural systems.

Key Results

• Reward Processing and Anhedonia:
  • Heightened Anhedonia-Apprehension was significantly associated with lower orbitofrontal cortex (OFC) activity during reward receipt in the MID task ($b = -.34, p = .021$).
  • Heightened Anhedonia-Apprehension was also associated with lower ventral striatal activation during reward anticipation. This association was driven by the unique variance of the ventral striatum ($b = -.53, p = .015$), as the broader reward-generating latent variable did not show a significant association.
  • These associations were specific to Anhedonia-Apprehension relative to General Distress and Fears, and specific to reward processing relative to threat processing.
• Threat Processing and Anhedonia (Unexpected Finding):
  • Contrary to the hypothesis that General Distress or Fears would relate to threat activity, Anhedonia-Apprehension was positively associated with heightened threat neural activity during the extinction phase of the fear learning task ($b = .21, p = .001$).
  • This association was specific to the extinction phase (stronger than acquisition or extinction recall) and specific to Anhedonia-Apprehension.
  • The effect was driven by the latent threat variable as a whole; no individual ROI within the threat latent variable showed unique variance associated with Anhedonia-Apprehension.
• Null Findings: Neither General Distress nor Fears showed significant associations with reward or threat neural activity in any phase of the tasks.

Significance and Claims
The paper claims that its findings highlight the importance of examining associations between specific brain systems and empirically derived symptom dimensions rather than categorical diagnoses.

• Transdiagnostic Specificity: The study demonstrates that low reward-related brain activity (specifically in the OFC and ventral striatum) is uniquely linked to the Anhedonia-Apprehension dimension, supporting the utility of dimensional models over categorical ones.
• Novel Threat-Anhedonia Link: The unexpected finding that heightened anhedonia is associated with heightened threat activity during fear extinction suggests a complex interplay where low positive affect may impair reward-dependent prediction error learning necessary for fear extinction.
• Methodological Contribution: By using a latent construct approach and FIML, the study successfully disentangled shared versus unique variance across brain regions and symptom dimensions, providing a more nuanced view of neural correlates in internalizing psychopathology.
• Limitations and Cautions: The authors explicitly note that the cross-sectional design precludes causal inference and that the sample consists of college students, necessitating replication in more diverse and longitudinal contexts. They also caution that the association between anhedonia and threat extinction requires further investigation in independent samples to confirm whether low positive affect drives fear persistence.