Beta activity in human anterior cingulate cortex mediates reward biases

Jiayang Xiao, Joshua A. Adkinson, John Myers, Anusha Allawala, Raissa Mathura, Victoria Pirtle, Ricardo A. Najera, Nicole R. Provenza, Eleonora Bartoli, Andrew J. Watrous, Denise Oswalt, Ron Gadot, Adrish Anand, Ben Shofty, Sanjay J. Mathew, Wayne K. Goodman, Nader Pouratian, Xaq Pitkow, Kelly R. Bijanki, Benjamin Y. Hayden, Sameer A. ShethNature Communications 2024 · DOI

TL;DR

Researchers have identified that beta-band (12–30 Hz) oscillations in the human anterior cingulate cortex (ACC) act as a neural "bridge" between past rewards and future decisions. This specific neural signature is significantly blunted in patients with treatment-resistant depression, offering a potential electrophysiological biomarker for anhedonia and a target for future neuromodulation therapies.

What Problem Does This Paper Address?

Human behavior is heavily influenced by "reward biasing"—the tendency to favor choices that have historically yielded better outcomes, even when sensory information is ambiguous. This adaptive mechanism allows us to maximize gains in uncertain environments. However, a hallmark of clinical depression is a breakdown in this process, manifesting as anhedonia (the inability to feel pleasure or motivation from rewards).

While the anterior cingulate cortex (ACC) has long been suspected as a hub for integrating reward information with executive control, the specific "neural language" it uses to mediate these biases remained unclear. This paper seeks to identify the electrophysiological signatures of reward learning and determine how these signals are altered in the depressed brain.

Methods

The study utilized a rare and powerful dataset: intracranial recordings (stereoelectroencephalography or sEEG) from two patient cohorts:

  1. Epilepsy Cohort (n=15): Patients undergoing seizure monitoring who did not have a mood disorder (serving as the primary model for "normal" reward processing).
  2. Depression Cohort (n=4): Patients with severe, treatment-resistant major depressive disorder (MDD) participating in a clinical trial for Deep Brain Stimulation (DBS).

The Task: Participants performed a Probabilistic Reward Task (PRT). They were shown cartoon faces with slightly different mouth lengths and asked to categorize them as "short" or "long." Crucially, one mouth type (the "rich" stimulus) was rewarded three times more frequently than the other (the "lean" stimulus).

Analysis: The researchers focused on Local Field Potentials (LFPs) across six frequency bands, with a primary hypothesis centered on the beta band (12–30 Hz). They analyzed neural activity during two critical windows: the delay period (after a choice is made but before feedback) and the feedback period (when the reward is revealed).

Datasets

The paper does not use standard public datasets; it relies on primary clinical data collected at the Baylor College of Medicine.

  • Intracranial LFP Data: Recorded from the ACC, medial orbitofrontal cortex (mOFC), lateral orbitofrontal cortex (lOFC), and amygdala.
  • Behavioral Data: Performance metrics from the PRT, specifically "response bias" (the systematic preference for the rich stimulus).

Key Insights and Ideas

  • Beta as a Reward Signal: While beta oscillations are traditionally associated with motor control or "maintaining the status quo," this study reinforces a growing view of beta as a carrier for motivational value. In the ACC, beta power was significantly higher when participants chose the "rich" stimulus compared to the "lean" one.
  • The ACC as an Integrator: The ACC didn't just react to rewards; its beta activity during the delay period predicted how much a participant would be biased on future trials. This suggests the ACC is where the brain "binds" reward history to sensory evidence to guide future behavior.
  • Neural Blunting in Depression: The most striking insight is the "flattening" of this system in depressed patients. These individuals showed both a reduced behavioral bias toward the rich stimulus and a corresponding lack of beta-power differentiation in the ACC. The brain essentially stops "caring" about the statistical advantage of one choice over another.
  • Anatomical Specificity: While the OFC and amygdala showed some reward-related activity, only the ACC consistently showed a positive relationship between beta power and reward across both the anticipation and receipt phases.

Key Results

  • Predictive Power: In the epilepsy cohort, ACC beta power (12–30 Hz) was significantly higher for rich vs. lean trials during the delay period ($p = 0.039$).
  • Behavioral Correlation: There was a significant positive correlation ($r = 0.38, p = 0.013$) between the difference in ACC beta response and the magnitude of the participant's behavioral response bias.
  • Depression vs. Control: Depressed patients showed a marked reduction in response bias compared to the epilepsy group ($p = 0.030$ in the final task block).
  • Blunted Feedback: The reward-induced increase in ACC beta power was significantly smaller and delayed in the depression cohort compared to the epilepsy cohort ($p < 10^{-4}$).

Limitations and Future Work

The most obvious limitation is the small sample size of the depression cohort (n=4). While the results are statistically significant, intracranial studies in psychiatric populations are inherently limited by the number of patients undergoing such invasive procedures.

Future Directions: * Non-invasive Validation: Can these ACC beta signatures be detected via high-density scalp EEG or MEG? If so, they could become a routine clinical tool for diagnosing or subtyping depression. * Closed-Loop Therapy: The authors suggest that ACC beta could serve as a "biomarker" for closed-loop DBS. A stimulator could theoretically monitor ACC beta levels and deliver a pulse specifically when reward-processing signals are flagging, "boosting" the impact of positive experiences for the patient. * Causal Links: Future studies using optogenetics (in animals) or targeted electrical stimulation (in humans) are needed to prove that increasing ACC beta power directly causes an increase in reward-biased behavior.

Summary updated

Abstract

The rewards that we get from our choices and actions can have a major influence on our future behavior. Understanding how reward biasing of behavior is implemented in the brain is important for many reasons, including the fact that diminution in reward biasing is a hallmark of clinical depression. We hypothesized that reward biasing is mediated by the anterior cingulate cortex (ACC), a cortical hub region associated with the integration of reward and executive control and with the etiology of depression. To test this hypothesis, we recorded neural activity during a biased judgment task in patients undergoing intracranial monitoring for either epilepsy or major depressive disorder. We found that beta (12-30 Hz) oscillations in the ACC predicted both associated reward and the size of the choice bias, and also tracked reward receipt, thereby predicting bias on future trials. We found reduced magnitude of bias in depressed patients, in whom the beta-specific effects were correspondingly reduced. Our findings suggest that ACC beta oscillations may orchestrate the learning of reward information to guide adaptive choice, and, more broadly, suggest a potential biomarker for anhedonia and point to future development of interventions to enhance reward impact for therapeutic benefit.
BibTeX
@inproceedings{xiao2024beta,
  author = {Jiayang Xiao and Joshua A. Adkinson and John Myers and Anusha Allawala and Raissa Mathura and Victoria Pirtle and Ricardo A. Najera and Nicole R. Provenza and Eleonora Bartoli and Andrew J. Watrous and Denise Oswalt and Ron Gadot and Adrish Anand and Ben Shofty and Sanjay J. Mathew and Wayne K. Goodman and Nader Pouratian and Xaq Pitkow and Kelly R. Bijanki and Benjamin Y. Hayden and Sameer A. Sheth},
  title = {{Beta activity in human anterior cingulate cortex mediates reward biases}},
  booktitle = {Nature Communications},
  year = {2024},
  doi = {10.1038/s41467-024-49600-7},
}
Figure