Effects of Social Feedback on Intracranial EEG

This study aims to assess the impact of social influence on direct neural recordings in human patients undergoing surgical treatment of epilepsy. A total of 24 patients, ranging in age from 18 to 65 years, will be recruited from the Epilepsy Center at Dartmouth Health Medical Center (DHMC), which serves a catchment area of approximately 2.5 million people in Northern New England. Patients eligible for intracranial electroencephalography (iEEG) are determined by a clinical-based multidisciplinary board, with 10-20 iEEG studies performed annually in the Epilepsy Monitoring Unit.

Each patient will undergo electrode implantation throughout the brain to localize epileptogenic zones, with 12-20 electrode leads implanted, each containing 8-18 channels. This results in a total of 150 to 220 channels of continuous electrophysiological data per patient. The iEEG signals will be recorded using a Natus clinical system, and precise timing synchronization will be ensured between the iEEG recordings and the experimental paradigm presentation.

The protocol will involve a single one-hour session for each patient. Patients will participate in a doctor rating task, where they provide impressions of warmth, competence, and other qualities related to the doctors presented in the experiment. Additionally, patients will perform a simplified version of the Multimodal Negative Affect Task (MNAT) used in Experiment 2, with the omission of pain stimuli to increase patient acceptability and reduce time and fatigue burden. Trial blocks will consist of brief videos of the doctors explaining what to expect in the upcoming trials, followed by four task trials involving fear or math stimuli from the MNAT task. Each doctor will have four threat-suggestion videos and four safety-suggestion videos, resulting in a total of 144 trials (8 doctors X 2 low/high threat cues X 3 repetitions x 4 trials per block).

The analysis of the neural data will involve time-frequency decomposition of local field potentials (LFPs) to examine broadband gamma-frequency power (40-200 Hz) locked to cue and stimulus onset time. Effects of cues (high vs. low-expectancy), self-reported fear or effort within each cue condition, and their interaction on broadband gamma signals will be assessed for each electrode contact. A focus will be placed on affect-encoding electrodes showing relationships with fear and/or effort within conditions. Proportions of affect-encoding electrodes and other contacts demonstrating social influence effects will be quantified. The analysis will also explore assimilation and contrast effects with cue-predicted values, investigating whether social information influences the neural correlates of affective experience.

To ensure appropriate sample size, it is anticipated that 10-15 eligible patients will participate each year during the project period. Biological sex differences and age will be controlled for in all analyses, and hypotheses will be tested while considering these factors. The study aims to provide valuable insights into the effects of social influence on neural activity, expanding our understanding of the mechanisms underlying placebo effects and affective experiences.