Targeted Alternating Current and Cognitive Training Intervention for Cognitive Control in Bipolar Disorders (TACTIC-BD)

Veterans with bipolar disorder randomized to this arm receive 10 sessions of active frontal theta-frequency transcranial alternating current stimulation (tACS; 5Hz, 1mA, delivered via electrodes at F3/F4), administered concurrently with computerized cognitive training (BrainHQ) targeting working memory, inhibitory control, set-shifting, and attention. Sessions are self-administered at home under real-time remote supervision.

The Soterix 1x1 tES mini-CT is a compact, battery-powered transcranial electrical stimulation device manufactured by Soterix Medical, Inc., designed to deliver low-intensity electrical current (including tDCS and tACS waveforms) through scalp electrodes for research use. The device is part of Soterix's Remote Neuromodulation platform, specifically engineered to support safe, self-administered stimulation in the home setting under remote supervision. It is used in conjunction with the SNAPstrap headgear and SNAPpad electrodes for standardized, simplified electrode placement, and is programmed via subject- and session-specific, one-time-use activation codes that pre-set stimulation parameters (intensity, duration, and active/sham condition), preventing participant alteration of stimulation settings. The device is labeled by the manufacturer for investigational use only under U.S. federal law.

• Soterix Medical Remote Neuromodulation

Veterans with bipolar disorder randomized to this arm receive 10 sessions of sham (inactive) tACS, using identical electrode placement, session structure, and setup as the active arm, but without sustained active current delivery beyond a brief onset sensation. This arm receives the same concurrent cognitive training (BrainHQ) as the active arm, allowing isolation of tACS-specific effects beyond cognitive training alone.

The Soterix 1x1 tES mini-CT is a compact, battery-powered transcranial electrical stimulation device manufactured by Soterix Medical, Inc., designed to deliver low-intensity electrical current (including tDCS and tACS waveforms) through scalp electrodes for research use. The device is part of Soterix's Remote Neuromodulation platform, specifically engineered to support safe, self-administered stimulation in the home setting under remote supervision. It is used in conjunction with the SNAPstrap headgear and SNAPpad electrodes for standardized, simplified electrode placement, and is programmed via subject- and session-specific, one-time-use activation codes that pre-set stimulation parameters (intensity, duration, and active/sham condition), preventing participant alteration of stimulation settings. The device is labeled by the manufacturer for investigational use only under U.S. federal law.

• Soterix Medical Remote Neuromodulation

Veterans without major psychopathology complete the same baseline and follow-up assessment schedule (self-report, behavioral, and EEG measures of cognitive control) as the BD cohort, without receiving cognitive training, tACS, or sham tACS. This arm establishes practice-effect benchmarks to distinguish true intervention effects from repeated-testing effects. This arm is assigned by eligibility status, not randomization.

The Affective Go/NoGo task assesses response inhibition within an emotionally salient context, in contrast to purely non-affective measures of cognitive control. On each trial, a word with positive, negative, or neutral emotional valence is presented at the center of the screen; participants are instructed to respond as quickly as possible to target words of a designated valence (e.g., positive) while withholding responses to distractor words of the other two valences (e.g., negative or neutral). This design requires participants to inhibit prepotent responses to emotionally salient but task-irrelevant stimuli, capturing the interaction between affective processing and cognitive control. Affective response inhibition will be indexed by overall task accuracy using d', consistent with signal detection approaches to response inhibition performance.

The Dot Pattern Expectancy (DPX) is a dot-based variant of the AX-Continuous Performance Test (AX-CPT) used to assess cognitive control in the context of proactive response preparation and contextual updating. On each trial, participants are presented with a cue (white) followed by a probe (light blue), separated by a 2500-3500 ms interval; a target response is required only when a valid cue ("A") is followed by a valid probe ("X"), yielding four trial types: AX (target), AY, BX, and BY (nontarget). The outcome measure is overall task accuracy, calculated using the discriminability index (d'), a signal detection measure calculated as the z-transformed hit rate minus the z-transformed false alarm rate, adjusted for performance. The d' score is a continuous measure with a range of -4 to 4; higher d' scores indicate better discrimination between target and non-target trials, reflecting stronger cognitive control.

The Trail Making Test-B (TMT-B) is a widely used, paper-and-pencil neuropsychological measure of cognitive flexibility and set-shifting. Participants are asked to draw a line connecting a series of numbered and lettered circles in alternating sequential order (e.g., 1-A-2-B-3-C), as quickly and accurately as possible. Because successful performance requires continuously shifting attention between two different response sets, TMT-B is considered a sensitive index of executive/cognitive control, with longer completion times reflecting greater difficulty in adaptive, flexible responding. The primary outcome measure is time to completion.

EEG will be recorded during a Go/NoGo task in which participants are instructed to respond via button press to any letter other than "X" (Go trials) and to withhold response when an "X" appears (NoGo trials), while responding as quickly as possible and minimizing errors. Participants complete two blocks totaling 228 trials, with approximately 24% designated as NoGo trials; letters are presented for 2000 ms with a randomized intertrial interval of 800-1200 ms. Because Go trials occur more frequently than NoGo trials, the task design induces a prepotent tendency to respond, increasing demand on inhibitory control during NoGo trials. Behavioral accuracy will be indexed by d'. The primary neural outcome of interest is the event-related potential (ERP) time-locked to the button-press response reflecting conscious error detection, indexed by error-related positivity (Pe) amplitude.

The Monetary Incentive Delay (MID) task assesses neural responses associated with anticipating and receiving monetary outcomes. Participants complete 180 trials across three blocks; each trial begins with a 250 ms cue indicating a potential gain, loss, or neutral outcome, denoted by line markings corresponding to magnitude ($0.25-$5.00). Following a 1,500 ms fixation period, a brief target (180-280 ms) appears, requiring a speeded button press to win or avoid losing money; target duration is adaptively titrated to maintain approximately 70% accuracy across participants. Trial outcome feedback is presented for 1,650 ms, followed by a 500 ms display of cumulative earnings, and participants are compensated with a portion of their total winnings at the conclusion of the task. EEG is recorded continuously throughout the task, with frontal theta power during the reward anticipation period serving as the primary neural outcome measure.

The UPPS-P is a self-report measure conceptualizing impulsivity as a multi-faceted, multi-dimensional construct comprising five distinct trait subscales: Negative Urgency (tendency to act rashly during negative emotions), Lack of Premeditation (tendency to act without forethought), Lack of Perseverance (inability to remain focused on a task), Sensation Seeking (tendency to pursue novel and thrilling experiences), and Positive Urgency (tendency to act rashly during positive emotions). This study will use the 20-item short-form version of the UPPS-P, which has demonstrated high reliability and validity.

The MPQ-BF is a self-report personality assessment comprising 11 primary trait scales that load onto three higher-order temperament factors. This study will use the Control primary scale, which assesses trait impulsivity related primarily to response inhibition and lack of premeditation, as a self-report index of cognitive control.

A semi-structured exit interview will be administered to participants in the bipolar disorder (BD) cohort to assess acceptability of the combined tACS and cognitive training protocol and to inform future clinical implementation. The interview will assess participants' perceptions of each intervention component (tACS and cognitive training), including perceived benefits and concerns, suggestions for protocol modifications to enhance future benefit, and likelihood of recommending the intervention to other Veterans. All participants in the BD group will be approached for the exit interview, including those who do not complete all 10 intervention sessions, in order to capture factors contributing to low adherence; responses will be compared between active and sham tACS conditions to help distinguish concerns specific to the neuromodulation approach from those related to the cognitive training or overall study protocol.