Hippocampal Temporal Interference Stimulation for Memory and Dual-Task Postural Control in Older Adults With Mild Cognitive Impairment

Retrieval accuracy is assessed using a computerized object-scene associative memory task programmed in E-Prime 3.0. Accuracy is calculated as the number of correct match/non-match responses divided by the total number of valid retrieval trials and multiplied by 100. The reported outcome is the post-intervention value minus the pre-intervention value within each stimulation period. Higher percentages indicate better associative memory retrieval performance.

Reaction time is recorded by E-Prime 3.0 during the retrieval phase of the computerized object-scene associative memory task. The measure is the mean reaction time across valid trials with correct responses. Trials with no response, reaction times shorter than 300 milliseconds, or reaction times more than 3 standard deviations above the participant-specific mean are excluded according to the prespecified data-cleaning procedure. The reported outcome is the post-intervention value minus the pre-intervention value within each stimulation period. Shorter reaction times indicate greater retrieval efficiency.

Task-state functional magnetic resonance imaging is acquired while participants perform the object-scene associative memory task. The measure is the mean blood-oxygen-level-dependent (BOLD) percent signal change within a prespecified anatomically defined left hippocampal region of interest during correct associative memory retrieval relative to the task control condition. The reported outcome is the post-intervention value minus the pre-intervention value within each stimulation period. Positive values indicate increased retrieval-related activation of the left hippocampus.

Participants perform the 3-meter Timed Up and Go test under single-task and dual-task conditions. During the dual-task condition, participants complete the test while continuously subtracting 3 from a randomly selected three-digit number. Total completion time is measured in seconds using the Mobility Lab wearable motion analysis system. Each condition is tested three times, and the mean completion time is used. Dual-task cost is calculated as: [(dual-task completion time - single-task completion time) / single-task completion time] × 100%. The reported outcome is the post-intervention value minus the pre-intervention value within each stimulation period. A higher dual-task cost indicates greater deterioration in mobility performance under cognitive load.

Participants complete the 7-meter walking test at their usual comfortable speed under single-task and dual-task conditions. During the dual-task condition, participants walk while continuously subtracting 3 from a randomly selected three-digit number. Walking speed is measured in meters per second using the Mobility Lab wearable motion analysis system. Each condition is tested three times, and the mean walking speed is used. Dual-task cost is calculated as: [(single-task walking speed - dual-task walking speed) / single-task walking speed] × 100%. The reported outcome is the post-intervention value minus the pre-intervention value within each stimulation period. A higher dual-task cost indicates a greater reduction in walking speed under cognitive load.

Safety is assessed using a standardized adverse-reaction questionnaire completed immediately after each stimulation session. Prespecified symptoms include tingling, itching, burning sensation, pain, skin redness, fatigue, visual sensations, difficulty concentrating, and mood changes. Each symptom is rated as none, mild, moderate, or severe, corresponding to scores of 0, 1, 2, and 3, respectively. Other spontaneously reported symptoms are also recorded. For each stimulation condition, a participant is counted as having an adverse symptom if at least one prespecified or spontaneously reported symptom is present with a severity score of 1 or higher during or immediately after stimulation.