Neurofeedback in Alzheimer's Disease

National Institutes of Health (NIH), National Institute on Deafness and Other Communication Disorders (NIDCD)

This study will evaluate behavioral and electrophysiological changes in a sample of adults with possible or probable Alzheimer's disease (AD), before and after undergoing training using a brain-computer interface (BCI) system with neurofeedback based on electroencephalography (EEG). Participants will repeatedly complete a letter viewing task and receive visual and auditory cues about their brainwaves. The study team hypothesizes that exposure to this EEG-based biofeedback intervention (neurofeedback) will result in a reduction of theta activity (brainwaves in the range of 4-8 Hz). The study team also predicts that exposure to the neurofeedback intervention will result in increased performance on reading, attention, and working memory tasks.

This study is a single case research design (SCRD) that will enroll adults with possible or probable AD. Participants will complete a 9 to 13 week testing schedule to deliver a BCI EEG-based neurofeedback intervention. All study visits will occur at participants' place of residence. Prospective participants will schedule an introductory consent/screening visit with the research team. All participants with AD will be required to enroll with a "study partner" (authorized representative or caretaker) to assist with scheduling and adherence as they progress through the study. Participants who pass screening criteria will continue to participate in the study. Eligible participants will complete between 3 and 7 weekly baseline assessments, beginning approximately 1 week after the screening visit. In accordance with SCRD standards, performance on outcome measures will be actively monitored during the baseline period. The intervention portion of the study will be triggered sometime during the 3-7 week baseline testing window, once participant performance on the primary outcomes is determined to be stable by the study team. Following baseline testing, participants will begin a 6-week intervention period with 3 BCI EEG-based neurofeedback sessions per week. A single follow-up visit will be completed approximately one month after the final intervention visit. Because of the inclusion of adults with decisional impairments in this study, participants with AD will be required to enroll with an accompanying "study partner" (e.g., spouse, caregiver or authorized representative), who will be required to attend all testing visits with the participant and assist with scheduling and adherence. Study partners will be formally enrolled and asked to complete a brief set of questionnaires (e.g., details about the primary participant and their interactions with the primary participant), but they will not receive other testing or intervention materials directly.

Three times per week, for six weeks, participants will receive biofeedback about the quality of their electroencephalograms (EEG; neurofeedback) during a letter identification task.

Participants receive intermittent visual and auditory cues to adjust their attentional engagement with a letter identification task.

EEG will be collected during the BCI letter identification task and spectral content will be assessed continuously throughout the task. Changes in theta band activity (4-8 Hz) between the start and end of the letter identification task will be quantified.

Once per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase); once, 4 weeks after the final intervention visit (follow-up)

EEG will be collected during the BCI letter identification task and spectral content will be assessed continuously throughout the task. Changes in average theta band activity (4-8 Hz) between subsequent study visits will be quantified.

Once per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase); once, 4 weeks after the final intervention visit (follow-up)

Participants must read and answer questions about the semantic accuracy of short sentences. Accuracy and speed are recorded.

Once per week for 9-13 weeks (throughout baseline and intervention phases); once, 4 weeks after the final intervention visit (follow-up)

Participants must identify all instances of a target letter (e.g., "Z") on a piece of paper. Speed of completion will be recorded in seconds.

Once per week for 9-13 weeks (throughout baseline and intervention phases); once, 4 weeks after the final intervention visit (follow-up)

Participants must identify all instances of a target letter (e.g., "Z") on a piece of paper. Accuracy will be quantified as the number of targets omitted (e.g., 2 targets missed).

Once per week for 9-13 weeks (throughout baseline and intervention phases); once, 4 weeks after the final intervention visit (follow-up)

Participants will view a series of letters and attempt to report them back to researchers either in sequence (forwards), or in reverse order (backwards). Letter series will differ in length, starting with two items and increasing in length. Accuracy will be reported and a working memory span will be generated based on performance.

Once per week for 9-13 weeks (throughout baseline and intervention phases); once, 4 weeks after the final intervention visit (follow-up)

Participants will answer a series of 22 self-reported likert-scaled questions (each item ranging 1-10) about their confidence and emotions regarding various aspects of reading. This inventory assesses general attitudes about the act of reading, and also perceived changes in reading ability for respondents who have experienced stroke, head injury, or other issues that may have affected their reading ability. Single item scores are not summed for this inventory.

Participants will read and/or listen to a series of short stories and answer questions about the content of the test passages.

Participants will listen to a series of digits and attempt to report them back to researchers either in sequence (forwards), in reverse order (backwards), or ordered based on value (sequencing). Digit series will differ in length, starting with two items and increasing in length, up to nine digits. Respondents continue within each trial (forward; backward; sequencing) until they incorrectly respond to two trials in a row for the same sequence length. Total number of points (1 point per correct trial) are tallied per condition: 16 points maximum per condition; 48 points total. This test provides a measure of working memory performance.

EEG signals acquired during the letter viewing task are used to generate on-line estimates of target and non-target EEG profiles. Model fitting accuracy (area under the curve; AUC) will be calculated from a calibration sub-task.

EEG signals acquired during the letter viewing task are used to generate on-line estimates of target and non-target EEG profiles during a calibration sub-task. Model fitting efficiency (information transfer rate; ITR) will be calculated from a copy-phrase sub-task as a measure of how quickly (bits/minute) the system is able to reverse-identify a target based on the pre-determined EEG profiles.

Weekly during baseline; three times per week during interventionOnce per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase)

EEG will be collected during the BCI letter identification task and spectral content will be assessed continuously throughout the task. Changes in alpha band activity (8-12 Hz) between the start and end of the letter identification task will be quantified.

Once per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase); once, 4 weeks after the final intervention visit (follow-up)

EEG will be collected during the BCI letter identification task and spectral content will be assessed continuously throughout the task. Changes in average theta band activity (8-12 Hz) between subsequent study visits will be quantified.

Once per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase); once, 4 weeks after the final intervention visit (follow-up)

EEG will be collected during the BCI letter identification task and spectral content will be assessed continuously throughout the task. Changes in beta band activity (13-30 Hz) between the start and end of the letter identification task will be quantified.

Once per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase); once, 4 weeks after the final intervention visit (follow-up)

EEG will be collected during the BCI letter identification task and spectral content will be assessed continuously throughout the task. Changes in average theta band activity (13-30 Hz) between subsequent study visits will be quantified.

Once per week for 3-7 weeks (baseline phase); three times per week for 6 weeks (intervention phase); once, 4 weeks after the final intervention visit (follow-up)

Inclusion Criteria: Consensus diagnosis of possible or probable AD as determined by the OHSU Alzheimer's Disease Center (ADC) weekly meeting Age 50-100 years old Possible or probable Alzheimer's disease as indicated by a Global Clinical Dementia Rating score of 0.5 or 1, with language impairment ≥0.5 on the supplemental Clinical Dementia Rating (CDR; form B4, section 2, #10) or similar clinical indicator of language difficulty Passed screening on the adapted BCI screening task presentations on a computer monitor, perceive visual/auditory feedback signals, and tolerate an EEG recording apparatus Reading impairment as measured by Discourse Comprehension Test Exclusion Criteria: Unstable medication regimen or use of EEG-altering prescription medications Anticipation of major medical interventions which may interrupt study proceedings, including upcoming surgeries Unwilling or unable to follow study protocol, including unstable schedule with frequent trips Note: AD participants must enroll with "study partners" (legally authorized representatives or designated caretakers) in order to participate. To qualify for the study, a study partner must spend an average of ≥10 hours per week interacting with the primary AD participant.

Galvin-McLaughlin D, Klee D, Memmott T, Peters B, Wiedrick J, Fried-Oken M, Oken B; Consortium for Accessible Multimodal Brain-Body Interfaces (CAMBI). Methodology and preliminary data on feasibility of a neurofeedback protocol to improve visual attention to letters in mild Alzheimer's disease. Contemp Clin Trials Commun. 2022 Jun 13;28:100950. doi: 10.1016/j.conctc.2022.100950. eCollection 2022 Aug.