The MedX sham intervention device is identical in all respects to the active device, except that the MedX console and diode clusters were modified to NOT deliver NIR light when turned on. The sham MedX devices were modified to deliver 'warmth', similar to that of the active devices. As with the active condition, a total of six sham interventions were given over a 2-week period, following the identical procedures described in the active condition.

Gait is assessed using the Primary Gait Screen (Schmidt et al., 2019) that requires participants to walk the length of an 8-meter pressure sensitive mat (Zeno Walkway, 120Hz, Zeno Metrics), turn around, and return to the beginning of the walkway. Gait variability during forward walking was selected because in PD, greater variability is associated with increased number of falls and is one of the most disabling symptoms in PD. For this outcome, variability in stride length is indexed by the individual's standard deviation (SD) of stride length measured in centimeters; higher SD scores indicate worse performance. A change score is calculated by subtracting baseline SD of stride length from the post-intervention scores. Negative change scores mean better performance, whereas positive change scores mean worse performance.

The RAVLT is a commonly used memory task in clinical settings. It is a 15-item word list learning task given over 5 trials followed by delayed recall of the list 20-30 minutes later. The major outcome is number of items freely recalled after the delay. Scores can range from 0 to 16 (maximum number of words), with higher score reflecting better performance. A change score is calculated by subtracting the baseline score from the post-intervention score. Higher difference scores indicate better performance due to intervention.

The Negative Affect scale is derived from the emotion module of the NIH Toolbox, a 12-22 minute self-report assessment that surveys current emotion health. The Negative Affect Scale is based on ratings of a series of emotion words. Questions comprise Likert-type items using computerized adaptive testing based on item response theory, resulting in a normed T score (mean of 50, SD of 10). The Negative Affect scale includes items pertaining to fear, anger, sadness. Scores below 40T indicate low levels of negative affect and scores above 60T indicate higher levels of negative affect. A difference score is computed by subtracting the baseline score from the post-intervention score.

Parkinson participants randomized to the Sham group underwent identical procedures as those in the 'active' group. However, during the intervention phase, the near infrared deliver device (MedX console) was modified and did not deliver NIR light when turned on. The Sham condition involved 6 intervention sessions over a 2-week period. Each session consisted of 40 minutes of sham NIR stimulation using a modified MedX delivery system.

Total of all reporting groups

The Montreal Cognitive Assessment (MoCA) is a widely used cognitive screener for dementia that was administered prior to beginning the trial. The MoCA scale ranges from 0 to 30 points, with higher scores reflecting better cognition. Participant scores were screened using age, education, sex, and race based norms from >8300 older adults (Sachs et al., 2020).

This refers to duration of Parkinson motor symptoms in years.

This refers to number of Parkinson participants whose initial motor symptoms were characterized as primarily involving tremors (versus a non-tremor presentation involving motor rigidity). This is important as some research suggests that tremor predominant PD is associated with less rapid progression of cognitive and other symptoms.

In the SHAM arm, Parkinson participants underwent identical intervention procedures as those in the 'active' group. However, the sham MedX device did not deliver NIR light when turned on. The Sham condition involved 6 sessions over a two-week period. Each session involved 40 minutes of sham NIR stimulation using modified MedX delivery system that delivered no NIR light.

ARENA is a task of spatial memory-navigation that has been linked to hippocampal function and is a human analogue to the Morris water maze, which has shown sensitivity to NIR stimulation in Alzheimer's transgenic mice. ARENA requires participants to learn and navigate to a hidden target location in a simulated environment. It involves use of a joystick over a series of 8 learning trials and one final probe trial. On each learning trial, the path length and time to reach the target are recorded. The Learning dependent variable is a composite score consisting of mean z-scores for path length and for time to reach the target (Learning Composite). A change score is computed by subtracting the baseline Learning Composite from the post-intervention Learning Composite z-score. Higher scores mean a better outcome.

Participants included nondemented individuals with Parkinson disease who were randomized to active or sham intervention groups. During outcome testing, PD participants were evaluated "off" their normal dose of dopamine medication, following the standard overnight washout period.

ARENA is a computer-based task of spatial memory-navigation that has been linked to hippocampal function and is a human analogue to the Morris water maze. ARENA consists of 9 learning trials and one final probe trial. On each learning trial, the path length and time to reach the target are recorded. On each probe trial, the percent time spent in the spatial quadrant where the target is located is recorded. The dependent variable is a total composite score consisting of mean z-scores for path length, time to reach the target, and %time in the target quadrant during the probe trial (Total Composite). A change score is computed by subtracting the baseline Total Composite z-score from the post-intervention Composite z scores. Higher scores mean better outcome.

Participants included nondemented individuals with Parkinson disease who were randomized to active or sham intervention groups. During outcome testing, PD participants were "off" their normal dose of dopamine medication, following the standard overnight washout period.

Gait is assessed using the Primary Gait Screen (Schmidt et al., 2019) that requires participants to walk the length of an 8-meter pressure sensitive mat (Zeno Walkway, 120Hz, Zeno Metrics), turn around, and return to the beginning of the walkway. Gait variability during forward walking was selected because in PD, greater variability is associated with increased number of falls and is one of the most disabling symptoms in PD. For this outcome, variability in stride length is indexed by the individual's standard deviation (SD) of stride length measured in centimeters; higher SD scores indicate worse performance. A change score is calculated by subtracting baseline SD of stride length from the post-intervention scores. Negative change scores mean better performance, whereas positive change scores mean worse performance.

Participants included nondemented individuals with Parkinson disease who were randomized to active or sham intervention groups. During outcome testing, PD participants were "off" their normal dose of dopamine medication, following the standard overnight washout period.

This Working Memory composite is derived from a computer-based battery of executive functioning tasks (NIH Examiner). This composite was chosen due to known working memory difficulties in individuals with Parkinson disease. The Working Memory composite consists of scores from the N-back task and a Dot Counting task, which are combined to yield a composite score that ranges from -3 to 3.0; higher scores correspond to better working memory. A change score is calculated by subtracting the baseline Working Memory Composite from the post-intervention Working Memory Composite.

Participants included nondemented individuals with Parkinson disease who were randomized to active or sham intervention groups. During outcome testing, PD participants were "off" their normal dose of dopamine medication, following the standard overnight washout period.

The RAVLT is a commonly used memory task in clinical settings. It is a 15-item word list learning task given over 5 trials followed by delayed recall of the list 20-30 minutes later. The major outcome is number of items freely recalled after the delay. Scores can range from 0 to 16 (maximum number of words), with higher score reflecting better performance. A change score is calculated by subtracting the baseline score from the post-intervention score. Higher difference scores indicate better performance due to intervention.

Participants included nondemented individuals with Parkinson disease who were randomized to active or sham intervention groups. During outcome testing, PD participants were "off" their normal dose of dopamine medication, following the standard overnight washout period.

The Negative Affect scale is derived from the emotion module of the NIH Toolbox, a 12-22 minute self-report assessment that surveys current emotion health. The Negative Affect Scale is based on ratings of a series of emotion words. Questions comprise Likert-type items using computerized adaptive testing based on item response theory, resulting in a normed T score (mean of 50, SD of 10). The Negative Affect scale includes items pertaining to fear, anger, sadness. Scores below 40T indicate low levels of negative affect and scores above 60T indicate higher levels of negative affect. A difference score is computed by subtracting the baseline score from the post-intervention score.

Individuals with idiopathic Parkinson disease who were tested 'off' dopaminergic medications

Part III of the Unified Parkinson Disease Rating Scale (UPDRS) is a standard clinical rating scale for motor symptoms in individuals with Parkinson disease. It is clinician administered by a trained rater and involves ratings of motor tremors, rigidity, and slowness. In the research setting, the scale is videotaped for later scoring by a trained rater who is blinded to group assignment. The Part III motor scale score ranges from 0 to 108 and involves 27 responses to 14 questions. Higher scores are worse and reflect more severe motor symptoms. A change score is computed by subtracting baseline UPDRS motor score from the post-intervention score. Greater change in negative direction reflects better outcome.

Patients with Parkinson disease who were tested "off" dopamine medication following standard overnight washout.

Parkinson's Disease participants in the sham control group will undergo identical procedures as the intervention group - screening, baseline testing, and LED cluster placement procedures. However, during the near infrared (NIR) session, the MedX console will not be turned on and no active stimulation will be applied.

Sham MedX 1116 Rehab Console: This intervention makes use of sham transcranially applied near infrared light using light emitting diodes. Light emitting diodes (but not near-infrared light) will be applied to the scalp for a period of 1 hour at each session. A total of six session will be held.