Cognitive Augmentation Via Multimodal Sensing and Auricular Neurostimulation

Overview

The goal of this clinical trial is to extend this period of optimal cognitive performance by applying neurostimulation to buffer health volunteers against the effects of increased levels of stress, distraction, and cybersickness. The main questions it aims to answer are: * Can we use OpenBCI's head-mounted Galea biosensor + eXtended Reality (XR) platform to measure participants' cognitive state in relation to stress, attention and cybersickness? * How does applying external neurostimulation via Spark Biomedical's Sparrow Link transcutaneous auricular neurostimulation (tAN) system enhance cognitive performance with a closed-loop interface that automatically applies neurostimulation as a function of physiologically determined stress, attention, and cybersickness metrics? Researchers will compare the active neurostimulation group to the sham neurostimulation group to see if cognitive performance is improved with stimulation. Participants will complete 4 virtual reality tasks in the lab: * 2 tasks related to attention - Flanker and Gradual-onset Continual Performance Task (GradCPT) * The Multi-Attribute Task Battery (MATB) * A cybersickness task * And a baseline session before each task * Neurostimulation intervention will occur in response to cognitive states

The objective of this study is to determine if active tAN stimulation can form a closed-loop system with the Galea biosensing headset to enhance cognitive performance. The study is designed as a randomized, blinded, sham-controlled trial to test the effects of active tAN stimulation in cognitive stimulation tasks (Flanker task, GradCPT, MATB and cybersickness tasks) towards improvement of cognitive performance. The study occurs in three phases. * Phase I is designed to establish a quantifiable relationship between biometric data and cognitive states (cognitive load, stress, attention and cybersickness) under cognitive stimulation tasks (Flanker task, GradCPT, MATB and cybersickness tasks). * Phase II involves evaluation of manual tAN to affect cognitive state in an open-loop paradigm. * The third phase involves using the cognitive state quantification from Phase I, and the quantified effects of tAN discovered in Phase II to examine a closed-loop cognitive augmentation system. Preliminary Cognitive State Assessment: Phase I of the experiment will establish a cognitive task performance baseline, and also allow for quantification of cognitive state metrics based on biomarkers. Participants will perform the following tasks: Flanker Task, GradCPT, MATB, and a cybersickness stimulation task. Each of these tasks have different quantification and scoring systems to evaluate performance, and each aim to induce different cognitive stimulation. Before each task, an idle baseline of cognitive state is recorded for 10 minutes. Open-loop Intervention: In Phase II, all participants (Groups 1-2) will undergo the same set of tasks, this time while wearing the Sparrow Link tAN device, and Group 1 will experience active tAN, controlled by the investigator to determine optimal settings for affecting cognitive state. Closed-loop Intervention: In Phase III, the information on quantifying cognitive state determined in phase I and information on affecting cognitive state determined in phase II will be used to create a closed-loop intervention system using active tAN (Group 1 only), triggered based on conditions determined by the investigating team. Both groups will undergo the same set of tasks as in Phase I and Phase II, but Group 1 will receive active tAN and Group 2 will not. Tasks Experimental paradigms and mechanisms that have been shown to induce stress, loss of attention, and cybersickness will be used to elicit physiological responses from study participants. These responses will be used to generate data for automatic characterization of the target cognitive states, determine how different levels of manual tAN affects these states, and test the efficacy of a closed-loop neurostimulation system. These paradigms and mechanisms are described below. Multi-Attribute Task Battery (MATB) The Multi-Attribute Task Battery is a computer-based task designed to evaluate operator performance and workload. MATB provides a benchmark set of tasks analogous to activities that aircraft crew members perform in flight, although it can also be used by non-pilot individuals. The MATB requires simultaneous performance of monitoring dynamic resource management, and task tracking. The simultaneous performance of multiple tasks is a central feature of the MATB, and is the feature that is consistent with most operational systems, making it useful as a research platform. In this case, we will use OpenMATB, an open-source implementation of the MATB experimental environment written in Python (Cegarra, et al. 2020). Flanker Task The flanker task can be used to measure processing and selective attention (Eriksen, 1974). The flanker task requires participants to respond to a central target stimulus while ignoring flanking stimuli that may or may not be congruent with the target stimulus (Eriksen, 1974). These tasks become more challenging over time either by decreasing the time allowed for reactions or by introducing more conflicting information. A subject's responses and error rate can be used to gauge their attention and stress levels. These metrics, combined with self-reported data, can be used to create labeled datasets. The task may be modified with numbers or shapes to add more variation or complexity, as required. GradCPT A continuous performance task requires the participants to respond to frequent stimuli and inhibit responding to infrequent stimuli (Robertson, et al., 1997). GradCPT eliminates the offsets and onsets of visual stimuli between trials by using gradual transitions. As a result, GradCPT is more dependent on internal attention control and useful for studying sustained attention processes using physiological methods. The GradCPT task is used to test a subject's sustained and selective attention as well as response inhibition. Insufficient attention to tasks can result in unintended or incorrect inputs. Images show the first image transitioning to the second at 100%, 75%, 50%, 25% and 0% image coherence. Participants are required to press a button each time a city scene is presented (90% of trials) and withhold responses when infrequent mountain scenes are presented on the remaining 10% of trials (Fortenbaugh, et al., 2017). The initial transition period is 800ms, and can be randomized or decreased over time to increase task difficulty and variance. Cybersickness Stimulation Mechanisms Cybersickness relates to the tendency of some users to exhibit symptoms similar to classical motion sickness both during and after a virtual reality (VR) experience. It is distinct from motion sickness in that the user is often stationary but has a compelling sense of self motion as a result of the immersive visual aspects of the simulation (LaViola Jr, 2000). The symptoms of cybersickness include dizziness, nausea, and lightheadedness (Davis \& Nalivaiko, 2014), and there are several factors that are known to induce cybersickness including latency (Stauffert et al. 2020) and loss of control (Davis \& Nalivaiko, 2014). Prevalence of sensitivity to cybersickness ranges from 20-95% with immersive VR experiences (Yildirim, 2020). Sensing Technologies This section describes the technologies that will be used during the experiment to record physiological responses to stimuli and characterize cognitive state metrics. Electroencephalography (EEG) Electroencephalography (EEG) records brain electrical activity via placement of electrodes on the scalp. One major advantage of using EEG is that it is noninvasive, meaning it is a safe and low-risk method for studying brain activity. It is also relatively inexpensive when compared to fMRI or MEG. EEG has a high temporal resolution and captures events on a millisecond time scale. However, the disadvantage of EEG is that it has very limited spatial resolution as the electrodes are placed on the scalp and there are multiple intermediate layers of non-neuronal tissue in between the brain and electrodes. This causes the signals to be dispersed and heavily biased towards peripheral structures in the brain. EEG is a useful tool for estimating cognitive states and emotion, and becomes more powerful when combined with other sensing modalities that monitor other physiological markers. The combination of EEG with other modalities increases the probability of detection and quantification of cognitive states (Kartsch et al., 2018; Yasemin et al., 2019; Antonenko et al., 2010). The Galea biosensing suite supports ten active EEG electrodes and two passive EEG electrodes. The ten active electrodes are located along the midline and in the frontal, parietal and occipital lobes. Specifically, their locations are F1, F2, C3, CZ, C4, P3, P4, PZ, O1, O2 as denoted by the 10-10 system for EEG recording. The two passive electrodes are located in the headset face mask and measure the Fp1 and Fp2 locations as denoted by the 10-10 system. The active EEG electrodes included in the Galea system use combs formed from conductive polymer and provide stable contact with the head through hair and other obstacles. The passive EEG electrodes are made from silver/silver chloride (Ag/AgCl) and make contact with the forehead of the subjects. The device has two attached clips that attach to the ears of the user. These ear clips provide common-mode signal rejection and serve as a reference voltage for the EEG electrodes. Face Electromyography (EMG) Electromyography (EMG) relates to the measurement of muscles and the nerve cells (motor neurons) that control them. To measure EMG, electrodes are placed on the skin or inserted into the muscle to detect electrical activity. The resulting signals are then analyzed to provide information about the functioning of muscles and nerves. The Galea biosensing suite includes dry EMG electrodes around a face mask that attaches to the XR component of the headset. These electrodes measure facial muscle activity and allow for estimation of face poses when the face is covered by the XR headset. Three pairs of electrodes are placed above the eyebrows in different orientations in order to target the Frontalis muscles that raise the eyebrows, as well as the Depressor Glabellae, Depressor Supercilli, and Currugator muscles that allow the eyebrows to lower. Additional electrode pairs are placed on each of the cheeks to target the Zygomatic Major and Zygomatic Minor muscles that allow the lips and cheeks to move. The electrodes are standard dry, flat electrodes coated in Ag/AgCl. Photoplethysmography (PPG) Photoplethysmography (PPG) is a non-invasive method of measuring blood flow in blood vessels close to the skin's surface. PPG provides information about the volume of blood in a particular area and can be used to infer the pulse rate, oxygen saturation levels. This information can contribute to the quantification of cognitive states by monitoring changes in physiological parameters that are associated with mental activities. PPG can be used to measure pulse rate variability, which closely approximates heart rate variability (Chuang et al., 2015). Heart rate variability (HRV) is a measure of the variation in time between successive heart beats. It provides information about the autonomic nervous system's activity, which regulates functions such as heart rate, breathing, and blood pressure. Studies have shown that HRV can be a useful indicator of cognitive states (McDuff et al., 2014). For example, low HRV has been associated with high levels of stress, anxiety, and depression, while high HRV has been associated with improved mood, cognitive performance, and overall well-being (McDuff et al., 2014). HRV can also provide information about the balance between the sympathetic and parasympathetic nervous systems. High HRV indicates a balance between the two systems, while low HRV indicates that one system is dominant over the other. In addition, HRV has been found to be a reliable indicator of mental effort and cognitive load (Solhjoo et al., 2019). For example, HRV has been found to decrease when individuals are engaged in mentally demanding tasks, such as problem-solving or decision-making. This decrease in HRV can be used as a marker of mental effort and cognitive load, and it can be used to monitor changes in cognitive states over time. A PPG sensor is integrated into one of the ear clips attached to the Galea biosensing suite. The clip is placed on the earlobe, which results in very high signal-to-noise (SNR) signals, relative to other locations and is a sufficient replacement for heart rate monitoring when compared to electrocardiogram (Vescio et al., 2018; Weiler et al., 2017). Electrodermal Activity (EDA) Electrodermal activity (EDA) refers to changes in the electrical conductance of the skin which are caused by changes in sweat gland activity. EDA can be measured by attaching electrodes to the skin and recording changes in skin conductance or impedance. Both methods provide adequate information on the changes in skin conductance. Sweat gland activity is often associated with increased emotional arousal, particularly with stress and anxiety, and EDA is a well established physiological measure that is used to assess changes in arousal, attention and emotional states (Leiner et al., 2012). However, the limitation of EDA is that it only provides information about changes in skin conductance relative to a baseline, and does not indicate what specific cognitive state the change in conductance is linked to. Galea measures EDA from the forehead, which provides a similar level of information to more peripheral measurement locations (Hossain et al., 2022). Electrooculography (EOG) The Galea biosensing suite supports integrated EOG, which measures eye electrical activity. The system detects eye movements along the vertical and horizontal axes. When these movements occur, the eye movement acts as a dipole and causes large measurable fluctuations that can be recorded. Although exact eye positions are difficult to estimate with EOG, it is possible to use EOG information to estimate attention and other cognitive metrics in the context of other neural and physiological signals (Perdiz et al., 2017; Morton et al., 2022). The Galea's EOG sensors use the same flat Ag/AgCl electrodes that the EMG face sensors. They are placed above and below the right eye for vertical EOG and on the outside periphery of the left and right eyes for horizontal EOG. Eye Tracking and Pupillometry Eye gaze tracking is achieved using infrared imaging sensors that map eye movements in real time. Eye gaze can be an indirect measure of attention, and how individuals shift attention between different stimuli and prioritize attentional resources (Ziv, 2016). Eye tracking combined with pupillometry and other physiological monitoring techniques can allow for insight regarding loss in attention due to stress and other factors (Miller \& Unsworth, 2020). Eye gaze tracking is useful for understanding when a subject is focused on a stimulus and when they shift their focus (Armstrong and Olatunji, 2012). Pupillometry is the measure of pupil size changes in response to different scenarios or stimuli. Pupillometry is critical in the field of cognitive augmentation because it is an indirect measure of adrenergic activity, particularly norepinephrine activity in the brain. This is useful because norepinephrine is associated with arousal and attention. Norepinephrine is also involved in the regulation of the pupil, as it causes pupil dilation by activating the dilator muscles in the iris. The ability to detect adrenergic activity and then modulate it with tAN allows for the mitigation of various stress and motion sickness related symptoms (Badran et al., 2018; Molefi et al., 2023). The Galea biosensing suite includes an integrated Varjo XR-3 head-mounted display (HMD). These HMDs use infrared eye cameras to track gaze, eye movement, and pupillometry. The integrated eye gaze tracking and pupillometry system samples at 200 Hz and is able to pick up subtle shifts in attention and concentration. Neurostimulation Techniques Transcutaneous auricular neurostimulation (tAN) is a paradigm which targets both the trigeminal and vagus nerves on the auricle. Recently, the auricle has become a target for neuromodulation in a wide variety of pathologies, including opioid withdrawal, chronic abdominal pain, and drug-resistant epilepsy (Kaniusas et al., 2019). tAN delivers mild electrical stimulation to modulate several cranial nerve branches in key dermatome regions through a small non-invasive device placed over the left ear (auricle). These regions cover or are adjacent to the afferent sensory innervation of several cranial nerves (V, VII, IX, X) and occipital nerves. As a result, tAN stimulation may have an impact on maintaining attention levels (Sharon et al., 2020), reducing acute stress (Ylikoski et al., 2020), and alleviating nausea (Molefi et al., 2023; Espinoza-Palavicino et al., 2023; Eren et al., 2018). tAN may also attenuate symptoms of motion sickness (MS) and spatial deviation (SD) (Molefi et al., 2023; Espinoza-Palavicino et al., 2023). Vagal and trigeminal afferents project to the nucleus tractus solitarius (NTS), which is a major relay station within the brain. The NTS contains trigemino-vestibulo-vagal neurocircuitry and plays an important role in MS (Babic \& Browning, 2014). It also receives direct or indirect input from the spinal tract, area postrema, hypothalamus, the cerebellum and vestibular / labyrinthine systems as well as the cerebral cortex, all of which play important roles in the regulation of medullary reflexes controlling nausea and vomiting (Babic \& Browning, 2014). tAN also modulates the locus coeruleus, a major brain nucleus for norepinephrine production, which has been hypothesized to mitigate symptoms of MS (Badran et al., 2018). These connections suggest a pathway by which tAN could modulate MS, reducing the impact of cybersickness symptoms associated with XR simulations.