Review of EEG, ERP, and Brain Connectivity Estimators as Predictive Biomarkers of Social Anxiety Disorder

Anxiety disorders impose substantial costs upon public health and productivity in the USA and worldwide. At present, these conditions are quantified by self-report questionnaires that only apply to behaviors that are accessible to consciousness, or by the timing of responses to fear- and anxiety-related words that are indirect since they do not produce fear, e.g., Dot Probe Test and emotional Stroop. We now review the conditioned responses (CRs) to fear produced by a neutral stimulus (conditioned stimulus CS+) when it cues a painful laser unconditioned stimulus (US). These CRs include autonomic (Skin Conductance Response) and ratings of the CS+ unpleasantness, ability to command attention, and the recognition of the association of CS+ with US (expectancy). These CRs are directly related to fear, and some measure behaviors that are minimally accessible to consciousness e.g., economic scales. Fear-related CRs include non-phase-locked phase changes in oscillatory EEG power defined by frequency and time post-stimulus over baseline, and changes in phase-locked visual and laser evoked responses both of which include late potentials reflecting attention or expectancy, like the P300, or contingent negative variation. Increases (ERS) and decreases (ERD) in oscillatory power post-stimulus may be generalizable given their consistency across healthy subjects. ERS and ERD are related to the ratings above as well as to anxious personalities and clinical anxiety and can resolve activity over short time intervals like those for some moods and emotions. These results could be incorporated into an objective instrumented test that measures EEG and CRs of autonomic activity and psychological ratings related to conditioned fear, some of which are subliminal. As in the case of instrumented tests of vigilance, these results could be useful for the direct, objective measurement of multiple aspects of the risk, diagnosis, and monitoring of therapies for anxiety disorders and anxious personalities.

Electroencephalography (EEG) serves as an effective diagnostic tool for mental disorders and neurological abnormalities. Enhanced analysis and classification of EEG signals can help improve performance in classifying the disorders and abnormalities. A new approach is examined here for enhancing EEG classification performance using a novel model of data representation that leverages knowledge of spatial layout of EEG sensors. An investigation of the performance of the proposed data representation model provides evidence of consistently higher classification accuracy of the proposed model compared with a model that ignores the sensor layout. The performance is assessed for models that represent the information content of the EEG signals in two different ways: a one-dimensional concatenation of the channels of the frequency bands and a proposed image-like two-dimensional representation of the EEG channel locations. The models are used in conjunction with different machine learning techniques. Performance of these models is examined on two tasks: social anxiety disorder classification, and emotion recognition using a dataset, DEAP, for emotion analysis using physiological signals. We hypothesize that the proposed two-dimensional model will significantly outperform the one-dimensional model and this is validated in our results as this model consistently yields 5-8% higher accuracy in all machine learning algorithms investigated. Among the algorithms investigated, Convolutional Neural Networks provide the best performance, far exceeding that of Support Vector Machine and k-Nearest Neighbors algorithms.

The commercial availability of many real-life smart sensors, wearables, and mobile apps provides a valuable source of information about a wide range of human behavioral, physiological, and social markers that can be used to infer the user’s mental state and mood. However, there are currently no commercial digital products that integrate these psychosocial metrics with the real-time measurement of neural activity. In particular, electroencephalography (EEG) is a well-validated and highly sensitive neuroimaging method that yields robust markers of mood and affective processing, and has been widely used in mental health research for decades. The integration of wearable neuro-sensors into existing multimodal sensor arrays could hold great promise for deep digital neurophenotyping in the detection and personalized treatment of mood disorders. In this paper, we propose a multi-domain digital neurophenotyping model based on the socioecological model of health. The proposed model presents a holistic approach to digital mental health, leveraging recent neuroscientific advances, and could deliver highly personalized diagnoses and treatments. The technological and ethical challenges of this model are discussed.