COGNITIVE ASSESSMENT BASED ON PSYCHOPHYSIOLOGICAL PARAMETERS

Abstract

Cognition is the ability of an organism to organize and process information to form knowledge. This includes acquiring information (perception), selecting (attention), representing (understanding) and retaining (memory) information, and using it to guide behavior (reasoning and coordination of motor outputs). The scienti c discipline dedicated to study relationship between the physiological and psychological elements of behavior is called psychophysiology. When human behavior and cognition are studied in context of relating to biological basis which control and sustains behavior, such studies add enrichment to behavioral studies. Psychophysiological study generally comprises of physiological parameters, such as, ECG, EMG, EOG, HRV, PPG, respiration, galvanic skin response, electroencephalogram (EEG), MEG, FNIR, FMRI, PET, etc. In this research, we have obtained psychophysiological parameters using EEG and HRV. Considering cognitive assessment as focus of the research, the work has been divided into four parts. First study is to obtain a relationship between psychological stress and EEG and HRV metric parameters, and to design a classi er to identify psychological stress level of a subject. Second work is nding mathematical correlate between perception-attention and EEG signal band powers. Third is to study the enhancement in various attentional parameters due to long-term intervention practice of Yoga and meditation, using psychophysiological parameters as EEG. Fourth is to develop feature sets using EEG signal to identify various states of cognitive state such as rest, active sensory-motor coordination, active perceptive attention, active sustained attention, etc. All the work was conducted in the coordinated laboratory environment of IIT Roorkee. In the rst study, an EEG features based system is developed to detect and classify the mental stress level of a person. Regular personal stress pro le generated can be used as neurofeedback for the clinical and other assessment. An EEG-metric features based binary and ternary stress classi er is developed. A probabilistic stress pro ler of di erential stress inventory (a questionnaire based evaluation using VTS system) is used as an output feature during the learning, validation and testing of classi ers. B-Alert X10 EEG system is used to extract metric features in form of cognitive state and workload outputs for 41 healthy volunteers (37 males and 4 females, age; 24 5 years All subjects were guided to perform three visual tasks of eyes closed, gazing on a red dot on a dark screen and gazing on a bright screen. Central tendencies (mean, median and mode) and deviations were extracted from EEG and HRV combined metric (sleep onset, distraction, low engagement, high engagement and cognitive states) as features. Either of the two or three stress classes were formulated based on probabilistic stress pro ler of di erential stress inventory and used as training output classes. A supervisory training of radial basis function based binary support vector machine classi er was used to detect stress class one by one. 40 subjects samples were used for training and interchanging one-by one 41th subjects stress class is determined from the designed classi er. An accuracy of 73.17% was obtained in detecting the binary class of psychological stress. Focus, alertness level, and reaction time are key cognitive factors and are dependent on visual and auditory perception. In the second study, the functional changes in cortical regions; through frequency domain features of EEG, in correlation to perception and attention level (visual and auditory), using a computer screen-control task, is studied. Psychophysiological data consisting nine channel EEG for 26 subjects were recorded during three conditions; 5 minutes eyes closed (EC), 5 minutes visual focus on a red dot on dark screen (DOT), and approximately 10 minutes battery test consisting 3 visual and 3 auditory subsections (intrinsic, cross-modal phasic and unimodal phasic) through perception and attention functions: alertness (WAFA). Relative power spectrum density (PSD) of di erent activity bands for each epoch of one second have been computed; corresponding to each task. For WAFA task, mean reaction time (MRT, in milliseconds) and dispersion of reaction time (DRT, in milliseconds) have been registered. Correlation values between PSD of every EC, DOT and WAFA; and, MRT and DRT of WAFA assignment was registered. For task EC, correlation values are positive for theta, alpha, negative and positive for MRT and DRT consecutively, throughout the cortical region. For task DOT, correlation values negative for delta, theta and gamma; positive for alpha and beta. For task WAFA, correlation values are negative for delta and theta, but for di erent subsections of visual and auditory (intrinsic, cross-modal phasic and unimodal phasic) are either of positive, negative or uncorrelated for alpha, beta and gamma throughout the cortical region. EEG band activities are observed to be directionally correlated with visual and auditory alertness level. Yogic activities have gained e cacy in physical tness, cognitive enhancement, rehabilitation and therapy. In the third study, the e cacy of Yogic exercises in improving cognitive attention is investigated. A comparative psychophysiological analysis was performed between experimental group \Y "practicing Yoga for more than a year (n=29 and a control group \C"(n=35) , age group 25 5 years, to study the di erence in cortical activation during visual and auditory attention engagement and comparison of task based evaluation. Volunteers were guided to perform four consecutive tasks involving visual attention blocking or activation, with EEG recorded in synchronization. EEG band powers and cognitive index features were extracted after wavelet decomposition and Wilcoxon signed-rank test based statistical comparison of features for two groups was performed. A signi cantly higher EEG theta, alpha and beta power were observed for Y during sustained attention. During high engagement attention tasks, for the Yoga group; task load index and neural activity index is lower, whereas attention resource index, arousal index, psychophysiological index are higher. During visuo-auditory attention task; the average response time for the control group is lesser. Similarly, during vigilance task average response time for the control group is lesser than Yoga group. Slower response time of Yoga group is an indicator of slowersensory-motor responses. Yoga group had higher alpha and beta powers at frontal and parieto-occipital region during high involvement attention, and higher response time during sustained attention tasks. Overall, it was concluded that long term practice of Yoga has an enhancing impact on cognitive attention. Locating cognitive task states by measuring changes in electrocortical activity due to various attentional and sensory-motor changes, has been in research interest since last few decades. In this part of work, various cognitive state while performing various attentional and visuo-motor coordination tasks, are classi ed using electroencephalogram (EEG) signal. A non-linear time-series method, multifractal detrended uctuation analysis (MFDFA) , is applied on respective EEG signal for features. Using MFDFA based features a multinomial classi cation is achieved. Nine channel EEG signal was recorded for 38 young volunteers (age:25 5 years, 30 male and 8 female), during six consecutive tasks. First three tasks are related to increasing levels of selective focus vision; and next three are re ex and response based computer tasks. Total of 90 features (ten features from each of nine channel) were extracted from Hurst and singularity exponents of MFDFA on EEG signals. After feature selection, a multinomial classi er of six classes using two methods: support vector machine (SVM) and decision tree classi er (DTC). An accuracy of 96.84% using SVM and 92.49% using DTC was achieved. Overall, the research work is comprised of assessment and analysis of cognitive functions, viz. mental stress, visual attention and perception. The research work touched varieties of study such as relating EEG and HRV parameters with psychological stress, evaluation of individual's stress level, relating EEG frequency band parameters with visual attention and perception, analysis and comparative study of visual attention function for Yoga practitioner group using EEG parameters and task performance parameters, and classi cation of cognitive task class using MFDFA features from EEG independent of individual participant. The presented research work will nd it's application in individual's cognitive function assessment and comparative analysis during intervention studies.