An Investigation of Analytic Decision During Driving Test

To examine the long-term causality between Cardiorespiratory Electromyography Galvanic signals for 17 drivers taken from Stress Recognition in Automobile Drivers database. Methods: Two statistical methods, co-integration to reveal an eventual existence of a long-term relationship between ECG (Electrocardiograph), EMG (electromyography), GSR (galvanic skin resistance), heart rate (HR) and respiration, well as the Application of the model of Granger causality. Results: ECG shows certain dependence to EMG, GSR, heart rate and respiration. The results for ECG dependent suggest that an increase of 1% in EMG, FOOTGSR, HAND GSR, HR and RESPIRATION implies a variation of ECG which take a value respectively of 0.016248%, 0.007241%, 0.028366%, 0.000511% and 0.000110% in the within dimension based on the FMOLS (Fully Modified Ordinary Least Squares). With same way, the result for ECG suggest that an increase of 1% in EMG, FOOT GSR, HAND GSR, HR and RESPIRATION implies a variation of ECG which take a value respectively, of 0.065684%, 0.014534%, 0.032800%, 0.000304%, 0.005986% in the between dimension based on the same method. The results of panel Granger causality show a bi-directional relationship between ECG and FOOT GSR, HAND GSR and respiration signals, it must be noted as a unidirectional causality from EMG to ECG. Conclusion: This study shows the long-term interaction between the bio signals, and reveal how the understanding of these interactions can help the doctors to understand the risks that may exist between these interactions. The main advantage of a multidimensional and multivariate model is to solve a multitude of problems that prevent doctors to treat the patients better and is not the case for studies in two dimensions. Keywords—Panel Co-integration; Panel Granger Causality; FMOLS and DOLS Estimators; Cardiorespiratory electromyography galvanic signals