Emotional state objective measurement method based on cardiac interval

Abstract

The invention discloses an emotional state objective measurement method based on a cardiac interval, and the method comprises the steps: simulating the activities of sympathetic nerves and parasympathetic nerves according to the collected cardiac pulse interval, and extracting the activity indexes of the sympathetic nerves and the activity indexes of the parasympathetic nerves through a regression coefficient, thereby eliminating the influence of noise and extreme values, and achieving the objective measurement of the emotional state. And forming a two-dimensional emotion label through an emotion awakening condition and an emotion titer condition provided by the participant, and further establishing an emotion prediction model by combining a comprehensive neural activity index, thereby obtaining a corresponding emotion state according to a subsequently output cardiac pulse interval.

Description

technical field [0001] The invention relates to emotional measurement in applied psychology, in particular to an objective measurement method of emotional state based on heart interval. Background technique [0002] In applied psychology, research on emotion measurement has always been a relatively important project. The existing related technology is the emotion detection technology based on Heart Rate Variability (HRV). This technology uses ECG sensor or pulse sensor to obtain heart fluctuation interval (R-R interval) and utilizes the variability of R-R interval length (such as R-R interval variance, root mean square, etc.) over a period of time (usually 60 seconds-1 hour). , high-middle-low frequency after Fourier transform) indicators to fit the individual's emotional state. The emotional state it measures is generally "psychological tension" or "strength of frustration". Its essence is the measurement of human sympathetic-parasympathetic activity. However, the validi...

AI Technical Summary

Problems solved by technology

[0004] This type of technology has the following technical limitations, one of which is: HRV parameters are used as machine learning features, but the unreliability of HRV is an objective problem, although technicians have adopted a large number of technologies (such as multi-model combination, advanced feature engineering ), but its model always has large fluctuations, so it is impossible to obtain a particularly accurate emotional estimate. Whether it is 2 or 4 emotional categories, each category spans a large range. According to common sense, human The difference in emotional state can be really nuanced, far from being summarized in 2-4 intervals; the second is that the time granularity of the data obtained by HRV technology is very large, as mentioned above, the fastest simultaneous measurement also takes about 60 seconds

[0005] In addition, the data measured by the existing technology is a single dimension, that is, the degree of tension, and the degree of "frustrated" (frustrated) called in some studies, which are collectively referred to as the degree of emotional arousal in the field of psychology (that is, from calm to agitated). Dimensions of change between emotions), but emotion is not only this dimension, emotional valence (that is, the nature of the positive-negative dimension of emotion) is also a very important discriminant indicator, losing the positive and negative judgment of emotion, emotion discrimination is not complete

Images