Study of Implicit Preferences in Facial Expression Recognition of Urban Internet-Addicted Left-Behind Children in China

Emotional image recognition of facial expression is an important and widely used measure and method for study of emotions. Standardized emotional image system carries out matching of emotional valence and emotional arousal towards emotional stimulation materials, increases possibility of equal positive and negative intensity from emotion elicitation and makes experimental results possess higher comparability and stability. After emotion elicitation, signal detection theory is a general method to detect recognition condition of participants against experimental materials. Variant GO/NO-GO (Go/ No-go association test, GNAT) test of implicit association test (IAT) that is used to measure internal mental process of individuals against a kind of thing or phenomenon tests participants’ hit rate and false alert rate under different unified tasks by adopting principles of signal detection theory. Response of participants to stimulus that represents target and attribute categories is named as hit rate (i.e. Go), and no response to other represented stimuli is referred to as false alert rate (i.e. No-go). The hit rate and false alert rate are converted to z-score and the difference value is taken as d’-score; then, through comparison of d’-score under different combinations of target concept and attribute concept, perceptibility, propensity and preference of participants are reflected [1].