Generalization Effect of Conditioned Sadness: An Innovation of the Approach to Elicit Implicit Sadness

Event-related potentials (ERPs) were recorded to explore electrophysiological correlates of conditioned sadness and happiness generalization processing in a classical conditioning paradigm when subjects performed a simple number estimation task during post-conditioning phase. Three emotion conditions (happy, neutral, and sad) were examined, each of which contained two phases: a conditioning phase and a post-conditioning phase. During conditioning phase of each emotion condition, we presented 56 three-digit-numbers (i.e. conditioned numbers) which were paired respectively with 56 facial expressions containing one emotion one by one; participants were instructed to attempt, to the best of their ability, to study and remember the number while ignoring the facial expression as much as possible. During post-conditioning phase of each emotion condition, we presented three kinds of three-digit-numbers [i.e. 56 conditioned numbers, 56 generalized numbers, and 56 control numbers] one by one; participants were instructed to judge whether the figure had been presented during conditioning phase or not. Our results showed that (1) the three-digit-numbers [i.e. Sad Generalized Numbers] which might contain the generalized conditioned sadness elicited a more negative ERP component (P300–600) than did the three-digit-numbers [i.e. Neutral Generalized Numbers] which might be neutral (P < 0.05); and (2) the three-digit-numbers [i.e. Happy Control Numbers] which might contain the biased attention toward novel stimuli induced by positive stimuli elicited a more negative ERP component (P450–750) than did the three-digit-numbers [i.e. Neutral Control Numbers] which might be neutral (P < 0.05). Thus, it is suggested that P300–600 may constitute the brain indices of generalized conditioned sadness. We discuss whether the generalized conditioned sadness elicited by Sad Generalized Numbers could be implicit. Whether P450–750 is related to the neural correlates of the biased attention toward novel stimuli elicited by positive stimuli is also discussed.