ISSN 0439-755X
CN 11-1911/B

Acta Psychologica Sinica ›› 2018, Vol. 50 ›› Issue (7): 739-749.doi: 10.3724/SP.J.1041.2018.00739

• Reports of Empirical Studies • Previous Articles     Next Articles

Effects of prediction error on post-retrieval extinction of fear to compound stimuli

Wei CHEN,Junjiao LI,Jingwen CAOYANG,Yong YANG,Yanjian Hu,Xifu ZHENG()   

  1. School of Psychology, South China Normal University, Center for Studies of Psychological Application, South China Normal University, Guangdong Key Laboratory of Mental Health and Cognitive Science, Guangzhou 510631, China
  • Received:2017-08-14 Published:2018-07-15 Online:2018-05-29


Memories with a strong fear emotional connotation play a pathogenic role in a variety of emotional disorders, including post-traumatic stress disorder (PTSD), anxiety disorder and phobia. So how to fundamentally eliminate the acquired fear memory and prevent relapse has been the difficulties of clinical treatment. The retrieval extinction paradigm, which is based on the memory reconsolidation theory, was demonstrated an effective and promising way in eliminating undesired memories. The key of this paradigm is how to retrieve fear memory to undergo reconsolidation, which makes the memory fragile and labile. Research has shown that when a mismatch between what is expected based on previous experiences and the actual state of events at retrieval, the prediction error will occur so that the memory will undergo reconsolidation. In the present study, we change CS-US (conditioned stimulus-unconditioned stimulus) matching rules during the retrieval to manipulate the conditions of PE to explore whether the amount of PE is a crucial factor to open reconsolidation window.

In the current study, two tones and four colored figures were used, one of the tones and two of the figures (CS+) were paired with a mild shock to the wrist (US) on 50% of the trails, while the other figures and tones were never paired with shock (CS-). the skin conductance response (SCR) was the measure of fear. Four groups of participants were fear conditioned on day1 using a 50% reinforcement schedule, in such a way that they could expect the CS to be followed by shock every other trial. During memory reactivation on day2, participants received one unreinforced CS+ (No PE group), two unreinforced CS+ (Negative PE group), two reinforced CS+ (Positive PE group) and four unreinforced CS+ (Multiple PE group) respectively, following extinction training. On day3, participants took part in tests of spontaneous recovery and reinstatement of fear through re-extinction and regaining.

The results showed that the SCR was not significantly different among the four groups in the fear conditioning of the first day and extinction of the second day. However, on the third day, No PE group and Multiple PE group showed increased SCR in spontaneous fear recovery and reinstatement test, whereas Negative PE group and Positive PE group did not show any significant increased SCR in spontaneous fear recovery or reinstatement.

Our study provided further evidence that the behavioral interference during reconsolidation (retrieval- extinction) can effectively eliminate fear and block fear relapse under certain circumstances. According to the results, we demonstrate that the prediction error is a necessary condition of initiating reconsolidation and a lack of prediction error during retrieval will leave the memory trace in an inactive state. Additionally, the amount of PE is a crucial factor and too much prediction error will cause failure. It is inferred that a limited degree of mismatch between the memory and events at the time of retrieval will induce memory destabilization.

Key words: prediction error, fear extinction, retrieval-extinction, compound stimuli

CLC Number: