| Self-reference effect(SRE)refers to that information encoded in relation to the self has a mnemonic advantage over information encoded in other ways,which is closely related to individuals’ social functions.Difficulties with social communication and interaction are key characteristics of autism spectrum disorder(ASD).Therefore,it is of great significance to study the SRE of ASD to understand their self-concept and explore the related causes of their social difficulties.Previous studies have adopted different paradigms to investigate the SRE of autistic individuals,but different paradigms have obtained inconsistent results.In view of this result,some researchers have proposed that different paradigms involve different levels of selfrepresentation,and put forward "different levels of self-representation",that is,the SRE of autistic individuals at the first-order self-representation level is intact.The SRE at the level of second-order self-representation is defective.But at present,this hypothesis has not been verified by experimental studies.Therefore,this study will further systematically investigate the SRE of autistic children to verify this view.Three groups of participants were included in the current experiment: Children with ASD,typical developmental(TD)children,and children with intellectual disorder(ID).The ASD group had a verbal and nonverbal IQ similar to that of the ID group,whereas both the ASD and ID groups’ IQs were lower than the TD groups’ IQ.Three groups were matched in terms of chronological ages.In Experiment 1,we aimed to investigate whether there are defects in the SRE of children with ASD at the level of first-order self-representation in the shape-label matching task.Experiment 1adopted 3(Groups: ASD/ID/TD)× 2(Person: you/other)× 2(Trail type:match/mismatch)three-factor mixed experimental design.Participants were first instructed to associate a shape(triangle or circle)to a label,which could either relate to themselves or an unfamiliar other(you or other),e.g.,you-circle and other-triangle were labeled matches,any alternative combinations were labeled mismatches.After the association phase,they were asked to judge if the presented shape and label was a match or a mismatch.The primary dependent measures for this task were the accuracy and reaction time(RT)of their judgments.We calculated a SRE score for matching judgments by subtracting the accuracy for other from the accuracy for self.We also calculated a SRE score for RT in the same manner.The larger the difference value,the greater the SRE.In Experiment 2,we aimed to explore whether there are defects in the ownership effect of children with ASD at the level of second-order selfrepresentation in the ownership task.Experiment 2 adopted 3(Groups: ASD/ID/TD)× 2(Ownership: self/other)two factors mixed experimental design.In the study phase,participants were presented with picture cards stacked in a random order.Half of the pictures had a red border and half had a blue border.Participants were also presented with a blue shopping basket and a red shopping basket.Participants were instructed to place cards with a red border into the red basket and cards with a blue border into the blue basket.In the test phase,participants were presented with each picture individually and asked to recognize whether or not it had appeared in the study phase.Half of the pictures were occurred in the study phase and half were new pictures which hadn‘t appeared in the study phase.If participant’s memory of pictures of their own items is better than that of pictures of other’s items,it indicates that they exhibit a SRE.The primary dependent measures for this task was the Correct Hit Rate(discrimination sensitivity,d’).We calculated a SRE score by subtracting the Correct Hit Rate about other from the Correct Hit Rate about self.The larger the resulting value,the greater the SRE.After the experiments,the head teacher of the children were asked to fill in the Autism Spectrum Quotient Questionnaire(AQ)and the Autism Rating Scale for Children(CARS)to investigate the relationship between SRE and autism symptoms.The experimental results are as follows:(1)In the shape-label matching task,on matching judgments,for three groups,accuracy for self was higher than that for other.RT for self was faster than RT for other.This suggests that participants in three groups showed a SRE.(2)In the ownership task,for ASD group,there was no significant difference for memory performance(d’)in “self-owned” and that in the “other-owned” items.But the d’ of the TD and ID groups were significantly higher for “self-owned” items than that for the “other-owned” items.This shows that the SRE was absent for the ASD group.For TD and ID groups,a clear SRE was present.(3)For ASD and ID groups,the SRE of the shape-label matching task was significantly greater than that of the ownership task.For TD group,the SRE in the two tasks was not significantly different.There was no correlation between the SRE of the three groups of participants on the two experimental tasks.This indicates that in different tasks,the SRE may involve different levels of self-representation,and the SRE in children with ASD is weaker in the high level of self-representation.(4)For three groups,whether it was the shape-label matching task or the ownership task,the SRE was not related to the autistic traits and symptoms.These results suggests that for ASD group,the SRE is partially intact.The SRE in the first-order self-representation level may be intact,but the SRE in the second-order self-representation level may be defective.This result verifies the theory of different levels of self-representation of autistic individuals,and provides more experimental evidence to support the hypothesis.It further reveals that autistic children’s self is atypical,which is of great significance to understand how children with ASD deal with the information related to themselves,form self-concept,and help teachers or parents create an environment that can make full use of their self-reference effect,and improve autistic children’s social adaptability and learning ability. |