Individuals with autism spectrum conditions (ASC) have profound impairments in the interpersonal social domain, but it is unclear if individuals with ASC also have impairments in the intrapersonal self-referential domain. We aimed to evaluate across several well validated measures in both domains, whether both self-referential cognition and empathy are impaired in ASC and whether these two domains are related to each other.
Thirty adults aged 19-45, with Asperger Syndrome or high-functioning autism and 30 age, sex, and IQ matched controls participated in the self-reference effect (SRE) paradigm. In the SRE paradigm, participants judged adjectives in relation to the self, a similar close other, a dissimilar non-close other, or for linguistic content. Recognition memory was later tested. After the SRE paradigm, several other complimentary self-referential cognitive measures were taken. Alexithymia and private self-consciousness were measured via self-report. Self-focused attention was measured on the Self-Focus Sentence Completion task. Empathy was measured with 3 self-report instruments and 1 performance measure of mentalizing (Eyes test). Self-reported autistic traits were also measured with the Autism Spectrum Quotient (AQ). Although individuals with ASC showed a significant SRE in memory, this bias was decreased compared to controls. Individuals with ASC also showed reduced memory for the self and a similar close other and also had concurrent impairments on measures of alexithymia, self-focused attention, and on all 4 empathy measures. Individual differences in self-referential cognition predicted mentalizing ability and self-reported autistic traits. More alexithymia and less self memory was predictive of larger mentalizing impairments and AQ scores regardless of diagnosis. In ASC, more self-focused attention is associated with better mentalizing ability and lower AQ scores, while in controls, more self-focused attention is associated with decreased mentalizing ability and higher AQ scores. Increasing private self-consciousness also predicted better mentalizing ability, but only for individuals with ASC.
We conclude that individuals with ASC have broad impairments in both self-referential cognition and empathy. These two domains are also intrinsically linked and support predictions made by simulation theory. Our results also highlight a specific dysfunction in ASC within cortical midlines structures of the brain such as the medial prefrontal cortex.
Citation: Lombardo MV, Barnes JL, Wheelwright SJ, Baron-Cohen S (2007) Self-Referential Cognition and Empathy in Autism. PLoS ONE2(9): e883. https://doi.org/10.1371/journal.pone.0000883
Academic Editor: Paul Zak, Claremont Graduate University, United States of America
Received: August 10, 2007; Accepted: August 17, 2007; Published: September 12, 2007
Copyright: © 2007 Lombardo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: MVL was supported by the Shirley Foundation and the Cambridge Overseas Trust. JLB was supported by the Fulbright Foundation. SJW and SBC were supported by the Medical Research Council (MRC) during the period of this work.
Competing interests: The authors have declared that no competing interests exist.
Autism spectrum conditions (ASC) involve impairments in social functioning, language or communication, and the presence of stereotyped repetitive behaviors and/or highly restricted interests. Historically, the “self” has been integral in what it means to have autism. The term “autism” (derived from the Greek word “autos”, which literally means “self”) was first coined by Bleuler to characterize many of the social withdrawal symptoms exhibited by schizophrenics. Later, Kanner applied this term to the children in his clinic whom he observed to be completely self-focused. Most recently, Frith has developed a theoretical perspective that puts the self at the core of all the observed impairments and strengths in ASC. Frith's theory differs from Kanner in that, instead of viewing ASC as a syndrome of complete self-focus, it is viewed under the notion of an “absent self”.
The first reason for proposing an absent self comes from several observations that detail abnormalities in ASC due to weak central coherence, executive dysfunction, and varying degrees of mindblindness. One commonality among all of these theories is a lack of what psychologists have historically called the “central executive”; that is, a lack of top-down control of bottom-up information processing in the brain. Recent neuroimaging studies have confirmed that all of these abnormalities are due to deficient top down modulation of bottom up information processing by frontal regions of the brain, , , , ,  and are concurrent with the notion of enhanced short range but diminished long range connectivity in the autistic brain , , , , , .
At the psychological level of analysis, the absent self theory also proposes that self-awareness is less developed in ASC. When individuals with ASC are asked to report on the content of randomly sampled daily experiences, their reports relied on physical descriptions of the moment rather than on their own mental and emotional states. During structured interviews that elicit statements about various aspects of the self, adolescents with ASC gave fewer descriptions of themselves in social contexts. Adults with ASC may also not show the typical facilitative effect of self-referential cognition on memory (the “self-reference effect in memory”). Individuals with ASC also report more difficulty with identifying and describing their own emotions; what is known clinically as “alexithymia”. Atypical first person pronoun usage has also been well documented in ASC, , , , . While older theories have related this abnormality to a form of echolalia or to abnormalities in person deixis, normative studies in social psychology regard first person pronoun usage as an index of self-focused attention. Thus, the available work implicates that self-referential cognition may be abnormal in ASC.
In the present study we aimed to further examine whether self-referential cognitive processing is abnormal in ASC and to examine the interaction of self-referential cognition and empathizing. First, we tested whether individuals with ASC would show the typical self-reference effect in memory (SRE). Typically developing adults show a robust SRE when self-referential information processing is compared to information processing in relation to other people or to semantic manipulations. The SRE is also positively related to trait self-consciousness and works through enhancing organizational and elaborative cognitive processing of information, . To compliment the SRE paradigm, we included several other well validated self-report and performance measures of self-referential cognition. Furthermore, since both cognitive and affective empathy impairments have been well documented in ASC, , , , several self-report and performance measures of empathy were included. We predicted that across all empathy and self-referential cognitive measures, individuals with ASC would show concurrent impairments.
Within the typically developing brain, both self-referential cognitive processing and mentalizing are consistently localized to neural processing in cortical midline structures (CMS) of the brain such as the medial prefrontal cortex (MPFC) and posterior cingulate/precuneus (PC), , , , ,  (see Figure 1). However, mentalizing does not selectively recruit CMS regions of the brain in ASC, , , , . This observation has led some to theorize that CMS regions of the brain play a role in using simulation as a mentalizing strategy, . Simulation theory suggests that one way we know about the psychological characteristics of others is by using information about ourselves to construct a default model of how other minds work. To the extent that someone is similar to oneself, one can then evaluate how much of one's own characteristics should be imputed onto others, , . Consistent with predictions made by simulation theory, we predicted that individuals with ASC would show decreased memory for themselves and for similar close others, but show no differences in memory for a dissimilar non-close other or for non-social information. Furthermore, to corroborate the link between self-referential cognition and mentalizing we also predicted that there would be significant individual differences in how self-referential cognitive processing is related to mentalizing in both ASC and normally developing controls.
All participants gave informed consent to participate in this study in accordance with the University of Cambridge Psychology Research Ethics Committee. There were 23 males and 7 females aged 19–45 that had an official diagnosis by internationally accepted criteria,  of either high-functioning autism (HFA; n = 4) or Asperger Syndrome (AS; n = 26). Controls consisted of 23 males and 7 females who were pairwise matched on age and gender and had no known psychiatric, developmental, or neurological disorders. All participants completed the Autism Spectrum Quotient (AQ) and the ASC group scored higher than controls on the AQ (p<0.001). All participants scored in the normal range on the Wechsler Abbreviated Scale of Intelligence (WASI) and there were no statistically significant group differences on verbal, performance, or full scale IQ (all p>0.90). See Table 1 for participant characteristics.
Procedure and Measures
The paradigm used to assess the SRE was a depth-of-processing paradigm. During the encoding phase, participants judged trait adjectives in one of four ways. In the “self” condition, participants judged how descriptive the adjective was of themselves. In the “similar close other” condition, adjectives were judged on how descriptive it was of their best friend. In the “dissimilar non-close other” condition participants judged whether the adjective was descriptive of Harry Potter. All of these judgments were made on a 6 point scale where 1 indicated “not at all descriptive” and 6 indicated “very descriptive”. Finally, in the non-social control condition, participants judged how many syllables each adjective contained (from 1–6). Each condition had 30 trials and all were presented in pseudorandom order. All adjectives were drawn from a previously validated and widely used set of trait adjectives. Half the adjectives in each condition were positively valenced (e.g., inventive) while the other half were negatively valenced (e.g., messy). Among all conditions there were no differences in number of characters, syllables, valence, or frequency of the adjectives.
After encoding there was a 30 minute delay before the retrieval phase. Participants were completely unaware of the subsequent recognition memory task to follow. During this delay participants completed the performance section of the WASI. These tasks were non-verbal and were administered to keep the participant occupied during the delay period.
After the delay, participants were given a surprise recognition memory test. All 120 adjectives from encoding and 120 new distracter adjectives were presented in pseudorandom order. Participants judged their confidence in whether the adjective was “old” or “new”. Confidence judgments were made on a 1–6 scale where 6 was “definitely OLD” and 4 was “OLD, but kind of unsure”. Conversely, a 1 indicated that they were “definitely NEW” and a 3 was “NEW but kind of unsure”. This 6 point scale was used to force participants to make finer grained recognition judgments and also to investigate whether there were any differences in how each group used different confidence judgments. There were no group differences among judgments on each scale point within any of the conditions (all p>0.10). Therefore, we collapsed judgments 1–3 into “New” and 4–6 into “Old” judgments.
Prior to testing, to ensure all participants were familiar with Harry Potter we set an inclusion criterion to have seen at least one Harry Potter film or to have read at least one book. We obtained a familiarity index for Harry Potter by adding up the number of movies seen and books read. Groups did not differ on this Harry Potter familiarity index (p = 0.334).
After the SRE paradigm participants were asked how similar and close they thought they were to their friend and to Harry Potter and how likeable they perceived their friend or Harry Potter to be. Similarity and likable judgments were rated on a 1–6 scale, where 1 was “not at all” and 6 was “very much”. For the closeness measure, we used the Inclusion of Other in Self scale (IOS). The IOS depicts closeness spatially with Venn diagrams of two circles. One circle is labeled “Self” and the other circle was either labeled “Friend” or “Harry Potter”. Participants are given 7 choices that vary in degree of overlap between the “Self” circle and the other circle. Larger numbers indicate judgments of feeling close and including another person within the self.
Confirming our similarity and closeness manipulations, within each group participants viewed themselves and their best friend to be closer and more similar than the Self-Harry Potter judgment. Both groups also liked their friend more than Harry Potter (all p<0.009). However, there were no differences between groups for either best friend or Harry Potter on similarity, closeness, or liking (all p>0.12). See Table 1.
The Empathy Quotient (EQ), Interpersonal Reactivity Index (IRI), and the Emotional Contagion Scale (ECS) were our three self-report measures of empathy. The EQ was rescored into cognitive and affective empathy based on items identified from previous factor analyses. On the IRI, our indices of cognitive and affective empathy were the perspective taking and empathic concern subscales respectively. The IRI also has two other subscales (Fantasy and Personal Distress), but these were not included as measures of cognitive or affective empathy. Finally, our performance measure of empathy, indexing mentalizing ability, was the Reading the Mind in the Eyes Test (Eyes).
Self-reported self-consciousness/awareness was measured with the Private Self-Consciousness Scale (PSCS) and the Toronto Alexithymia Scale (TAS). We also included a performance measure of self-focused attention called the Self-Focus Sentence Completion task. In this task, participants were given sentence stems that included a self-reference (e.g., I think…, If I had my way…). For each stem, we asked participants to complete the sentence in whatever way they liked. Self-focused attention (SFA) was the dependent variable and was computed automatically with the Linguistic Inquiry and Word Count program as the percentage of first person pronouns used to complete the sentences (e.g., I, me, myself). This index has been used several times as a valid quantitative index of SFA, , .
To evaluate our main hypotheses within the SRE paradigm, we ran a repeated measures ANOVA with four within subject levels and group as the between subject variable. Our dependent variable was a standard measure of memory sensitivity (d') formulated as the standardized score of correctly remembered words minus the standardized score of false alarms. Because we hypothesized that groups would differ on Eyes test performance and because it was theoretically relevant to test the effects of the SRE paradigm independent of mentalizing ability, we ran another repeated measures ANCOVA with Eyes test included as a covariate. Second, we computed two different SRE variables. The first SRE was a difference score for self versus friend (d' Self-d' Friend) while the second SRE was the difference score for self versus Harry Potter (d' Self-d' Harry Potter). Larger scores on these SRE variables indicate larger biases for self-referentially encoded information compared to other-referentially encoded information. One sample t-tests in each group were ran to evaluate whether each SRE variable was significantly different from 0. Independent samples t-tests were also ran to determine whether groups differed on the SRE variables. Third, we examined whether groups differed on empathy and self-consciousness measures using independent samples t-tests. Measures of effect size (Cohen's d) were also computed for each t-test comparison.
To examine individual differences between self-referential cognition and mentalizing we ran multiple regression analyses with Eyes test as the dependent variable and measures of self-referential cognition (self memory, TAS, PSCS, SFA) as independent variables. In our regression models we determined statistically whether relationships between mentalizing and self-referential cognition differed as a function of diagnosis using procedures outlined by Aiken and West. In all, four regression models were evaluated (one model for each self-referential independent variable) and each model had the main effects of the self independent variable and a dummy variable for group entered on the first step. In the second step of our regression model, we entered the product vector of the interaction between the self independent variable and group membership (Group x Self Independent Variable). If the product vector interaction term was significant, this meant that the relationship between self-referential cognition and mentalizing was statistically different between groups (e.g., slopes were not parallel between groups). If this occurred, we conducted a simple linear regression, but this time separately on each group to discern how the relationships varied by diagnosis. However, if the product vector was not significant, this meant that there was no difference in the relationship between self-referential cognition and mentalizing as a function of diagnosis (e.g., slopes were parallel between groups). Therefore, in this case we ran a simple linear regression model but this time with the data collapsed across groups. The same procedures were used to examine the relationship between self-referential cognition and the AQ. In these analyses however, the AQ was the dependent variable.
Our first analysis looked at performance on the SRE paradigm. The repeated measures ANOVA showed an interaction effect between encoding level and group (F(3,58) = 3.599, p = 0.015). Therefore, we looked at the simple effects of encoding level within and between groups to discern the basis of the interaction. Consistent with our predictions, between group comparisons at each level of encoding showed that individuals with ASC were not impaired on memory for words previously judged on the basis of counting syllables (t(58) = 0.105, p = 0.916, Cohen's d = 0.03) or when judging Harry Potter's personality (t(58) = 0.716, p = 0.450, Cohen's d = 0.19). However, individuals with ASC had decreased memory for words previously judged in relation to themselves (t(58) = 2.265, p = 0.027, Cohen's d = 0.59) or their best friend (t(58) = 2.814, p = 0.007, Cohen's d = 0.73). See Table 2 and Figure 2.
Within each group there was a linear trend replicating the effect of depth of processing within the SRE paradigm. The hierarchy of memory performance was Self>Friend>Potter>Syllable. Post-hoc Bonferroni corrected comparisons confirmed that all conditions for both groups were different from each other (all p<0.02). Furthermore, within each group, SREs were greater than zero for Self-Friend (Control: t(29) = 3.175, p = 0.004; ASC: t(29) = 4.068, p = 0.0003) and for Self-Harry Potter (Control: t(29) = 7.768, p<0.001; ASC: t(29) = 6.617, p<0.001). However, between groups, the SRE for Self-Harry Potter was larger among controls and approached statistical significance (t(58) = 1.858, p = 0.068, Cohen's d = 0.49) while the SRE for Self-Friend was not different between groups (t(58) = −0.064, p = 0.949, Cohen's d = 0.02). See Table 2 and Figure 2.
When Eyes test was included as a covariate, the interaction between group and encoding level approached statistical significance (F(3,57) = 2.471, p = 0.064). Testing between group differences, we found only memory performance for the best friend condition was different, with controls having better memory than ASC (t(57) = 2.001, p = 0.05, Cohen's d = 0.53). Within each group the hierarchy in memory performance still existed (Self>Friend>Other>Syllable; all p<0.02). The group difference for the SRE of Self-Harry Potter approached statistical significance, with controls having a large bias that ASC (t(57) = 1.710, p = 0.093, Cohen's d = 0.45) while no differences remained for the SRE of Self-Friend (t(57) = −0.015, p = 0.988, Cohen's d = 0.003).
In our next analysis we examined whether there were concurrent impairments in self-referential cognition and empathy in ASC. Individuals with ASC scored lower across all empathy measures (Cognitive EQ, Affective EQ, IRI Perspective Taking, IRI Empathic Concern, ECS, Eyes test; all p<0.05). On the IRI Fantasy subscale, individuals with ASC were also lower than controls (t(58) = 2.507, p = 0.015, Cohen's d = 0.66). However, on the IRI Personal Distress subscale, individuals with ASC scored higher (t(58) = −3.195, p = 0.002, Cohen's d = 0.84). Individuals with ASC showed no differences on the PSCS (t(58) = 0.561, p = 0.577, Cohen's d = 0.15), but did report more alexithymia on the TAS (t(58) = −5.315, p<0.001, Cohen's d = 1.40). Among the TAS subscales, individuals with ASC had more difficulty identifying and describing their own feelings (DIF, DDF) and also reported more externally oriented thinking (EOT) (all p<0.001). Individuals with ASC were also lower in SFA on the Self-Focus Sentence Completion test (t(58) = 2.136, p = 0.037, Cohen's d = 0.56). See Table 3.
Next, we examined whether individual differences in self-referential cognition were related to mentalizing ability. No differences existed between groups in the relationship between Eyes test and self memory performance or TAS (indicated by a non-significant Group x Self interaction term in the regression model). Thus, since slopes were similar across groups, we collapsed the data across groups and looked at the effect of self-referential cognition on mentalizing. We found that regardless of diagnosis, self memory performance was positively related to Eyes test performance (r = 0.34, p = 0.007). Alexithymia was also negatively related to Eyes test performance (r = −0.43 p<0.001). Private self-consciousness and self focused attention however, showed differential relationships with the Eyes test that were dependent upon diagnosis. There was a significant interaction effect between how PSCS and diagnosis predicted Eyes performance (F change(1,56) = 4.917, p = 0.031). Inspecting the relationship PSCS had with the Eyes test separately within each group revealed that the interaction effect was driven by a strong positive correlation between PSCS and Eyes test performance in individuals with ASC (r = 0.50, p = 0.005) but not controls (r = −0.01, p = 0.951). Regarding the relationship that SFA had with Eyes test performance, we found another significant interaction effect between SFA and diagnosis (F change (1,56) = 6.627, p = 0.013). Upon looking at the relationship between SFA and Eyes test performance separately for each group, we found the interaction effect was driven by a negative relationship between SFA and Eyes test performance in the control group while individuals with ASC showed a positive relationship between SFA and the Eyes test. Both of these correlations approached statistical significance (Controls: r = −0.32, p = 0.08; ASC: r = 0.35, p = 0.06). See Figures 3–6.
The index of self-focused attention is the percentage of first person pronouns used on the Self-Focus Sentence Completion test.
The index of self-focused attention is a reflective form of self-focused attention as measured by the Private Self-Consciousness Scale.
In similar analyses, we looked at how self-referential cognition predicts scores on the AQ. There was no interaction effect among group and self-memory, TAS, or PSCS. Therefore, after collapsing across groups we found that increasing TAS scores predicted more autistic traits (r = 0.63, p<0.001) and that increasing self memory predicted less autistic traits (r = −0.32 p = 0.01). There was no relationship between PSCS and AQ scores. When modeling the relationship between SFA and AQ scores, there was a significant interaction effect between SFA and diagnosis (F change (1,56) = 12.38, p<0.001). Thus, we looked at the relationship of SFA on AQ scores separately for each group. Among controls, higher SFA predicts higher AQ scores (r = 0.50, p = 0.005). The opposite is true for individuals with ASC; higher SFA predicts lower AQ scores (r = −0.37, p<0.05). See Figures 3–6.
This study confirmed through multiple measures that individuals with ASC have difficulties in the self-referential cognitive domain. In the SRE paradigm, individuals with ASC show the typical SRE but to a reduced extent compared to age and IQ matched controls. People with ASC also showed reduced memory for themselves and for a similar close other but did not show any memory impairments for a dissimilar non-close other or for non-social memory. Thus, the only group differences in memory occurred for conditions where self information was to be remembered, or for information about others who significantly overlap with the self; similar close others.
Individuals with ASC are not however, completely impaired in self-referential information processing. We did not replicate the findings from the Toichi et al., study. Unlike the Toichi study, which claimed that there was no SRE in ASC, we did find significant SREs in the ASC group. The differences in our study and the Toichi study are worth mentioning because they shed light on the differing findings. First, the Toichi study used a biased measure of memory performance in the percentage of words that were correctly recognized. We used a more unbiased measure of memory sensitivity (d') that corrects for the false alarm rate. Second, unlike the Toichi study, we included comparison conditions for thinking about other people. When self memory is compared to memory for others, both the control and ASC groups showed the typical SRE. SREs for self versus a similar close other are identical among the two groups. Given that there are SRE effects within the ASC group, it cannot be the case that people with ASC are not facilitated by self information.
Similar and close others have been hypothesized to overlap with the self. Several investigators have found effects supporting this hypothesized overlap and that the neural substrate involved in this processing may be MPFC, , , , , . Thus, our comparison of an SRE between self and best friend may not be the best measure of a pure SRE. In fact, the self and best friend overlap quite a bit within our manipulation checks of closeness and similarity. Our best comparison of whether individuals really do exhibit a bias for self information is the SRE for self versus a dissimilar non-close other. When this comparison was investigated, individuals with ASC did show a trend for having smaller SREs than controls. This result brings up the idea that the impairment in using self information in ASC is one of degree and not complete absence. Further work at the neural level will be able to tell us more about the degree of impairment and the extent to which self- and other-referential cognitive processing overlap in the brains of people with ASC.
Our findings on several other measures of self-referential cognition also highlight its importance in ASC. Individuals with ASC report more alexithymia and are less self-focused but do not differ in private self-consciousness. Our findings extend on past work,  and show that deficits in self-referential cognition are not simply because of differences in the amount of self-reflection they engage in (as indexed by no group differences in PSCS and intact SREs), but has more to do with the ease of knowing about one's own inner emotional life. We mirror the conclusions of Berthoz and Hill, in that individuals with ASC can and do monitor their own inner states to some degree. However, it remains unclear what is the direction of the relationship between their known difficulties in understanding their own emotions and their attentional biases and future research should be directed at parsing apart this question.
We also showed in several ways that self-referential cognition and empathy are inextricably linked. First, our results within the SRE paradigm showed that mentalizing accounted specifically for self memory (indicated by the lack of a group difference in self memory after the Eyes test was included as a covariate) but not for memory in other-referential or non-social encoding conditions. Furthermore, some individual differences in self-referential cognition were related to mentalizing, regardless of whether an individual had a diagnosis. As self memory increased, mentalizing also increased and the endorsement of autistic traits decreased. The same pattern of individual differences (regardless of diagnosis) was manifest in the relationship between alexithymia and mentalizing. Here, more alexithymia is related to worse mentalizing and more endorsement of autistic traits.
Although we found some individual differences in self-referential cognition and mentalizing that were independent of diagnosis, we also found some very interesting and important relationships between self-referential cognition and mentalizing that depend upon diagnosis. In particular, self-focused attention and mentalizing were differentially related depending on diagnosis. ASC individuals who are more self-focused are better at mentalizing. However, among controls, being more self-focused predicted less mentalizing ability. A similar pattern existed with self-focused attention and endorsement of autistic traits. ASC individuals who are more self-focused report having less autistic traits. Among controls, those who are more self-focused report having more autistic traits. Another measure that showed a differential relationship based on diagnosis was the private self-consciousness scale. In the social psychological literature, private self-consciousness is believed to be a reflective form of self-focused attention and is a good compliment to our performance measure of self-focused attention (the Self-Focus Sentence Completion task), which was more implicit and non-reflective in nature. On this measure of more reflective self-focused attention, a similar pattern emerged to that of implicit non-reflective self-focused attention. ASC individuals who report being higher in private self-consciousness are better at mentalizing. However, among controls private self-consciousness was not related to mentalizing. Because these results are correlational it is hard to tease apart why these relationships exist and further manipulations of self-focused attention will be needed to understand the mechanisms underlying these relationships. However, one finding is clear: being more self-focused appears to be beneficial for individuals with ASC. For now, we speculatively conclude that these findings might point to the idea that people with autism need to be more self-focused and have more metacognitive ability to accurately reflect on themselves in order to mentalize with others. If this is true, it would support the idea that simulation is a strategy that individuals with ASC could benefit from and may be helpful in informing novel treatments for more able high-functioning individuals on the autism spectrum.
Overall, our results implicate dysfunction within CMS regions of the brain in ASC. The concurrent impairments and relationships observed for both self-referential cognition and empathy are striking given that CMS regions are consistently recruited for both thinking about the psychological characteristics of oneself or others and when engaging in the exact same paradigm we employed in this study; the “SRE levels-of-processing paradigm”, , . However, within ASC, recent evidence points towards disruption of CMS function when mentalizing, , , ,  and during rest. The idea that CMS dysfunction is important in ASC is also corroborated by studies in individuals with high levels of alexithymia. Moriguchi and colleagues found similar behavioral results to our study, in that individuals with high levels of alexithymia are also impaired at mentalizing. Individuals with high levels of alexithymia also exhibited a similar pattern of reduced activation in CMS regions that were previously found using the same mentalizing task in ASC. Thus, our study converges on the idea that CMS dysfunction is paramount to the social and self-referential impairments in ASC. Future work investigating the developmental trajectory of CMS structural and functional organization in the brain is needed. Furthermore, future investigation on the development of self-referential cognition and its causal relationship with empathy may prove fruitful for informing our understanding of the neurodevelopment of ASC.
We are grateful to all the volunteers in this study. We would also like to thank Erin Ingudomnukul, Liliana Ruta, and Greg Pasco for help with recruitment and data collection. We are also grateful for all the insight comments and discussion provided by Bhismadev Chakrabarti and Ilona Roth.
Conceived and designed the experiments: SB SW ML. Performed the experiments: ML JB. Analyzed the data: ML. Wrote the paper: SB SW ML JB. Other: Designed the study: ML SW SC. Recruited participants: ML JB. Collected data: ML JB.
- 1. Kanner L (1943) Autistic disturbance of affective contact. Nerv Child 2: 217–250.L. Kanner1943Autistic disturbance of affective contact.Nerv Child2217250
- 2. Frith U (2003) Autism: Explaining the enigma. Malden, MA: Blackwell. U. Frith2003Autism: Explaining the enigma.Malden, MABlackwell
- 3. Happe F (1999) Autism: cognitive deficit or cognitive style? Trends Cogn Sci 3: 216–222.F. Happe1999Autism: cognitive deficit or cognitive style?Trends Cogn Sci3216222
- 4. Pennington BF, Ozonoff S (1996) Executive functions and developmental psychopathology. J Child Psychol Psychiatry 37: 51–87.BF PenningtonS. Ozonoff1996Executive functions and developmental psychopathology.J Child Psychol Psychiatry375187
- 5. Baron-Cohen S (1995) Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press. S. Baron-Cohen1995Mindblindness: An essay on autism and theory of mind.Cambridge, MAMIT Press
- 6. Wang AT, Lee SS, Sigman M, Dapretto M (2007) Reading affect in the face and voice: neural correlates of interpreting communicative intent in children and adolescents with autism spectrum disorders. Arch Gen Psychiatry 64: 698–708.AT WangSS LeeM. SigmanM. Dapretto2007Reading affect in the face and voice: neural correlates of interpreting communicative intent in children and adolescents with autism spectrum disorders.Arch Gen Psychiatry64698708
- 7. Lee PS, Foss-Feig J, Henderson JG, Kenworthy LE, Gilotty L, et al. (2007) Atypical neural substrates of Embedded Figures Task performance in children with Autism Spectrum Disorder. Neuroimage. PS LeeJ. Foss-FeigJG HendersonLE KenworthyL. Gilotty2007Atypical neural substrates of Embedded Figures Task performance in children with Autism Spectrum Disorder.Neuroimage
- 8. Manjaly ZM, Bruning N, Neufang S, Stephan KE, Brieber S, et al. (2007) Neurophysiological correlates of relatively enhanced local visual search in autistic adolescents. Neuroimage 35: 283–291.ZM ManjalyN. BruningS. NeufangKE StephanS. Brieber2007Neurophysiological correlates of relatively enhanced local visual search in autistic adolescents.Neuroimage35283291
- 9. Ring HA, Baron-Cohen S, Wheelwright S, Williams SC, Brammer M, et al. (1999) Cerebral correlates of preserved cognitive skills in autism: a functional MRI study of embedded figures task performance. Brain 122 ( Pt 7): 1305–1315.HA RingS. Baron-CohenS. WheelwrightSC WilliamsM. Brammer1999Cerebral correlates of preserved cognitive skills in autism: a functional MRI study of embedded figures task performance.Brain122 ( Pt 7)13051315
- 10. Just MA, Cherkassky VL, Keller TA, Kana RK, Minshew NJ (2007) Functional and anatomical cortical underconnectivity in autism: evidence from an FMRI study of an executive function task and corpus callosum morphometry. Cereb Cortex 17: 951–961.MA JustVL CherkasskyTA KellerRK KanaNJ Minshew2007Functional and anatomical cortical underconnectivity in autism: evidence from an FMRI study of an executive function task and corpus callosum morphometry.Cereb Cortex17951961
- 11. Kana RK, Keller TA, Minshew NJ, Just MA (2007) Inhibitory Control in High-Functioning Autism: Decreased Activation and Underconnectivity in Inhibition Networks. Biol Psychiatry 62: 198–206.RK KanaTA KellerNJ MinshewMA Just2007Inhibitory Control in High-Functioning Autism: Decreased Activation and Underconnectivity in Inhibition Networks.Biol Psychiatry62198206
- 12. Just MA, Cherkassky VL, Keller TA, Minshew NJ (2004) Cortical activation and synchronization during sentence comprehension in high-functioning autism: evidence of underconnectivity. Brain 127: 1811–1821.MA JustVL CherkasskyTA KellerNJ Minshew2004Cortical activation and synchronization during sentence comprehension in high-functioning autism: evidence of underconnectivity.Brain12718111821
- 13. Courchesne E, Pierce K (2005) Why the frontal cortex in autism might be talking only to itself: local over-connectivity but long-distance disconnection. Curr Opin Neurobiol 15: 225–230.E. CourchesneK. Pierce2005Why the frontal cortex in autism might be talking only to itself: local over-connectivity but long-distance disconnection.Curr Opin Neurobiol15225230
- 14. Murias M, Webb SJ, Greenson J, Dawson G (2007) Resting State Cortical Connectivity Reflected in EEG Coherence in Individuals With Autism. Biol Psychiatry 62: 270–273.M. MuriasSJ WebbJ. GreensonG. Dawson2007Resting State Cortical Connectivity Reflected in EEG Coherence in Individuals With Autism.Biol Psychiatry62270273
- 15. Welchew DE, Ashwin C, Berkouk K, Salvador R, Suckling J, et al. (2005) Functional disconnectivity of the medial temporal lobe in Asperger's syndrome. Biol Psychiatry 57: 991–998.DE WelchewC. AshwinK. BerkoukR. SalvadorJ. Suckling2005Functional disconnectivity of the medial temporal lobe in Asperger's syndrome.Biol Psychiatry57991998
- 16. Hurlburt RT, Happe F, Frith U (1994) Sampling the form of inner experience in three adults with Asperger syndrome. Psychol Med 24: 385–395.RT HurlburtF. HappeU. Frith1994Sampling the form of inner experience in three adults with Asperger syndrome.Psychol Med24385395
- 17. Lee A, Hobson RP (1998) On developing self-concepts: a controlled study of children and adolescents with autism. J Child Psychol Psychiatry 39: 1131–1144.A. LeeRP Hobson1998On developing self-concepts: a controlled study of children and adolescents with autism.J Child Psychol Psychiatry3911311144
- 18. Toichi M, Kamio Y, Okada T, Sakihama M, Youngstrom EA, et al. (2002) A lack of self-consciousness in autism. Am J Psychiatry 159: 1422–1424.M. ToichiY. KamioT. OkadaM. SakihamaEA Youngstrom2002A lack of self-consciousness in autism.Am J Psychiatry15914221424
- 19. Hill E, Berthoz S, Frith U (2004) Brief report: cognitive processing of own emotions in individuals with autistic spectrum disorder and in their relatives. J Autism Dev Disord 34: 229–235.E. HillS. BerthozU. Frith2004Brief report: cognitive processing of own emotions in individuals with autistic spectrum disorder and in their relatives.J Autism Dev Disord34229235
- 20. Hobson PR, Chidambi G, Lee A, Meyer J (2006) Foundations for self-awareness: An exploration through autism. Monogr Soc Res Child Dev 71: vii–166.PR HobsonG. ChidambiA. LeeJ. Meyer2006Foundations for self-awareness: An exploration through autism.Monogr Soc Res Child Dev71vii166
- 21. Lee A, Hobson RP, Chiat S (1994) I, you, me, and autism: an experimental study. J Autism Dev Disord 24: 155–176.A. LeeRP HobsonS. Chiat1994I, you, me, and autism: an experimental study.J Autism Dev Disord24155176
- 22. Jordan RR (1989) An experimental comparison of the understanding and use of speaker-addressee personal pronouns in autistic children. Br J Disord Commun 24: 169–179.RR Jordan1989An experimental comparison of the understanding and use of speaker-addressee personal pronouns in autistic children.Br J Disord Commun24169179
- 23. Loveland KA, Landry SH (1986) Joint attention and language in autism and developmental language delay. J Autism Dev Disord 16: 335–349.KA LovelandSH Landry1986Joint attention and language in autism and developmental language delay.J Autism Dev Disord16335349
- 24. Mor N, Winquist J (2002) Self-focused attention and negative affect: a meta-analysis. Psychol Bull 128: 638–662.N. MorJ. Winquist2002Self-focused attention and negative affect: a meta-analysis.Psychol Bull128638662
- 25. Rogers TB, Kuiper NA, Kirker WS (1977) Self-reference and the encoding of personal information. J Pers Soc Psychol 35: 677–688.TB RogersNA KuiperWS Kirker1977Self-reference and the encoding of personal information.J Pers Soc Psychol35677688
- 26. Symons CS, Johnson BT (1997) The self-reference effect in memory: a meta-analysis. Psychol Bull 121: 371–394.CS SymonsBT Johnson1997The self-reference effect in memory: a meta-analysis.Psychol Bull121371394
- 27. Hull JG, Van Treuren RR, Ashford SJ, Propson P, Andrus BW (1988) Self-consciousness and the processing of self-relevant information. J Pers Soc Psychol 54: 452–465.JG HullRR Van TreurenSJ AshfordP. PropsonBW Andrus1988Self-consciousness and the processing of self-relevant information.J Pers Soc Psychol54452465
- 28. Klein SB, Kihlstrom JF (1986) Elaboration, organization, and the self-reference effect in memory. J Exp Psychol Gen 115: 26–38.SB KleinJF Kihlstrom1986Elaboration, organization, and the self-reference effect in memory.J Exp Psychol Gen1152638
- 29. Klein SB, Loftus J (1988) The nature of self-referent encoding: The contributions of elaborative and organizational processes. J Pers Soc Psychol 55: 5–11.SB KleinJ. Loftus1988The nature of self-referent encoding: The contributions of elaborative and organizational processes.J Pers Soc Psychol55511
- 30. McIntosh DN, Reichmann-Decker A, Winkielman P, Wilbarger JL (2006) When the social mirror breaks: deficits in automatic, but not voluntary, mimicry of emotional facial expressions in autism. Dev Sci 9: 295–302.DN McIntoshA. Reichmann-DeckerP. WinkielmanJL Wilbarger2006When the social mirror breaks: deficits in automatic, but not voluntary, mimicry of emotional facial expressions in autism.Dev Sci9295302
- 31. Baron-Cohen S, Knickmeyer RC, Belmonte MK (2005) Sex differences in the brain: implications for explaining autism. Science 310: 819–823.S. Baron-CohenRC KnickmeyerMK Belmonte2005Sex differences in the brain: implications for explaining autism.Science310819823
- 32. Baron-Cohen S (2002) The extreme male brain theory of autism. Trends Cogn Sci 6: 248–254.S. Baron-Cohen2002The extreme male brain theory of autism.Trends Cogn Sci6248254
- 33. Minio-Paluello I, Baron-Cohen S, Avenanti A, Walsh V, Aglioti SMAbsence of ‘sensorimotor contagion’ during pain observation in Asperger Syndrome. I. Minio-PaluelloS. Baron-CohenA. AvenantiV. WalshSM AgliotiAbsence of ‘sensorimotor contagion’ during pain observation in Asperger Syndrome
- 34. Amodio DM, Frith CD (2006) Meeting of minds: the medial frontal cortex and social cognition. Nat Rev Neurosci 7: 268–277.DM AmodioCD Frith2006Meeting of minds: the medial frontal cortex and social cognition.Nat Rev Neurosci7268277
- 35. Kelley WM, Macrae CN, Wyland CL, Caglar S, Inati S, et al. (2002) Finding the self? An event-related fMRI study. J Cogn Neurosci 14: 785–794.WM KelleyCN MacraeCL WylandS. CaglarS. Inati2002Finding the self? An event-related fMRI study.J Cogn Neurosci14785794
- 36. Macrae CN, Moran JM, Heatherton TF, Banfield JF, Kelley WM (2004) Medial prefrontal activity predicts memory for self. Cereb Cortex 14: 647–654.CN MacraeJM MoranTF HeathertonJF BanfieldWM Kelley2004Medial prefrontal activity predicts memory for self.Cereb Cortex14647654
- 37. Ochsner KN, Beer JS, Robertson ER, Cooper JC, Gabrieli JD, et al. (2005) The neural correlates of direct and reflected self-knowledge. Neuroimage 28: 797–814.KN OchsnerJS BeerER RobertsonJC CooperJD Gabrieli2005The neural correlates of direct and reflected self-knowledge.Neuroimage28797814
- 38. Saxe R, Moran JM, Scholz J, Gabrieli J (2006) Overlapping and non-overlapping brain regions for theory of mind and self reflection in individual subjects. Soc Cogn Affect Neurosci 1: 229–234.R. SaxeJM MoranJ. ScholzJ. Gabrieli2006Overlapping and non-overlapping brain regions for theory of mind and self reflection in individual subjects.Soc Cogn Affect Neurosci1229234
- 39. Gilbert SJ, Spengler S, Simons JS, Steele JD, Lawrie SM, et al. (2006) Functional specialization within rostral prefrontal cortex (area 10): a meta-analysis. J Cogn Neurosci 18: 932–948.SJ GilbertS. SpenglerJS SimonsJD SteeleSM Lawrie2006Functional specialization within rostral prefrontal cortex (area 10): a meta-analysis.J Cogn Neurosci18932948
- 40. Wang AT, Lee SS, Sigman M, Dapretto M (2006) Neural basis of irony comprehension in children with autism: the role of prosody and context. Brain 129: 932–943.AT WangSS LeeM. SigmanM. Dapretto2006Neural basis of irony comprehension in children with autism: the role of prosody and context.Brain129932943
- 41. Baron-Cohen S, Ring HA, Wheelwright S, Bullmore ET, Brammer MJ, et al. (1999) Social intelligence in the normal and autistic brain: an fMRI study. Eur J Neurosci 11: 1891–1898.S. Baron-CohenHA RingS. WheelwrightET BullmoreMJ Brammer1999Social intelligence in the normal and autistic brain: an fMRI study.Eur J Neurosci1118911898
- 42. Happe F, Ehlers S, Fletcher P, Frith U, Johansson M, et al. (1996) ‘Theory of mind’ in the brain. Evidence from a PET scan study of Asperger syndrome. Neuroreport 8: 197–201.F. HappeS. EhlersP. FletcherU. FrithM. Johansson1996‘Theory of mind’ in the brain. Evidence from a PET scan study of Asperger syndrome.Neuroreport8197201
- 43. Castelli F, Frith C, Happe F, Frith U (2002) Autism, Asperger syndrome and brain mechanisms for the attribution of mental states to animated shapes. Brain 125: 1839–1849.F. CastelliC. FrithF. HappeU. Frith2002Autism, Asperger syndrome and brain mechanisms for the attribution of mental states to animated shapes.Brain12518391849
- 44. Mitchell JP, Banaji MR, Macrae CN (2005) The link between social cognition and self-referential thought in the medial prefrontal cortex. J Cogn Neurosci 17: 1306–1315.JP MitchellMR BanajiCN Macrae2005The link between social cognition and self-referential thought in the medial prefrontal cortex.J Cogn Neurosci1713061315
- 45. Mitchell JP, Macrae CN, Banaji MR (2006) Dissociable medial prefrontal contributions to judgments of similar and dissimilar others. Neuron 50: 655–663.JP MitchellCN MacraeMR Banaji2006Dissociable medial prefrontal contributions to judgments of similar and dissimilar others.Neuron50655663
- 46. Humphrey N (1984) Consciousness regained. Oxford: Oxford University Press. N. Humphrey1984Consciousness regained.OxfordOxford University Press
- 47. Heal J (1986) Simulation, theory, and content. In: Carruthers P, Smith PK, editors. Theories of theories of mind. Cambridge, UK: Cambridge University Press. pp. 75–89.J. Heal1986Simulation, theory, and content.P. CarruthersPK SmithTheories of theories of mindCambridge, UKCambridge University Press7589
- 48. Ames DR (2004) Inside the mind reader's tool kit: projection and stereotyping in mental state inference. J Pers Soc Psychol 87: 340–353.DR Ames2004Inside the mind reader's tool kit: projection and stereotyping in mental state inference.J Pers Soc Psychol87340353
- 49. Association AP (1994) Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Association. AP Association1994Diagnostic and Statistical Manual of Mental Disorders.Washington, DCAmerican Psychiatric Association
- 50. ICD-10 (1994) International Classification of Diseases. Geneva, Switzerland: World Heath Organization. ICD-101994International Classification of Diseases.Geneva, SwitzerlandWorld Heath Organization
- 51. Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E (2001) The autism-spectrum quotient (AQ): evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. J Autism Dev Disord 31: 5–17.S. Baron-CohenS. WheelwrightR. SkinnerJ. MartinE. Clubley2001The autism-spectrum quotient (AQ): evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians.J Autism Dev Disord31517
- 52. Weschler D (1999) Wechsler abbreviated scale of intelligence. San Antonio, TX: The Psychological Corporation. D. Weschler1999Wechsler abbreviated scale of intelligence.San Antonio, TXThe Psychological Corporation
- 53. Craik FIM, Tulving E (1975) Depth of processing and the retention of words in episodic memory. J Exp Psychol Gen 11: 268–294.FIM CraikE. Tulving1975Depth of processing and the retention of words in episodic memory.J Exp Psychol Gen11268294
- 54. Pfeifer JH, Lieberman MD, Dapretto M (2007) “I know what you are but what am I?!”: Neural bases of self- and social knowledge retrieval in children and adults. J Cogn Neurosci 19: 1323–1337.JH PfeiferMD LiebermanM. Dapretto2007“I know what you are but what am I?!”: Neural bases of self- and social knowledge retrieval in children and adults.J Cogn Neurosci1913231337
- 55. Anderson NH (1968) Likableness ratings of 555 personality-trait words. J Pers Soc Psychol 9: 272–279.NH Anderson1968Likableness ratings of 555 personality-trait words.J Pers Soc Psychol9272279
- 56. Aron A, Aron EN, Smollan D (1992) Inclusion of other in the self scale and the structure of interpersonal closeness. J Pers Soc Psychol 63: 596–612.A. AronEN AronD. Smollan1992Inclusion of other in the self scale and the structure of interpersonal closeness.J Pers Soc Psychol63596612
- 57. Baron-Cohen S, Wheelwright S (2004) The empathy quotient: an investigation of adults with Asperger syndrome or high functioning autism, and normal sex differences. J Autism Dev Disord 34: 163–175.S. Baron-CohenS. Wheelwright2004The empathy quotient: an investigation of adults with Asperger syndrome or high functioning autism, and normal sex differences.J Autism Dev Disord34163175
- 58. Davis MH (1980) A multidimensional approach to individual differences in empathy. JSAS Catalogue of Selected Documents in Psychology 10. MH Davis1980A multidimensional approach to individual differences in empathy.JSAS Catalogue of Selected Documents in Psychology 10
- 59. Doherty RW (1997) The emotional contagion scale: A measure of individual differences. Journal of Nonverbal Behavior 21: 131–154.RW Doherty1997The emotional contagion scale: A measure of individual differences.Journal of Nonverbal Behavior21131154
- 60. Lawrence EJ, Shaw P, Baker D, Baron-Cohen S, David AS (2004) Measuring empathy: reliability and validity of the Empathy Quotient. Psychol Med 34: 911–919.EJ LawrenceP. ShawD. BakerS. Baron-CohenAS David2004Measuring empathy: reliability and validity of the Empathy Quotient.Psychol Med34911919
- 61. Baron-Cohen S, Wheelwright S, Hill J, Raste Y, Plumb I (2001) The “Reading the Mind in the Eyes” Test revised version: a study with normal adults, and adults with Asperger syndrome or high-functioning autism. J Child Psychol Psychiatry 42: 241–251.S. Baron-CohenS. WheelwrightJ. HillY. RasteI. Plumb2001The “Reading the Mind in the Eyes” Test revised version: a study with normal adults, and adults with Asperger syndrome or high-functioning autism.J Child Psychol Psychiatry42241251
- 62. Fenigstein A, Scheier MF, Buss AH (1975) Public and private self-consciousness: Assessment and theory. J Consult Clin Psychol 43: 522–527.A. FenigsteinMF ScheierAH Buss1975Public and private self-consciousness: Assessment and theory.J Consult Clin Psychol43522527
- 63. Bagby RM, Parker JD, Taylor GJ (1994) The twenty-item Toronto Alexithymia Scale–I. Item selection and cross-validation of the factor structure. J Psychosom Res 38: 23–32.RM BagbyJD ParkerGJ Taylor1994The twenty-item Toronto Alexithymia Scale–I. Item selection and cross-validation of the factor structure.J Psychosom Res382332
- 64. Exner JE (1973) The self-focus sentence completion: A study of egocentricity. J Pers Assess 37: 437–455.JE Exner1973The self-focus sentence completion: A study of egocentricity.J Pers Assess37437455
- 65. Pennebaker JW, Francis ME, Booth RJ (2001) Linguistic Inquiry and Word Count (LIWC): A computerized text analysis program. Mahwah, NJ: Erlbaum. JW PennebakerME FrancisRJ Booth2001Linguistic Inquiry and Word Count (LIWC): A computerized text analysis program.Mahwah, NJErlbaum
- 66. Davis D, Brock TC (1975) Use of first person pronouns as a function of increased objective self-awareness and prior feedback. J Exp Soc Psychol 11: 381–388.D. DavisTC Brock1975Use of first person pronouns as a function of increased objective self-awareness and prior feedback.J Exp Soc Psychol11381388
- 67. Ingram RE, Cruet D, Johnson BR, Wisnicki KS (1988) Self-focused attention, gender, gender role, and vulnerability to negative affect. J Pers Soc Psychol 55: 967–978.RE IngramD. CruetBR JohnsonKS Wisnicki1988Self-focused attention, gender, gender role, and vulnerability to negative affect.J Pers Soc Psychol55967978
- 68. Rude SS, Gortner E, Pennebaker JW (2004) Language use of depressed and depression-vulnerable college students. Cognition and Emotion 18: 1121–1133.SS RudeE. GortnerJW Pennebaker2004Language use of depressed and depression-vulnerable college students.Cognition and Emotion1811211133
- 69. Aiken LS, West SG (1991) Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. LS AikenSG West1991Multiple regression: Testing and interpreting interactions.Newbury Park, CASage
- 70. Aron A, Aron EN, Tudor M, Nelson G (1991) Close relationships as including other in the self. J Pers Soc Psychol 60: 241–253.A. AronEN AronM. TudorG. Nelson1991Close relationships as including other in the self.J Pers Soc Psychol60241253
- 71. Berthoz S, Hill EL (2005) The validity of using self-reports to assess emotion regulation abilities in adults with autism spectrum disorder. Eur Psychiatry 20: 291–298.S. BerthozEL Hill2005The validity of using self-reports to assess emotion regulation abilities in adults with autism spectrum disorder.Eur Psychiatry20291298
- 72. Kennedy DP, Redcay E, Courchesne E (2006) Failing to deactivate: resting functional abnormalities in autism. Proc Natl Acad Sci U S A 103: 8275–8280.DP KennedyE. RedcayE. Courchesne2006Failing to deactivate: resting functional abnormalities in autism.Proc Natl Acad Sci U S A10382758280
- 73. Moriguchi Y, Ohnishi T, Lane RD, Maeda M, Mori T, et al. (2006) Impaired self-awareness and theory of mind: an fMRI study of mentalizing in alexithymia. Neuroimage 32: 1472–1482.Y. MoriguchiT. OhnishiRD LaneM. MaedaT. Mori2006Impaired self-awareness and theory of mind: an fMRI study of mentalizing in alexithymia.Neuroimage3214721482