Abstract:

Individuals with autism spectrum disorders (ASD) are typically characterized by impaired social interactions that are thought to be related to deficits in empathy. While cognitive empathy deficit in ASD is widely recognized, it remains controversial whether individuals with ASD have a deficiency in emotional empathy. According to the shared representation theory, psychological and neuronal mechanisms involved in the personal experience of an emotional or somatosensory state are also engaged while empathizing with other individuals in those states. It suggests that the deficits of empathy seen in the ASD population could arise from the atypical experience of first-hand pain. Mild, subclinical forms of the characteristics associated with ASD are referred to as autistic traits. Individuals with high autistic traits exhibit sensory, emotional, and social behaviors similar to those with ASD. Given the relationship between pain empathy and first-hand pain as well as the similarity between autistic traits and ASD, the present study tested the hypothesis that autistic traits in the general population would influence pain empathic responses, which could be contributed by first-hand pain-related profiles.

In Experiment 1, we recruited 1131 healthy participants to complete the mandarin version of the Autism-spectrum Quotient (AQ) questionnaire, as a reliable measurement of ASD symptomatology in typically developing adults. The total AQ score ranges from 0 to 50, with higher scores indicating higher levels of autistic traits. According to the overall score distribution (20.08 ± 0.17, M ± SE), we divided participants into HAQ (with high autistic traits, AQ score ≥ 27) and LAQ (with low autistic traits, AQ score ≤ 13) groups by taking the top and bottom 10% of the overall distribution. Then, a subset of participants from both HAQ (n= 30, 15 males, aged 21.30 ± 0.31 years) and LAQ (n= 30, 16 males, aged 20.77 ± 0.34 years) groups were randomly recruited to participate in the EEG experiment. We adopted an ecological pain empathy paradigm and compared behavioral and neural activity between individuals with HAQ and those with LAQ. During the pain empathy paradigm (Figure 1), the participants either perceived the painful electrical stimuli themselves or witnessed the delivery of painful electrical stimuli to their partners in certain and uncertain contexts.

When perceiving pain themselves, behavioral responses and brain responses were comparable between HAQ and LAQ groups (all p> 0.05). When witnessing others in pain, participants in the HAQ group had greater amplitudes of the P2 component (5.68 ± 0.74 μV vs. 3.50 ± 0.74 μV, p = 0.041) on the event-related potentials and reported higher ratings of unpleasantness (3.53 ± 0.29 vs. 2.44 ± 0.29, p = 0.009) than those in the LAQ group. HAQ reported higher ratings of fear of pain of predictability cue (2.97 ± 0.33 vs. 1.78 ± 0.33, p = 0.014). The between-group differences in the behavioral and neural responses related to pain empathy were not moderated by certainty of the context (certain or uncertain) (Figure 2 & Figure 3).

We further tested the mediating role of fear of pain on the link between autistic traits on emotional empathy for others’ pain. Autistic traits showed a total effect (c = 0.40, SE = 0.12, 95% CI = [0.16, 0.64]) on emotional empathy for others’ pain. Autistic traits showed a direct effect (c’ = 0.29, SE = 0.12, 95% CI = [0.05, 0.53]) on emotional empathy for others’ pain. Autistic traits showed an indirect effect (a×b = 0.11, SE = 0.06, 95% CI = [0.01, 0.25]) on emotional empathy for others’ pain through fear of pain. These results revealed that the between-group differences in the unpleasantness elicited by witnessing others’ pain could be contributed by the greater fear of pain while anticipating the upcoming painful stimuli (Figure 4).

In Experiment 2, the relationship among autistic traits, pain-related profiles, and trait empathy was assessed in randomly recruited participants (381 healthy college students; 202 males, aged 20.95 ± 0.12 years). We found that autistic trait levels were negatively correlated with scores on the perspective-taking subscale of the Interpersonal Reactivity Index (r(379) = −0.19, p < 0.001) and positively correlated with the personal distress subscale (r(379) = 0.27, p < 0.001) (Table 1). Importantly, pain-related fear and pain catastrophizing mediated the link between autistic traits and personal distress (total effect: c = 0.27, SE= 0.05, 95% CI = [0.17, 0.36], p < 0.001; indirect effect: a×b = 0.11, SE= 0.04, 95% CI = [0.05, 0.19], p < 0.001; direct effect: c’ = 0.16, SE= 0.05, 95% CI = [0.07, 0.25], p = 0.002) (Figure 5).

Data from Experiments 1 and 2 demonstrated that autistic traits heighten emotional empathy, which can be explained by the negative emotion and cognition toward pain. Given the similarities between individuals with high autistic traits and ASD, this finding may help to expand the biological mechanisms underlying ASD, such as explaining empathy deficits or other social difficulties seen in the ASD from the perspective of atypical pain-related profiles. Future studies should combine multiple modalities of painful stimulations and multidimensional pain assessments to comprehensively characterize pain-related profiles among individuals with high autistic traits or ASD, and establish linkage between pain-related profiles and empathy or social deficits. This understanding has the potential to provide targets for clinical interventions and treatments of ASD.

Key words: autistic traits, pain empathy, pain-related fear, pain catastrophizing, event-related potentials