Emotional empathy in autism spectrum conditions: weak, intact, or heightened?

Dear Editor, We read with interest the Letter by Smith (2009) suggesting that individuals with Autism Spectrum Conditions (ASC) have heightened emotional empathy (EE). We are pleased to be invited to contribute to the discussion. We do this by commenting on both the evidence cited by Smith, and also the evidence he does not cite. We conceptually reframe the question being posed and argue that, in the light of a broader review of the evidence, empathic responses in people with ASC are in general reduced. However, a basic type of empathic response that would fall into Smith’s definition of EE may indeed be heightened (a self-oriented personal distress response to others’ distress). We end by suggesting additional avenues to be explored. Empathy is a process where another person’s mental state (which include intentions, desires, emotions and proprioceptive states) automatically activates in the observer a representation of the observed state. These representations, unless inhibited, prime associated autonomic and somatic responses in the observer, thus generating a state in the observer that is more appropriate to the state of the person being observed (Baron-Cohen and Wheelwright 2004). The empathic process can therefore be considered as a superordinate category that includes many phenomena sharing this same underlying mechanism (Preston and de Waal 2002). Minio-Paluello et al. (2009) addressed the embodied side of empathy in people with ASC by investigating the response to others’ pain using a paradigm that involved watching a hand being penetrated by a needle. Facial or vocal representations of pain were not shown, to avoid any confound due to emotion recognition difficulties (Humphreys et al. 2007; Ashwin et al. 2006; Kuusikko et al. 2009) or atypical face processing (Spezio et al. 2007; Behrmann et al. 2006) in ASC. This study found that when observing another person’s physical pain, unlike controls, participants with ASC did not show a neurophysiological response as if they were experiencing pain themselves, and nor did their neurophysiological response correlate with the imagined pain sensory qualities. Instead, participants with ASC represented others’ pain in relation to the level of self-oriented arousal experienced while watching pain affecting another person. Further, participants with high empathy and a low number of autistic traits showed a stronger embodied empathy response. Thus, embodiment of another person’s state is one possible mechanism underlying the empathic process (Niedenthal 2007; Gallese 2007) and one that seems reduced in ASC (Minio-Paluello et al. 2009; Williams 2008). Reduced embodiment in people with ASC does not directly relate to their level of EE, defined by Smith (2009) as ‘an emotional response in an individual that stems from and parallels the emotional state of another individual’ (italics added). In the study by Minio-Paluello et al. (2009), for example, neither participants with ASC nor controls reported vicariously experiencing any pain while watching the model’s hand being penetrated by the needle. I. Minio-Paluello (&) Dipartimento di Psicologia La Sapienza, Universita’ di Roma, Via dei Marsi 78, 00185 Rome, Italy e-mail: [email protected]