Main findings

The specific objectives of the project could be reached by answering the following questions:

1) What are the evolutionary origins of the potential to imitate? and
2) What types of experience enhance the potential for imitation?

The EDICI project has shown that the capacity to imitate is not something that is innate—present or absent in a creature at birth. Rather, the ability to imitate is something that an individual develops as a consequence of observing their own actions and crucially, engaging in social interaction with others. In the animal kingdom, humans are the most gifted imitators, not because we are intrinsically smarter than other animals, but because our culture-soaked world provides rich sources of the right kind of experience for the development of imitation. However, our work has also shown that other animals, including monkeys and dogs, can also imitate when they receive this kind of sensorimotor experience. Furthermore, they sometimes imitate with remarkable precision, and sometimes highly selectively, taking the specific circumstances of the demonstration into account.

3) How does intentional control of imitation change in the course of human development? and
4) Do the neuro-cognitive mechanisms that distinguish self from others play a key role in intentional control of human imitative performance?

It is now widely acknowledged that the observation of an action leads to an activation of a corresponding motor representation in the observer (shared representations). Such shared representations form the basis for our ability to imitate. However, one fundamental question regarding such shared representations is how we can avoid automatic imitation and how we can distinguish self-generated from externally triggered motor representations. Our research identified the brain regions that are involved in the control of shared representations. We could show that the posterior superior temporal sulcus and the anterior fronto-median cortex are critical for self/other distinction and avoiding imitative response tendencies. Furthermore, our research provides a new perspective on the question how shared representations are related to other social cognitive skills such as mentalizing. In contrast to the dominant view that tries to link shared representations directly to mentalizing abilities, we rather assume that it the control of shared representations that forms the basis for mentalizing.

5) Is intentional control of imitative performance uniquely human?

Whether the intentional control of imitation is uniquely human is still an open question. However, the dominant role of the prefrontal cortex in the control of imitative behaviour and the involvement of prefrontal brain regions in higher order social cognitive skills such as mentalizing suggests that humans have developed extraordinary skills in the control of imitation. Our results have furthermore demonstrated that human imitative learning is a unique mechanism of naïve pedagogy that facilitates fast and efficient cultural knowledge transfer. Unlike imitation in other species, human imitative learning is inferential, context-sensitive, selective, and it is being guided by ostensive social cuing that convey the overt communicative intentions of others.

Clinical relevance

Our results have implications of two kinds with respect to the assistance of people with impaired imitative ability. First, they suggest that remedial programmes designed to overcome this impairment specifically (rather than to treat its sequelae) are most likely to be successful when they are based on sensorimotor training. Second, our results clarify some conceptual issues with clinical relevance. For example, it was assumed that imitation forms the basis for our ability to understand other peoples behavior. However, our results suggest that it is rather the control of imitative behavior that shows a functional overlap with the ability to mentalize. This opens a different perspective on our understanding of developmental disorders such as autism. While research has recently focused on imitative deficits in this patient group we would rather suggest to look at deficits in the control of shared representations. This would also mean that potential intervention programs should focus on improving the ability for self/other distinction rather than imitative abilities.