Abstract
Several theoretical frameworks have attempted to account for interpersonal synchronization processes. Cognitive theories suggest that synchronization is achieved through discrete and mutual corrections of asynchronies between the two partners. The dynamic theories are based on the assumption of a continuous coupling between the two systems, conceived as self-sustained oscillators. Finally, the complexity matching model is based on the assumption of a multi-scale coordination between the two interacting systems. As a first step, we develop statistical tests in order to identify, in experimental data, the typical signatures of these three modes of coordination. In particular, we propose a multifractal signature, based on the analysis of correlations between the multifractal spectra characterizing the series produced by the two interacting systems. We also develop a windowed cross-correlation analysis, which aims at revealing the nature of the local synchronization processes. These studies allow us to revisit a number of previous works. We show that if the synchronization of discrete tasks such as tapping relies on discrete correction processes of asynchronies, the synchronization of continuous tasks such as pendulum oscillations is essentially based on the same principles of discrete correction, and not on a continuous coupling of effectors. Our results also indicate that synchronization could be sustained by hybrid mechanisms mixing notably asynchronies correction and complexity matching. Finally we highlight that synchronized walking is based on a dominant effect of complexity matching, especially when partners are closely coupled (arm-in-arm walking). We propose in a second step to exploit this result to test the possibility of a restoration of complexity in the elderly. Aging has indeed been characterized as a process of gradual loss of complexity, and this effect has been particularly documented in the field of locomotion. In particular, it has been shown that the loss of complexity correlates in older people with the propensity to fall. Complex matching theory assumes that two interacting systems tend to align their complexity levels. It also assumes that when two systems of different levels of complexity interact, the more complex system tends to attract the less complex, causing an increase in complexity in the second. We show, in a protocol in which older people are invited to walk arm-in-arm with a younger companion, that synchronization between the two partners is achieved through a complexity matching effect, and that prolonged training in such synchronized walking allows a restoration of the complexity of locomotion in the elderly. This effect persists during a post-test conducted two weeks after the end of the training sessions. This result, in addition to reinforcing one of the essential aspects of the theory of complexity matching, opens new avenues of research for the design of rehabilitation and fall prevention strategies.