French Face-to-Face Interaction: Repetition as a Multimodal Resource - Coverbal Synchrony in Human-Machine Interaction - page 150

co-presence, and not necessarily an indication of understanding or
alignment with the speaker's proposition”.
There is more consensus on the fact that a conversation can be
considered as a joint action (Garrod and Pickering, 2004; Holler and
Wilkin, 2011; Kimbara, 2006; Shockley et al., 2009; Tabensky, 2001, to
cite but a few studies) which entails the co-construction of meaning by
all the participants to the interaction as suggested in the introduction.
3.2 Behavior and gesture repetition
There has already been quite a large body of work on the role of
behavior and gesture-pattern repetition and on the conditions for their
emergence. Lakin et al. (2003) observe that some situations activate a
desire to affi liate in the participants to an interaction and thus encourage
mimicry. This work is derived from Chartrand and Bargh (1999) who
noted the social role of mimicry. On experimental data, they observed
that postures and adaptors (in their study, the shaking of one's foot) were
regularly mimicked by the participants, and that when the confederate
mimicked the participants, the latter felt greater empathy with the model.
Working on posture and gaze, Shockley et al. (2009) fi nd that similar
gaze patterns emerge in participants together with the increase of joint
understanding. Also working on experimental data, they observe that
participants adopt more postural coordination when they see the same
words on a screen than when the words are different. Mol et al.
(2009) go further on experimental data as well. They find that
reproduced iconic gestures are not just imitation: only gestures
that are consistent with verbal content are copied. Much in the
same vein, Holler and Wilkin (2011) fi nd that mimicked hand gestures
in experimental conditions play an active role in the grounding process
and help create mutually shared understanding. Their classifi cation of
mimicked gestures is both semantic and formal. To count as repeated,
a gesture has to represent the same meaning or have the same referent,
use the same mode of representation and have the same overall
form. Drawing on data from a joint narration task, Kimbara (2006:45)
adds that temporal proximity together with co-referentiality between
a gesture and its repetition show “realizations of a shared image
construal”. Besides, she observes that not all the features of gestures
are repeated, but a subset has to be present in the repetition for the
gesture to be considered as mimicked. From her study, she concludes
that hand-gesture mimicry creates gesture catchments (McNeill, 2001)
across speakers.
In a later study, Kimbara (2008) notes that gesture repetition is
not a chance phenomenon. In experimental conditions, she notices