Graphics Reference
In-Depth Information
used until recently in an indefinite way. Alignment refers then to the
endorsement of the sequence/activity in progress and contrasts with
the notion of affiliation which refers to the endorsement of the previous
speaker's evaluative positioning, or stance (cited by Gorish et al.,
2012:7). More specifically, Stivers (2008) uses the term of alignement
to describe actions by a second speaker which support the activity
being undertaken by the first speaker. In this way, the production of
backchannel signals in conversation can be considered as adapted and
expected responses from the listener during conversation (Bertrand et
al., 2007; Heldner et al., 2010 among others) and more particularly in
a storytelling activity in which the main speaker (narrator) is indeed
ratified as main speaker by the listener (Stivers, 2008). At last, Gorish
et al.'s study also constitutes a first attempt to develop a method that
enables the measurement of the acoustic similarity (in terms of f0
and intensity) of pitch contours in naturally occurring data. Similar
parameters were considered by De Looze et al. (2011) in a study on
prosodic convergence in spontaneous conversations. In Gorish et al.'s
study, the prosodic matching observed is then considered as a resource
used to demonstrate alignment with the prior action. In a similar
way, Bertrand and Priego-Valverde (2011) have shown that a copy
of some prosodic cues by both participants could be a resource to
demonstrate orientation to a humorous utterance expressed by the
speaker. A series of prosodic matching repetitions by both participants
is leading to the creation of a short sequence called joint fantasy
(Kotthoff, 2006).
4. Identification of Multimodal Repetition
4.1 Gesture repetition
The criteria we adopted for the repetition of hand gestures were very
much inspired from von Raffler-Engel (1986). For a hand gesture to
be considered as repeated, gesture phrase, lexicon and movement
trajectory have to be identical, that is the functional category of the
gesture, whereas other descriptive features like gesture space, tension,
amplitude and velocity do not have to be strictly identical to the
model for the gesture to be considered as a repetition of the model.
The criteria for the repetition of head movements and gaze orientation
are stricter than for hand gestures since we considered a gesture was
repeated if the movement in the repetition was strictly identical or
mirrored between the repetition and the model.
Search WWH ::




Custom Search