Figure 4 - uploaded by Mariapaola D'Imperio
Content may be subject to copyright.
Distributions of time points of F0 peaks and head nod gesture phases for alignment strategy 1 (i.e. k-means cluster 1).
Source publication
Previous research has shown that rhythmic head movement accompanies F0 modulations in speech (Munhall et al., 1994) and that this co-verbal head movement may be linked to prosodic features such as pitch accents and prosodic boundaries (Esteve-Gibert et al., 2017; Hadar et al., 1984; House et al., 2001). In this study, we examined how the production...
Contexts in source publication
Context 1
... the data were split by performing a twogroup k-means clustering of the time point of maximum velocity of the head nod gesture, in order to determine the nature of the two groups underlying the global pattern observed in Figure 3. This clustering resulted in 64 items (55% of the total data) in cluster 1 and 52 items (45% of the total data) in cluster 2. Figure 4 displays the normalized probability distributions of the data in cluster 1; we will refer to the results for this cluster as "alignment strategy 1." Here, the onset of the head nod stroke occurs before the target word, the maximum velocity of the nod is roughly aligned with the start of the target word, and the apex of the head nod stroke is aligned with the F0 peak. An example of alignment strategy 1 can seen above in Figure 2. Figure 5 displays the normalized probability distributions of the data in cluster 2; we will refer to the results for this cluster as "alignment strategy 2." Here, the entire stroke of the head nod gesture is shifted forward in time compared to alignment strategy 1. ...
Context 2
... the data were split by performing a twogroup k-means clustering of the time point of maximum velocity of the head nod gesture, in order to determine the nature of the two groups underlying the global pattern observed in Figure 3. This clustering resulted in 64 items (55% of the total data) in cluster 1 and 52 items (45% of the total data) in cluster 2. Figure 4 displays the normalized probability distributions of the data in cluster 1; we will refer to the results for this cluster as "alignment strategy 1." Here, the onset of the head nod stroke occurs before the target word, the maximum velocity of the nod is roughly aligned with the start of the target word, and the apex of the head nod stroke is aligned with the F0 peak. An example of alignment strategy 1 can seen above in Figure 2. Figure 5 displays the normalized probability distributions of the data in cluster 2; we will refer to the results for this cluster as "alignment strategy 2." Here, the entire stroke of the head nod gesture is shifted forward in time compared to alignment strategy 1. ...
Similar publications
In this paper, we present a novel interface named GazeHand which is designed for distant object interaction in virtual reality. The GazeHand interface translates the virtual hands near to a distant object that the user is looking at and allows direct hand gesture interaction for manipulating an object. Either eye-gaze or head-gaze can be applied to...
Citations
... We move our hands (manual gestures), our heads (head gestures), and our facial expressions in temporal and functional synchrony with our speech. From a timing point of view, prosodic events in speech serve as anchoring landmarks for gestural alignment, as points of maximal displacement in gestures usually occur within the temporal limits of prominent words or syllables in speech (e.g., Carignan et al., 2020;Esteve-Gibert & Prieto, 2013;Esteve-Gibert et al., 2017;Krivokapic et al., 2017;Leonard & Cummins, 2011). From a functional point of view, gestures convey meanings that can be referential (representing an entity or event deictically, iconically, or metaphorically) and non-referential (signaling information structure, modal information, or discourse cohesion) (Kendon, 1980;McNeill, 2000; and many others thereafter). ...
Previous evidence suggests that children's mastery of prosodic modulations to signal the informational status of discourse referents emerges quite late in development. In the present study, we investigate the children's use of head gestures as it compares to prosodic cues to signal a referent as being contrastive relative to a set of possible alternatives. A group of French-speaking pre-schoolers were audio-visually recorded while playing in a semi-spontaneous but controlled production task, to elicit target words in the context of broad focus, contrastive focus, or corrective focus utterances. We analysed the acoustic features of the target words (syllable duration and word-level pitch range), as well as the head gesture features accompanying these target words (head gesture type, alignment patterns with speech). We found that children's production of head gestures, but not their use of either syllable duration or word-level pitch range, was affected by focus condition. Children mostly aligned head gestures with relevant speech units, especially when the target word was in phrase-final position. Moreover, the presence of a head gesture was linked to greater syllable duration patterns in all focus conditions. Our results show that (a) 4- and 5-year-old French-speaking children use head gestures rather than prosodic cues to mark the informational status of discourse referents, (b) the use of head gestures may gradually entrain the production of adult-like prosodic features, and that (c) head gestures with no referential relation with speech may serve a linguistic structuring function in communication, at least during language development.
