Content uploaded by Leonardo Barón Birchenall
Author content
All content in this area was uploaded by Leonardo Barón Birchenall on Mar 29, 2019
Content may be subject to copyright.
Synchronization of speech rhythm between Spanish-speaking interlocutors
Leonardo Barón Birchenall, Noël Nguyen
Aix Marseille Univ, CNRS, LPL, Aix-en-Provence, France
In spoken language communication, accommodation refers to the many processes
employed by talkers to adapt to each other. Within these processes, phonetic convergence is
associated with an increase in similarity in speech patterns between interlocutors over the
course of a conversation. In some theoretical frameworks, phonetic convergence is thought to
occur in an involuntary and automatic manner rather than intentionally [1].
The present work focuses on the interaction between speech rhythm and phonetic
convergence in a semi-interactive task (understanding rhythm as a “temporally regular
[iteration] of events which embody alternating strong and weak values of an observable
parameter” [2]). Specifically, given that a repeated speech stimulus requires both less
processing time and lower neural activation across repetitions, and that multiple repetitions
significantly enhance memory and learning [3], we propose that the use of regular rhythmic
structures during conversations produces more convergence between speakers with respect to
irregular rhythmic structures. To our best knowledge, the only existing research on this
particular topic is the one conducted by Späth et al. [4], who found more rhythmic
convergence between a healthy person and a model speaker than between individuals with
Parkinson’s disease and the same model speaker.
To test our hypothesis, we created a set of stimuli consisting of seven groups of 16 nine-
or eight-syllable Spanish sentences each. Each group had a particular rhythmic structure,
obtained through the arrangement of different types of words (oxytones, paroxytones,
proparoxytones and unstressed words) in feet of different length. Rhythmic structures were
composed as follows (unstressed syllables are represented by a lowercase x and stressed
syllables by an uppercase X and in uppercase within the sentences):
Regular structures: (1) Three feet, head to the right: xxXxxXxxX (e.g. la re-PÚ-bli-ca
NO ter-mi-NÓ) [the republic did not end]. (2) Three feet, head to the left: XxxXxxXxx (e.g.
E-llos es-PE-ran al MÉ-di-co) [they wait for the doctor]. (3) Four feet, head to the left:
XxXxXxXx (e.g. JUAN es-TÁ bus-CAN-do BA-rro) [John is looking for mud].
Irregular structures: (4) Three feet, head to the right: xxXxXxxXx (e.g. la es-PO-sa CA-
e sin VI-da) [the wife falls down dead]. (5) Three feet, head to the left: XxxxXxxXx (e.g.
CAR-los ter-mi-NÓ mi lla-MA-da) [Charles ended my call]. (6) Four feet, head to the left:
XxXXxxxXx (e.g. JUAN es-TÁ SIEM-pre so-me-TI-do) [John is always under control]. (7)
Four feet, head to the left: XxXXxxXx (e.g. JUAN sa-LIÓ RÁ-pi-do SIEM-pre) [John always
left quickly].
We tested four dyads of Spanish native speakers separately in a reading - repetition task
with different combinations of the rhythmic structures. In the task, each member of the dyad
must read a sentence and the other one must immediately repeat it. Participants alternate
between reading and repeating the sentences of each group. The order of presentation of the
sentences within a group, and that of the groups themselves, are randomized. A rhythmic
distance score, proposed by Späth et al. [4], was then used to determine the degree of
convergence between the interlocutors’ rhythms.
Results indicate a greater amount of convergence between regular structures than between
irregular ones, when feet nuclei are left aligned. We observed an overall tendency for the
regular utterances to present more similar metrical timing patterns between interlocutors than
the irregular ones, rather than a gradual augmentation of the resemblance between regular
utterances’ rhythms over the course of the task. Details will be given on the response patterns
observed in the other conditions (right-aligned feet nuclei), and implications for current
models of phonetic convergence in speech will be discussed.
74
[1] Louwerse, M., Dale, R., Bard, E., & Jeuniaux, P. 2012. Behavior matching in multimodal
communication is synchronized. Cognitive Science, 36(8), 1404-1426.
[2] Gibbon, D. 2015. Speech rhythms – modeling the groove. In R. Vogel & R. van de Vijver
(Eds.), Rhythm in Cognition and Grammar: A Germanic Perspective (pp. 108-161).
Berlin: De Gruyter Mouton.
[3] Falk, S., Rathcke, T., & Dalla Bella, S. 2014. When speech sounds like music. Journal of
Experimental Psychology: Human Perception and Performance, 40(4), 1491-1506.
[4] Späth, M., Aichert, I., Ceballos, A., Wagner, E., Miller, N., & Ziegler, W. 2016.
Entraining with another person’s speech rhythm: Evidence from healthy speakers and
individuals with Parkinson’s disease. Clinical Linguistics and Phonetics, 30(1), 68-85.
75
It is with great pleasure that we welcome you to Cologne to PaPE 2017!
The Phonetics and Phonology in Europe (PaPE) conference series is a forum that has the aim of
exploring disciplinary and interdisciplinary approaches to all areas of phonetics and phonology, with a
special focus on Laboratory Phonology. This includes both theoretical and applied research, and in
particular the relationship between the two. The series covers a wide variety of topics including tone
and intonation, phonological theory, audiovisual prosody and gesture, language development, linguistic
typology, language pathology, and language teaching. Methodologically, the conference also aims at
bridging the gap between the fields of phonetics and phonology and fields such as psycholinguistics,
neurolinguistics, and computational linguistics.
This is the second PaPE conference, following a highly successful first conference in Cambridge, UK,
in June 2015. Prior to 2015, a series of biennial PaPI (Phonetics and Phonology in Iberia) conferences
dates back to 2003. The broadening of the scope of the conference will hopefully lead to fruitful
exchange in Europe and beyond.
For PaPE 2017, over 200 submissions were received, with authors from 40 countries. Of these, 95 will
be presented at the conference.
Looking forward to your presentations and to stimulating discussions in and around the sessions.
Martine Grice
On behalf of the organising committee
Organising committee:
Stefan Baumann
Anna Bruggeman
Francesco Cangemi
Martine Grice
Local organisation:
Martina Krüger
Christine Riek
We are especially grateful to our sponsors, the Deutsche Forschungsgemeinschaft (DFG), Language
Science Press (LSP), the International Phonetic Association (IPA) and the Association for Laboratory
Phonology (ALP).
Association for Laboratory Phonology
Deutsche Forschungsgesellschaft
Language Science Press
International Phonetic Association
1