When the Words are Not Everything: The Use of Laughter, Fillers, Back-Channel, Silence, and Overlapping Speech in Phone Calls

Abstract

This article presents an observational study on how some common conversational cueslaughter, fillers, back-channel, silence, and overlapping speech-are used during mobile phone conversations. The observations are performed over the SSPNet Mobile Corpus, a collection of 60 calls between pairs of unacquainted individuals (120 subjects for roughly 12 h of material in total). The results show that the temporal distribution of the social signals above is not uniform, but it rather reflects the social meaning they carry and convey. In particular, the results show significant use differences depending on factors such as gender, role (caller or receiver), topic, mode of interaction (agreement or disagreement), personality traits, and conflict handling style.

What an “Ehm” Leaks About You: Mapping Fillers into Personality Traits with Quantum Evolutionary Feature Selection Algorithms

“…In other words, the fillers are those vocalizations that speakers utter when they want to hold the floor, but they do not know what to say next. Fillers occur frequently during spontaneous conversations and, in particular, the analysis presented in [7] shows that the speakers involved in the experiments of this work utter, on average, one filler every 10.9 seconds. To the best of our knowledge, no theory explains why fillers should carry personality-relevant information and, according to the literature, "[...] little research examines the correlation between self-report personality traits and filler words" [8].…”

“…The fillers have been extracted from 60 dyadic conversations between unacquainted individuals (see [4], [7] for a full description of the data) for a total of 120 participants (63 female and 57 male), all native English speakers of British nationality. The conversations are based on the Winter Survival Task (WST) [59]: The participants are said to be part of a rescue team that will assist the survivors of a plane crash in a polar area.…”

The discourse value of social signals at topic change moments

“…Results show that pauses and overlaps on their own are good markers of the topic structure of meetings conversation, reaching performance comparable with lexical based methods. Vinciarelli et al [15] also analyze laughter and topics, finding that laughs decreases when participant are concentrated on a topic rather than in social chitchats. In previous work, we have investigated the discourse function of laughter in isolation by looking at its distribution at a macro level throughout a topic [16]; here we investigate laughter and other signals at a micro level, providing a better insight into influence of the social signals on topic change.…”

“…Some [15] have investigated the relation among paralinguistic events and contextual factors, such as topic changes, gender, speaker role, mode of interaction. In previous work [16,17,18] we have analyzed the relation between laughter and topic changes, showing that the distribution of laughter is not random, nor uniform, and does not exclusively depend on the distribution of mirthful events, but rather correlates with discourse events.…”

Body Language Without a Body

“…Early works addressing explicitly such a question show that the typing pa erns that people manifest in online textual chats are su ciently speci c of an individual to allow person recognition [1], that people speaking on the phone invest up to one quarter of their conversation time in nonverbal communication [6] and that the images that people tag as favourite on Flickr convey personality impressions [5,9]. e works above -and the others that the literature presentsprovide initial indications, but several questions remain to be addressed.…”