Measuring Synchrony in Task-Based Dialogues

Abstract

In many contexts from casual everyday conversations to formal discussions, people tend to repeat their interlocutors, and themselves. This phenomenon not only yields random repetitions one might expect from a natural Zipfian distribution of linguistic forms, but also projects underlying discourse mechanisms and rhythms that researchers have suggested establishes conversational involvement and may support communicative progress towards mutual understanding. In this paper, advances in an automated method for assessing interlocutor synchrony in taskbased Human-to-Human interactions are reported. The method focuses on dialogue structure, rather than temporal distance, measuring repetition between speakers and their interlocutors last n-turns (n = 1, however far back in the conversation that might have been) rather than utterances during a prior window fixed by duration. The significance of distinct linguistic levels of repetition are assessed by observing contrasts between actual and randomized dialogues, in order to provide a quantifying measure of communicative success. Definite patterns of repetitions where identified, notably in contrasting the role of participants (as information giver or follower). The extent to which those interacted sometime surprisingly with gender, eyecontact and familiarity is the principal contribution of this work.

Interaction Analysis and Cognitive Infocommunications

“…Intersections of topic areas are also addressed in the coginfocom literature. For instance, language use in situations that demand communication of spatial directions has been studied [32], [33], [34]. Some coginfocom researchers have studied linguistic representation of reasoning [35].…”

Linguistic and Behaviour Interaction Analysis within Cognitive Infocommunications

“…In the manner that we have addressed emotion, gesture, language and thought as coginfocom technologies, others have explored other representational systems humans have adopted for reasoning, in particular, maps, for spatial reasoning [11], [12]. Intersections are addressed as well, for example, language use in scenarios requiring communication about spatial directions [13], [14], [15]. Some have addressed linguistic representation of reasoning [16].…”

Linguistic repetitions, task-based experience and a proxy measure of mutual understanding

“…2) Extended Method: The method was extended in [19] with the aim of exploring more levels of linguistic description than the tokens as transcribed from the dialogue (lemmas and part-of-speech labels were considered alongside tokens, and sequences thereof) and doing so in the context of the HCRC Map Task data in order to relate repetition effects to task success (and other variables controlled in the maptask experiment [15]). It was extended for two reasons: first to observe the scope in which different linguistic levels of repetitions provide information reliably as indicators of synchrony within the frame of the base method, and second, to which extent success in communication is associated with repetitions.…”

“…We labelled the HRCR Map Task transcripts with the default version of the TreeTagger as trained for English [24]. As we observed in [19], although the method is not designed directly to look at syntactic repetitions, the POS labelling allows us to observe two different form of repetitions; lexical categories for N1 and structural repetitions for N2+ in combination with Level 4 (POS).…”

“…Of course, one may also consider the speaker's and other interlocutors' prior n-turns, and boundary conditions can be identified for which the methods are the same, but in general, they are different. The structural construal of repetitions has been explored before [18], [10], and scrutiny of the HCRC Map-Task data has addressed levels of linguistic representation at which repetition may occur [19]. We adopt this structural approach to analysing dialogues, using memory registers for each interlocutors' last contribution, and comparing the content of the speaker's current utterance to the contents of each register (rather than within a specific temporal distance).…”