Information Transfer Capacity of Articulators in American Sign Language

Abstract

The ability to convey information is a fundamental property of communicative signals. For sign languages, which are overtly produced with multiple, completely visible articulators, the question arises as to how the various channels co-ordinate and interact with each other. We analyze motion capture data of American Sign Language (ASL) narratives, and show that the capacity of information throughput, mathematically defined, is highest on the dominant hand (DH). We further demonstrate that information transfer capacity is also significant for the non-dominant hand (NDH), and the head channel too, as compared to control channels (ankles). We discuss both redundancy and independence in articulator motion in sign language, and argue that the NDH and the head articulators contribute to the overall information transfer capacity, indicating that they are neither completely redundant to, nor completely independent of, the DH.

Shared neural and cognitive mechanisms in action and language: The multiscale information transfer framework

“…The quantitative difference between action and the signal produced with communicative intent (i.e., language, either spoken or signed) is that the linguistic signal contains more information (defined in terms of entropy) across multiple timescales (Malaia, Borneman, & Wilbur, 2016;Singh & Theunissen, 2003). The most recent contribution to the existing evidence on this distinction comes from entropy measures in the visual domain (Borneman, Malaia, & Wilbur, 2018;Bosworth, Bartlett, & Dobkins, 2006;Malaia et al, 2016;Malaia, Borneman, & Wilbur, 2017). A quantitative comparison between the action and sign language signal can be formulated in terms of the amount of motion information in the visual signal, as measured, for example, by optical flow.…”

“…ASL, as any other language, is highly efficient in transferring information in time; there is less to comprehend in action than in language, despite the fact that they are in the same modality. ASL is known to operate on multiple timescales using hands, face, head, and posture as linguistic articulators across different temporal scales (Malaia et al, 2017). For example, the eyebrow furrow scopes over the interrogative clause, regardless of how many hand signs are in it; the syntactic structure is put together at a different temporal scale than the lexical word.…”

“…In mathematical terms, a power law is a relationship between two parameters of the system (e.g., the number of different lexical items in a language, and the relative frequency of their use), such that a relative change in one quantity results in a proportional relative change in the other quantity. The example of the relationship between a lexical item's ranking on the frequency list and its usage frequency was first formulated as Zipf's law, but the application of the analysis to corpora diachronically and cross-linguistically made it clear that the specific parameters of power-law-based models (such as the fractal complexity parameter, β) reflect not just the properties of the linguistic system at the lexical level, but can be used to track relative changes in syntax from synthetic (high use of morphological modification on lexical items) to analytic (reliance on function words), or to compare the amount of information from various sources in visual communication (Malaia et al, 2016(Malaia et al, , 2017.…”

The Impact of Transitional Movements and Non-Manual Markings on the Disambiguation of Locally Ambiguous Argument Structures in Austrian Sign Language (ÖGS)

“…More complex information, including, for example, spatial arrangement/orientation in addition to the physical properties of an object, can also be expressed this way. This complex simultaneity with each manual articulator (hand) is further enhanced by the availability of two hands, which can independently indicate distinct objects and can in their interaction indicate the spatial relationship or interaction between those objects (although they are not completely free to act independently; Malaia, Borneman & Wilbur, 2017). The pairing of manual articulators further enables the simultaneous expression of morphological information such as marking of number (Lepic, Börstell, Belsitzman & Sandler, 2016).…”

Current and future methodologies for quantitative analysis of information transfer in sign language and gesture data

“…to investigate the models of information transfer between complex systems using, for example, complex systems analysis and machine learning techniques (cf. Malaia et al, in press; Barbu et al 2014).…”