A Systematic Investigation of Gesture Kinematics in Evolving Manual Languages in the Lab

How accurately can we estimate spontaneous body kinematics from video recordings? Effect of movement amplitude on OpenPose accuracy

Estimating how the human body moves in space and time-body kinematics-has important applications for industry, healthcare, and several research fields. Gold-standard methodologies capturing body kinematics are expensive and impractical for naturalistic recordings as they rely on infrared-reflective wearables and bulky instrumentation. To overcome these limitations, several algorithms have been developed to extract body kinematics from plain video recordings. This comes with a drop in accuracy, which however has not been clearly quantified. To fill this knowledge gap, we analysed a dataset comprising 46 human participants exhibiting spontaneous movements of varying amplitude. Body kinematics were estimated using OpenPose (video-based) and Vicon (infrared-based) motion capture systems simultaneously. OpenPose accuracy was assessed using Vicon estimates as ground truth. We report that OpenPose accuracy is overall moderate and varies substantially across participants and body parts. This is explained by variability in movement amplitude. OpenPose estimates are weak for low-amplitude movements. Conversely, large-amplitude movements (i.e., > ~ 10 cm) yield highly accurate estimates. The relationship between accuracy and movement amplitude is not linear (but mostly exponential or power) and relatively robust to camera-body distance. Together, these results dissect the limits of video-based motion capture and provide useful guidelines for future studies.

Extending the Architecture of Language From a Multimodal Perspective

Language is inherently multimodal. In spoken languages, combined spoken and visual signals (e.g., co-speech gestures) are an integral part of linguistic structure and language representation. This requires an extension of the parallel architecture, which needs to include the visual signals concomitant to speech. We present the evidence for the multimodality of language. In addition, we propose that distributional semantics might provide a format for integrating speech and co-speech gestures in a common semantic representation.

The CABB dataset: A multimodal corpus of communicative interactions for behavioural and neural analyses

We present a dataset of behavioural and fMRI observations acquired in the context of humans involved in multimodal referential communication. The dataset contains audio/video and motion-tracking recordings of face-to-face, task-based communicative interactions in Dutch, as well as behavioural and neural correlates of participants' representations of dialogue referents. Seventy-one pairs of unacquainted participants performed two interleaved interactional tasks in which they described and located 16 novel geometrical objects (i.e., Fribbles) yielding spontaneous interactions of about one hour. We share high-quality video (from three cameras), audio (from head-mounted microphones), and motion-tracking (Kinect) data, as well as speech transcripts of the interactions. Before and after engaging in the face-to-face communicative interactions, participants' individual representations of the 16 Fribbles were estimated. Behaviourally, participants provided a written description (one to three words) for each Fribble and positioned them along 29 independent conceptual dimensions (e.g., rounded, human, audible). Neurally, fMRI signal evoked by each Fribble was measured during a one-back working-memory task. To enable functional hyperalignment across participants, the dataset also includes fMRI measurements obtained during visual presentation of eight animated movies (35 min total). We present analyses for the various types of data demonstrating their quality and consistency with earlier research. Besides high-resolution multimodal interactional data, this dataset includes different correlates of communicative referents, obtained before and after face-to-face dialogue, allowing for novel investigations into the relation between communicative behaviours and the representational space shared by communicators. This unique combination of data can be used for research in neuroscience, psychology, linguistics, and beyond.

The multimodal nature of communicative efficiency in social interaction

How does communicative efficiency shape language use? We approach this question by studying it at the level of the dyad, and in terms of multimodal utterances. We investigate whether and how people minimize their joint speech and gesture efforts in face-to-face interactions, using linguistic and kinematic analyses. We zoom in on other-initiated repair-a conversational microcosm where people coordinate their utterances to solve problems with perceiving or understanding. We find that efforts in the spoken and gestural modalities are wielded in parallel across repair turns of different types, and that people repair conversational problems in the most cost-efficient way possible, minimizing the joint multimodal effort for the dyad as a whole. These results are in line with the principle of least collaborative effort in speech and with the reduction of joint costs in non-linguistic joint actions. The results extend our understanding of those coefficiency principles by revealing that they pertain to multimodal utterance design.

A multi-scale investigation of the human communication system's response to visual disruption

In human communication, when the speech is disrupted, the visual channel (e.g. manual gestures) can compensate to ensure successful communication. Whether speech also compensates when the visual channel is disrupted is an open question, and one that significantly bears on the status of the gestural modality. We test whether gesture and speech are dynamically co-adapted to meet communicative needs. To this end, we parametrically reduce visibility during casual conversational interaction and measure the effects on speakers' communicative behaviour using motion tracking and manual annotation for kinematic and acoustic analyses. We found that visual signalling effort was flexibly adapted in response to a decrease in visual quality (especially motion energy, gesture rate, size, velocity and hold-time). Interestingly, speech was also affected: speech intensity increased in response to reduced visual quality (particularly in speech-gesture utterances, but independently of kinematics). Our findings highlight that multi-modal communicative behaviours are flexibly adapted at multiple scales of measurement and question the notion that gesture plays an inferior role to speech.