Characteristics of bi-directional unimanual and bimanual drawing movements: The application of the Delta-Lognormal models and Sigma-Lognormal model

Abstract The use of large digitizers, allowing users to interact using both hands simultaneously, are getting more and more popular in applications requiring human computer interaction. In most of these applications, gesture commands are used and these commands to activate specific actions can provide also information about the neuromuscular system of the user. As shown for single hand interactions, understanding of the neuromotor coordination processes using both hands could improve interface design. In the present paper, we address this topic with a first study aimed to characterize the bimanual and unimanual coordination. We compare the kinematic properties of movements performed unimanually and bimanually. Rapidly drawn lines are assessed using kinematic features in addition to the parameters estimated by two models, i.e., the Delta-Lognormal model and Sigma-Lognormal model. Sixteen right-handed participants were instructed to conduct 120 rapid drawing actions with the dominant hand, non-dominant hand, and both hands simultaneously. Kinematic variables (i.e. reaction time (RT) and movement time (MT)) together with parameters of the Delta-Lognormal model and Sigma-Lognormal model were extracted from the movements. The latter parameters estimate the agonist and antagonist synergies that contribute to the overall movement production of the rapid drawn lines. A 2 × 2 repeated measures ANOVA was applied to all dependent variables with manual condition (uni- vs bimanual) and hand used (dominant vs non-dominant) as independent factors. Parameters of the Delta-Lognormal model and Sigma-Lognormal model revealed that the prolonged preparation time of the bimanual task had both a central and peripheral origin, i.e., it originated at the CNS and neuromuscular levels. Furthermore, findings from the Sigma-lognormal model suggested that the dominant and non-dominant hand contribute differently to the manual conditions, which may lead to new heuristics for implementing command systems for computer interaction systems requiring bimanual manipulation as well as designing automatic systems for rehabilitative training.