We use motion capture data to investigate and statistically compare the kinematic fluency patterns within and in between the groups of four L1 and four L2 signers of Finnish Sign Language.
First and second language fluency in sign language is not only an understudied but also a complex topic. Part of the complexity is caused by the heterogeneity of the L1 group, another part by the fact that L2 users often have to learn sign language through a different modality. For this reason, Kanto and Haapanen (2019) emphasize that, in the field of sign language, fluency comparisons between L1 and L2 signers have to be made with caution and by taking into account, for example, (dis)fluency patterns of both groups.
An obvious feature contributing to the fluency in sign language is the smoothness of the articulatory movements. Kinematically, smoothness of the movement can be estimated, for example, with the help of velocity and acceleration information. When the velocity of the movement is high and the acceleration low, the movement is regarded as smooth. On the other hand, when the velocity of the movement is low and the acceleration high, the movement is regarded as jerky. (Burger et al. 2013.) In practice, estimating fluency this way kinematically corresponds to calculating the degree of circularity of the movements computationally (Burger & Toiviainen 2013).
In this study we use motion capture data to investigate and statistically compare the kinematic fluency patterns within and in between the groups of four L1 and four L2 signers of Finnish Sign Language. In the recording session, each signer produced a set of five sentences while wearing 25 reflective markers whose three-dimensional locations were tracked with an eight-camera optical motion capture system recording at the speed of 200 frames per second. In our presentation, we will show and discuss the results of our analysis focusing on both global (i.e. all markers) and local (i.e. selected manual and nonmanual markers) calculations of kinematic fluency.
