Successful integration of augmentative technology in everyday life relies, amongst other things, on individuals’ capacity to learn to successfully operate an extra limb. In a previous study (Kieliba, 2021), participants were able to use an extra robotic thumb – the Third Thumb (TT, Dani Clode Design) – and make significant progress over the course of 5 days of in lab training on various motor tasks. Our current aim is to expand on Kieliba’s findings and explore the motor learning curves over a longer period of time. We’ll include a wider variety of tasks in order to target an array of motor skills. We’ll also investigate if participants are able to meaningfully make progress on their training with a majority of unsupervised training.

We aim to recruit 30 participants (currently n=22) to learn to use the TT and train for 7 consecutive days. Days 1 and 7 of training occur in lab, supervised by an experimenter, whilst Days 2 to 6 take place remotely with some limited supervision via Zoom (version 5.15.7). Training lasts between 2 to 3h a day, with the possibility of doing extra hours. Participants enter their own scores via interactive videos hosted on a learning platform. The at-home training was designed to explore several facets of motor learning with augmentation devices and the majority of tasks can be broadly divided in 4 categories: collaboration, shared supervision, expansion, and individuation. Participant are tested on their skills in-lab before and after training, in a pre- and post-behavioural session.

Participants’ scores improved for all tasks between Day 1 and Day 7, including every day in between while training was unsupervised. This suggests they were able to successfully improve their motor skill using the Third Thumb in various types of scenarios. A plateau effect can be observed in some tasks towards Days 6 and 7. Score improvement between first and last day of training is positively correlated to in-lab pre and post behavioural measures improvement. Additionally, in-lab pre and post behavioural measures were also positively correlated with extra hours of unsupervised training. This seems to show that training with no direct supervision translates into their performance in-lab, proportionally with the number of hours spent training.

So far, this study demonstrates participants’ performance on manual tasks using an extra robotic thumb show progress with only 7 days of training, and little supervision. This is an important insight for the field of augmentative technologies, as it shows that learning to operate an extra limb is within reach of most individuals with regular training. Future studies should investigate how extra robotic body parts are incorporated into the motor synergies and their impact on individuals’ bodies’ sensory model after repeated use. Indeed, it is essential to explore the ways in which augmentative technologies can interact with individuals’ already established motor resources in order to develop products that can be implemented in harmony with them.