Background: The primary motor cortex (M1) is part of the neural substrates of skills learning but has finite processing resources. Typically, M1 representation of the trained body parts has been shown to enlarge at the expense of the representations of less trained body parts (e.g. 1). According to this competitive model of cortical resources, it might be possible to surmount training limits by “weakening” the adjacent fingers’ representations. In this study, we tested the hypothesis that the combination of training and concomitant immobilization of the adjacent fingers would effectively enhance neural changes and boost motor learning. Methods: Participants (planned n=60) had to train at a visuomotor tracking task that requires abduction-adduction movements of one finger for one week (3 x 10 min/day). They used either their index (Group 1) or their little finger (Group 2) and half of them had their three non-trained fingers immobilized (Groups 3 & 4). The two control groups consisted in wearing the same splints without training (Groups 5 & 6). We recorded their performances before and after the intervention and examined the neurobiological underpinnings of motor skill learning through local changes in corticospinal excitability (CSE) profiles of hand muscles using neuronavigated Transcranial Magnetic Stimulation (nTMS) and changes in connectivity patterns using functional Magnetic Resonance Imaging.Results: Our preliminary data (n = 45) show that immobilization-only deteriorates performance of the deprived finger but this “de-learning” can be prevented by training the adjacent finger. This was associated with preserved CSE profiles of the deprived muscle while immobilization-only induces a drastic down-regulation. Concomitant immobilization of the adjacent fingers did not facilitate tracking performance with the trained finger despite of a stronger enlargement of the trained muscle’s CSE profile and different patterns of neural connectivity changes affected by the training task. Conclusion: The data show that concomitant immobilization of the adjacent fingers does not boost motor learning, at least for continuous visuomotor tracking. Although it needs to be confirmed in larger trials, the results rather show that it might be possible to avoid a decay in performance after limb immobilization by concurrent training of the adjacent body part.