Transcranial magnetic stimulation (TMS) is a non-invasive and painless way of stimulating the brain in conscious healthy individuals and is a common tool in many labs worldwide. Repetitive TMS (rTMS) in which a large number of stimuli are applied within a few minutes produces after-effects on the function and excitability of the stimulated site that outlast the period of stimulation for several minutes or hours. Depending on the area stimulated, the after-effects can influence performance of cognitive tasks or learning. Given the success of these methods in healthy individuals, there has been enormous interest in applying them therapeutically, for example in rehabilitation after stroke. Unfortunately the results are mixed and evidence for success limited. One reason for this is that the after-effects vary considerably both within and between individuals. For rTMS of motor cortex, a common target for therapy after stroke, most protocols only produce the "expected" effects in 45-60% participants. The variation in effect occurs because TMS activates a mixture of neurones within a cortical area, i.e. different populations of excitatory and inhibitory neurones with different functions. It is likely that TMS activates a variable proportion of each type in different people, giving rise to the inter-individual differences in effect. We have already shown that it is possible to increase the selectivity of stimulation by using a controllable TMS device in which we can modify the shape and directionality of the pulse. Pilot data strongly suggests that this leads to much more reliable outcomes. The first aim of this project is to confirm this is correct in a large group of healthy individuals by measuring the after-effects on motor cortical excitability of a popular form of rTMS known as theta burst stimulation (TBS). Although measures of motor cortex excitability are the standard way of comparing effects of rTMS, they are not in practice the most useful because rTMS effects on cortical excitability may not correlate with rTMS effects on behaviour. The second aim of these experiments is therefore to show that better controlled TBS protocols also produce effects on movement. Given the importance of motor learning in rehabilitation, we have chosen to test the effects of TBS on motor learning. Furthermore, since our previous work has suggested that different types of motor learning involve different sets of neurones in motor cortex, we will examine two forms of learning: adaptation learning and model-free learning. The former involves adapting a movement that has already been acquired (e.g. adapting to a misalignment between the actual and visually perceived position of a cursor when moving a computer mouse) whereas the latter involves exploring the best combination of muscle activation to achieve a new aim (such as learning to maximise the acceleration of the thumb in a novel direction). We expect different TBS protocols to improve each type of learning. The implication is that future therapeutic applications may need to adapt the TBS protocols to the deficits of individual patients. The third aim of the project is to confirm that the same principles apply in chronic stroke survivors. If controllable TBS is to be a useful therapy, then it is vital to confirm that the conclusions from studies based in healthy volunteers are also true in the damaged brain after stroke. We will test the effects of the optimal forms of TBS on the two types of motor learning in stroke.