55To date there exists no reliable method to non-invasively upregulate or downregulate 56 the state of the resting motor system over a large dynamic range. Here we show that 57 an operant conditioning paradigm which provides neurofeedback of the size of motor 58 evoked potentials (MEPs) in response to transcranial magnetic stimulation (TMS), 59 enables participants to self-modulate their own brain state. Following training, 60 participants were able to robustly increase (by 83.8%) and decrease (by 30.6%) their 61 MEP amplitudes. This volitional up-versus downregulation of corticomotor 62 excitability caused an increase of late-cortical disinhibition (LCD), a read-out of 63 presynaptic GABA B disinhibition which was accompanied by an increase of gamma 64 and a decrease of alpha oscillations in the trained hemisphere. This approach paves 65 the way for future investigations into how altered brain state influences motor 66 neurophysiology and recovery of function in a neurorehabilitation context. 67 68 69 70 71 72Introduction 73 74Rhythmic oscillatory brain activity at rest is associated with high versus low 75 neuronal responsiveness, or 'excitability' of a region 1,2 . Measuring these momentary 76 fluctuations of neural activity via electro-or magnetoencephalography (EEG/MEG) 77 over human primary motor cortex (M1), it has been demonstrated that frequency, 78 amplitude and phase of the ongoing oscillation cycle systematically modulate 79 responses evoked by transcranial magnetic stimulation (TMS) 3-5 6-8 . In particular, it 80 has been shown that corticomotor excitability is significantly higher when the power 81 (amplitude) of sensorimotor rhythms in the alpha band (8-14 Hz, also called the 'mu'-82 rhythm), or beta band (15-30 Hz) are low, or when M1 is stimulated during the 83 trough of the oscillatory cycle of these rhythms 9 . This concept has inspired 84 neurofeedback interventions whereby, for example, stroke patients learn to 85 volitionally desynchronize sensorimotor rhythms with the goal of bringing the 86 sensorimotor system into a more excitable state as a precursor for enhanced neural 87 plasticity and accelerated recovery 10-12 . 88 89 Previous research has focussed on interactions between corticomotor 90 excitability and cortical dynamics at rest, but much less is known about whether it is 91 possible to voluntarily control the excitability of sensorimotor circuits while keeping 92 motor output and sensory feedback constant. In the case of stroke rehabilitation, this 93 mechanism may become particularly relevant as patients are unable to move or 94 receive sensory feedback from the paretic limb. Therefore, interventions that 95 optimally harness the residual ability to voluntarily and endogenously activate 96 relevant brain circuits in the days and weeks early after the incident, may provide the 97 crucial innervation necessary to promote re-wiring for functional recovery 13 . 98
99It is well known that primates 14,15 , and humans 10,eg. 16-19 can gain volitional 100 control of neural activity by...