In this study we investigated sensory feedback driven modulation of target force tremor amplitude in p-ET and HC.
In summary, we found that target force tremor amplitude is modulated by visual and auditory sensory feedback scaling in a comparable measure in p-ET. During the high visual, auditory or combined audio-visual feedback tasks the tremor amplitude was significantly increased. Augmented sensory feedback coincided with an increased pupil diameter in p-ET, but not in HC. Combined audio-visual feedback evoked the largest increase of tremor amplitude and pupil diameter in p-ET and additionally, a significant increase of tremor force in HC.
While it is well described, that visual feedback modulates action tremor amplitude in different underlying disease conditions like multiple sclerosis, ET and dystonic tremor 8–10,15, our study is the first to show that the amplitude of target force tremor in ET is modulated by a different quality of sensory feedback (i.e. auditory) in a comparable scale.
The increase of the tremor amplitude during the auditory-only condition cannot be explained by an increased error since the MF, RMSE and 0–3 Hz force power as markers for non-tremulous movements did not differ between the conditions or groups.
Our findings raise the question, whether there is a common underlying mechanism for sensory feedback induced tremor modulation in the context of different sensory qualities.
A recent functional MRI study found -apart from the well-known cerebello-thalamo-motor cortical tremor circuit- a widespread visually sensitive network including key regions in the visual cortex and parietal lobule associated with alterations of essential tremor amplitude during visual feedback manipulation in a grip force task 12. Interestingly, by the same group visual feedback-induced tremor exacerbation in patients with dystonic tremor was found as well, but in this patient group tremor amplitude modulation was not coupled to an altered BOLD signal of visual cortex regions 11. Taken together with our finding that force tremor amplitude is comparably modulated by auditory feedback as well, this underlines the role of a common underlying mechanism for sensory feedback induced tremor modulation apart from the visual network.
Our finding that combined audio-visual feedback evoked the largest increase of tremor amplitude in p-ET but also a significant increase of tremor in HC, underlines that the magnitude of sensory feedback per se correlates with a tremorgenic effect.
We hypothesized, that an increased arousal has an effect on the intensification of the tremor amplitude.
Recently, a modulatory role of cognitive effort during a serial seven task, as measured by a coincident pupillary dilation, onto the rest tremor network of Parkinson´s disease (PD) was shown 16. This effect was most likely exerted by direct bottom-up noradrenergic influences onto the thalamus and indirectly by top-down cognitive influences onto the cerebello-thalamo-cortical circuit. Since the thalamus is a key node not only within the PD resting tremor network but also the action tremor network in ET as well 7, an amplification of action tremor by ascending noradrenergic systems seems possible.
Enhanced feedback of any sensory quality during target driven physical tasks might increase the arousal/perceived effort level and thereby activate the ascending noradrenergic system, with the locus coeruleus (LC) as main effector 17. Recent neuroimaging studies have confirmed a close relationship between the LC and bilateral thalamus and the cerebellum, both key regions within the action tremor network 18. Therefore, cognitive arousal/perceived effort during motor tasks, induced by enhanced sensory feedback of any quality, might activate the LC-noradrenergic system and thereby mediate an amplification of action tremor amplitude via thalamic and cerebellar projections of the LC.
Therefore, in our experiment, pupil diameter was measured as a marker for cognitive arousal and an increase of pupil size during the enhanced auditory and audio-visual feedback trials was found. Only during the enhanced visual-only feedback there was no significant pupil dilation (although a non-significant trend), which is most likely explained by the changes in external illumination during the visual-only feedback, triggering a pupil constriction and hampering the pupil dilation. Since external illumination remained constant during the auditory feedback trials, pupil dilation occurred independently of external visual input. It´s rather probable, that the pupil dilation reflects an increased arousal during the large-scale feedback trials. Pupil size coincides with cognitive arousal due to activation of the sympathic system and the task evoked pupillary response is known to reflect the mental effort to perform the task 19, which was also shown in p-ET by our group 20. Apart from mental effort, pupil diameter also increases during physical effort, thereby reflecting not only the actual intensity of the physical activity but also the individual perception of the effort 21. In summary, pupil size mirrors the level of effort, which is invested in a task, irrespective of whether it is physical or mental. Therefore, we hypothesize that tremor p-ET perceived a higher effort during the large-scale feedback tasks, as reflected by the larger pupil diameter. Thus, the effort itself could exert a modulatory role on target force tremor amplitude.
Another explanation for sensory feedback dependent tremor modulation could encompass the interaction between somatosensory cortex (S1) and the primary motor cortex (M1). M1 plays a crucial role as a feedback controller for motor control, performing dynamic updates of internal motor commands, which receive input from the somatosensory cortex (S1). However, when sensory feedback is manipulated, such as in our paradigm where visual feedback is altered and does not match the tactile feedback, it might lead to incorrect updating in M1 22,23.
This idea is supported by the fact, that S1 and the cerebellum are closely interconnected and work together during movement control (Diedrichsen et al., 2005). Dysfunction of this interaction seems to contribute to the development of action tremor 24–27. Therefore, understanding the complex interactions between M1, S1, and the cerebellum seems essential for understanding how action tremor emerges.
Our data of the pupillometry is intended as a primer of the LC activity (Aston-Jones & Cohen, 2005). Studies have shown that the LC projects into the thalamus and basal ganglia and acts as modulator of these regions. Both, the basal ganglia, and thalamus are involved within tremor generation28,29. In our task, two mechanisms might contribute to the fact that p-ET show a higher tremor force in harder task conditions, feedback modality independent. First, a bottom-up process triggered by the LC activity in a higher arousal state mutes down inhibition on subcortical tremor-generating structures. This is partially supported by our pupil data. Secondly, the cerebellum and sensoricortical structures integrate different sensory information (visual, auditory, and somatosensory) which are supposed to work as an efference copy for the feedback control of M1.
Limitations
Our study has several limitations. The main limitation is that, by our experiment setup, we cannot finally prove that the altered arousal (mirrored by pupil dilation) is directly caused by the enhanced feedback. The enhanced arousal could also be just a secondary effect of the increased difficulty to perform the task with increased tremor. However, in this case we would expect a correlation of the pupil dilation with the PSD in the tremor relevant frequency spectrum (4–12 Hz) independently of the feedback condition or with the individual TETRAS score, but both were not given. Therefore, the increase of arousal seems to be caused by the enhanced sensory feedback itself and is not a secondary effect of the tremor increase.
Another limitation of the auditory feedback paradigm is that hitting the target tone might be easier (and therefore cause less arousal) for participants who are familiar with making music or singing. At least we excluded a manifest hypoacusis in all participants by a hearing test.