Cerebellar patients have intact feedback control that can be leveraged to improve reaching

Abstract

1It is thought that the brain does not simply react to sensory feedback, but rather uses an 2 internal model of the body to predict the consequences of motor commands before sensory 3 feedback arrives. Time-delayed sensory feedback can then be used to correct for the 4 unexpected-perturbations, motor noise, or a moving target. The cerebellum has been implicated 5 in this predictive control process. Here we show that the feedback gain in patients with cerebellar 6 ataxia matches that of healthy subjects, but that patients exhibit substantially more phase lag. 7 This difference is captured by a computational model incorporating a Smith predictor in healthy 8 subjects that is missing in patients, supporting the predictive role of the cerebellum in feedback 9 control. Lastly, we improve cerebellar patients' movement control by altering (phase advancing) 10 the visual feedback they receive from their own self movement in a simplified virtual reality 11 setup.12 16 slow movements accurately. However, feedback is time delayed, and thus it never represents the 17 current state of the body during movement. Because of this, it is thought that we depend on 18 internal models of the body that are built based on prior experience. These models can be rapidly 19 accessed and thus provide a fast internal prediction system to estimate how a movement will 20 unfold, enabling us to better understand where our limbs are at any given moment. This allows 21 us to make fast and accurate movements despite long-latency feedback. 22People with cerebellar damage show a characteristic pattern of incoordination during 24 movement that is referred to as ataxia. When reaching, they make curved movements that miss 25 intended targets and require multiple corrections. This pattern of over-and undershooting a 26 target (dysmetria) and oscillatory corrections (intention tremor) are hallmarks of cerebellar 27 ataxia. One hypothesis that might explain ataxia is that the predictive estimation and control 28 provided by cerebellar circuits is dysfunctional or lost (Wolpert, Miall, and Kawato 1998; R. 29 Chris Miall et al. 2007).30 31 Normally, the estimation of limb state (i.e., position and velocity) benefits from 32 integrating proprioceptive measurements with an internal predictive control model during a 33 movement (Paillard and Brouchon 1974; Adamovich et al. 1998; Fuentes and Bastian 2010). 34However, patients with cerebellar damage do not seem to receive this benefit (N. H. Bhanpuri, 35 Okamura, and Bastian 2013; Weeks, Therrien, and Bastian 2017). Worse, it is possible that their 36 predictive model actually conveys incorrect state information during active movements, which 37 could corrupt rather than enhance proprioceptive estimation of limb state. This difficulty of 38 predicting the future state of limbs during active movement leads to movements that are poorly 39 directed and scaled, requiring ongoing corrections to reach a goal location. 41Patients with cerebellar ataxia may rely more heavily on visual feedback to correct 42 dysmet...

De novolearning versus adaptation of continuous control in a manual tracking task

“…Our characterization of learning made use of frequency-based system identification, a powerful tool that has been previously used to study biological motor control such as insect flight ( Sponberg et al, 2015; Roth et al, 2016 ), electric fish refuge tracking ( Cowan and Fortune, 2007; Madhav et al, 2013 ), human posture ( Oie et al, 2002; Kiemel et al, 2006 ), and human reaching ( Zimmet et al, 2019 ). This approach has a number of practical advantages over other methods for studying motor control (e.g., point-to-point reaches)—including time-efficiency for data collection and the availability of a rich suite of tools in the frequency domain ( Schoukens et al, 2004 ).…”

“…Although the primary goal of our frequency-based analysis was to establish how participants mapped target motion into hand motion, system identification yields more detailed information than this; in principle, it provides complete knowledge of a linear system in that knowing how the system responds to sinusoidal input at different frequencies enables one to predict how the system will respond to arbitrary inputs. This data can be used to formally compare different possible control system architectures ( Zimmet et al, 2019 ) supporting learning, and we plan to explore this more detailed analysis in future work.…”

Transient deactivation of dorsal premotor cortex or parietal area 5 impairs feedback control of the limb in macaques

“…It is also interesting to note that reductions in parameters associated with the forward model led to large errors even for small reductions in gain and even oscillatory behavior for reductions at $50% (Figures 1C and S1A). Accumulating evidence supports a role of the cerebellum for forward models in motor control [74][75][76][77][78][79] . The forward model is essential to make a prediction of the state to overcome sensorimotor delays that can destabilize control 39,80 .…”

Cerebellar patients have intact feedback control that can be leveraged to improve reaching

“…All analysis was performed using custom scripts (Zimmet, Cao, Bastian, and Cowan 2020) in MATLAB (The Mathworks Inc., Natick, MA, USA). To obtain the steady-state frequency response of the subject, the first period (20 seconds) was discarded as transient, leaving four periods per trial.…”