Goal-dependent tuning of muscle spindle receptors during movement preparation

See allHide authors and affiliations

Science Advances  24 Feb 2021:
Vol. 7, no. 9, eabe0401
DOI: 10.1126/sciadv.abe0401
  • Fig. 1 First experimental setup and representative single-trial data.

    (A) The general setup of experiment 1. Participants performed the classic instructed-delay reaching task using their right hand. From an initial semipronated position, wrist flexion-extension moved a visual cursor in the horizontal dimension, and wrist ulna-radial deviation moved the cursor in the vertical dimension. The participant’s task was to move the cursor to reach one of eight peripheral visual targets. On each trial, a target would suddenly turn into a red filled circle, representing the target “cue,” and participants were instructed to move to this target as soon as the go cue appeared (target turned into a green outline). The targets/trials were presented in a block-randomized manner; hence, there was no systematic difference in movement history across a particular group of targets. (B) Representative data from a single trial where reaching the target required ulna deviation of the wrist. Muscle length and velocity estimates pertain to the spindle-bearing muscle, which in this case is the radial wrist extensor (RWE; i.e., extensor carpi radialis). Also shown is surface EMG from the ulna wrist extensor muscle (UWE; i.e., extensor carpi ulnaris), which mostly powered the reaching movement. Despite no overt changes in kinematics or EMG during the preparatory period (gray background), primary spindle afferent (Ia) firing rate decreased, particularly at the latter half of this period. (C) The same neuron as in (B), but here, the visual target was in the opposite direction, requiring radial deviation at the wrist and therefore shortening of the radial wrist extensor. No decrease in firing rate was observed during the preparatory period. Throughout, dashed gray lines represent zero values. a.u., arbitrary units.

  • Fig. 2 Goal-dependent tuning of muscle spindle receptors during movement preparation.

    (A) The visual targets were categorized on the basis of whether reaching them required stretching or shortening of the spindle-bearing muscle. According to published physiological models for each muscle (see Materials and Methods), six targets represented clear and substantial change in muscle length, whereas two “intermediate” targets (circle outlines) represented little or no muscle stretch or shortening. (B) Top: Mean stretch velocity of the recorded spindle-bearing muscles, essentially indicating that no overt movement occurred in the preparatory period (see, e.g., velocity scales in Fig. 1, B and C, and fig. S1). Bottom: Mean change in primary spindle afferent (Ia) firing rates (eight afferents recorded from six individuals). The traces are aligned to onset of the target cue (time “0”). Purple and blue traces represent targets associated with stretch and shortening of the spindle-bearing muscle, respectively. Shading represents ±1 SEM. (C) Average Ia firing rates in the three epochs (1 to 3) as shown in (B). Thin gray lines represent individual Ia afferents from wrist extensor muscles, and thin black lines represent Ia afferents from digit extensors. The shaded bars represent 95% CIs, and asterisk represents P < 0.05 following a paired t test. The same color scheme is used throughout. In the absence of changes in the muscle’s mechanical state, goal-dependent decreases in tonic Ia firing rate indicate a goal-dependent change in the fusimotor drive to spindles; such fusimotor supply may possibly have a stronger effect on the spindles’ sensitivity to dynamic muscle stretch (i.e., gain). n.s., not significant.

  • Fig. 3 Spindle Ia firing rates at late movement preparation predict performance during reaching.

    Throughout, each data point represents the average (median) value of a single participant/afferent across trials where reaching the target required stretch of the spindle-bearing muscle. The left column of panels represents all Ia afferents, including those originating from digit extensor muscles (black dots), and the right pertains to Ia from wrist muscles (gray dots). (A) Horizontal axes represent firing rates during the late preparation epoch (epoch 3 as defined in Fig. 2B), and vertical axes represent reaction time, i.e., the time between onset of the go cue and onset of the reaching movement. (B) Left: Vertical axes represent time between onset of reaching and the point of initial peak velocity during the reaching movement. With the exception of one afferent (black star), there was a strong positive relationship between Ia firing during preparation and time to peak velocity. Right: For the subset of muscles engaged in powering hand movement in the current task, movement performance was well described by the same relationship (i.e., 3-ms delay in attaining peak velocity for every additional spike per second). The relationship between spindle Ia responses at late preparation and subsequent reaching performance can be understood in terms of the spindle’s role in negative feedback circuits (i.e., stretch reflexes).

  • Fig. 4 The second experimental setup.

    (A) In experiment 2, participants held the graspable end of a robotic manipulandum. Vision was directed at a one-way mirror, on which the contents of a monitor were projected. Hand position was represented by a visual cursor. Although not shown here, the right forearm rested on an airsled, and the hand was immobile around the wrist (see Materials and Methods for more details). (B) Timeline of experimental manipulations. Each trial began by slowly loading the hand to 4 N in the upper left direction (i.e., +Y direction) or lower right direction (−Y direction), or there was no load (“null” load). The participants had to maintain the hand immobile at origin despite any loading. One of two visual targets (+Y or −Y direction) was then suddenly cued by turning red, and this state lasted for a relatively short delay (250 ms) or long delay (750 or 1250 ms). These preparatory delays correspond to the middle of epochs 1 to 3 (Fig. 2, B and C). At the end of the delay, the hand was rapidly perturbed toward or in the opposite direction of the cued target. The perturbation lasted for 150 ms; at its end, the go signal was given (cued target turned green), and movement to the target had to be actively completed. Cursor position was frozen during the perturbation. Trials were block-randomized; hence, perturbation direction was unpredictable even after experiencing a particular load, cue, and delay.

  • Fig. 5 Representative data from a single participant in experiment 2.

    Relevant median signals from a single participant when perturbations in the −Y direction stretched the pectoralis muscle following a 750-ms preparatory delay. The above occurred after first applying a load in the direction of pectoralis shortening (A), when there was no external load (B), or after first applying a load in the direction of pectoralis stretch, promoting increased pectoralis activity for maintaining the start position (C). Throughout, purple traces represent trials where reaching the cued target required pectoralis stretch, and blue traces represent trials where the cued target required pectoralis shortening. Deviations of the hand along the x axis were negligible during the perturbation and hence are not plotted in the current figure for clarity. Data are aligned to the onset of the position-controlled haptic displacement (time 0), defined as the point where movement speed reached 5% of initial peak value.

  • Fig. 6 The goal- and delay-dependent modulation of stretch reflex gains is congruent with the preparatory tuning profile of muscle spindles.

    (A to C) Mean hand position (posn.) and mean rectified pectoralis EMG activity across participants (N = 14) when an external (pre-)load was first applied in the direction of pectoralis shortening (A), when there was no external load (B) (but note increased EMG levels before time 0 due to co-contraction), and when an external load was applied in the direction of pectoralis stretch (C). Shading represents ±1 SEM. Data are aligned to the onset of the haptic perturbation (time 0). As the schematic on the far left indicates, the data represent trials where the preparatory delay was relatively long and the subsequent perturbation stretched the pectoralis. SLR denotes the epoch associated with the spinal stretch reflex and LLR the epoch associated with the long-latency stretch reflex or R3 (for LLR analyses, see Results and Fig. 7). Kinematic data pertaining to the blue condition are also plotted but are obscured. (D) Difference in mean pectoralis EMG activity (purple minus blue) in the spinal SLR epoch, corresponding to the data shown in (A) to (C). Dots represent individual participants, and thick vertical lines represent 95% CIs. (E to H) As top row of panels but representing trials where the preparatory delay was relatively short (0.25 s).

  • Fig. 7 LLR gains reflect the stronger goal-dependent suppression of spindle signals observed at longer preparatory delays.

    Goal-dependent difference in EMG responses of all recorded shoulder muscles at the LLR epoch (as indicated in Fig. 6A), with regard to the relatively short (250-ms) and long (≥750-ms) preparatory delays used in experiment 2. More negative values indicate stronger goal-appropriate behavior (i.e., relative suppression of stretch reflex gains for muscles that must stretch when reaching the cued target). Throughout, each data point represents the average value of a different participant (N = 14), and thick vertical lines represent 95% CIs. Asterisks indicate P values following a within-measures t test, with double asterisks indicating P < 0.01 and single asterisk indicating P < 0.05. These results demonstrate a weaker goal-dependent modulation of LLRs when the preparatory delay is short, regardless if contrasted across all load conditions (A), or only for the cases where the muscle was externally loaded, i.e., the load was applied in the direction of muscle stretch (B). The short delay here was 250 ms, which is substantially longer than the previously reported minimum delay for inducing full expression of goal-dependent LLR responses following perturbations of the upper limb (i.e., 100 to 150 ms). In contrast, the effect of delay length on LLR gains is generally congruent with the temporal evolution of spindle tuning (Fig. 2, B and C).

  • Fig. 8 LLR gains of biceps and triceps are also suppressed as a function delay when a larger workspace is involved.

    As in Fig. 7, but here, the z-normalized EMG data originate from experiment 3, where six targets were used (i.e., three axes of motion: vertical, horizontal, and diagonal). The data are collapsed across all load conditions. More negative values indicate stronger goal-appropriate behavior (i.e., relative suppression of stretch reflex gains for muscles that must stretch when reaching the target). Throughout, each data point represents a different participant, and thick vertical lines represent 95% CIs. P values resulted from within-measures t tests. As the case in experiment 2, the LLR responses of biceps and triceps were not significantly different as a function of delay length when preparing to reach targets along the diagonal axis (left column; only axis used in experiment 2). However, such effects are observed for the biceps brachii and triceps lateralis when preparing to act along the horizontal axis (middle column) and vertical axis (right column), respectively. This suggests that the larger workspace used in experiment 3 (versus 2) induced goal-dependent proprioceptive control of a larger group of muscles, but this control occurred selectively across the task’s dimensions.

Supplementary Materials

Stay Connected to Science Advances

Navigate This Article