Day 154: Review – Spinal stretch reflexes support efficient hand control

Fig 1 (a) from the paper, showing the multijoint perturbation away from target (red) with simultaneous flexion at the elbow and either flexion, extension, or no perturbation at the wrist joint.

Today is that critical review paper I promised. Everything following this introduction explains how the experiment was done, what they found, and why I think it is particularly interesting. To me the experiment was so well thought out I couldn’t think of anything I would change. Instead I focused on the methodology they used and why it highlights the importance of a well thought out experiment. This is really my first attempt at a “critical review” so take it how you will.

This paper consists of seven different experiments, each experiment was done to test the modulating effect wrist placement has on the stretch reflex of the elbow when perturbed away from a desired target. The experiments show that the stretch reflex integrates inputs from multiple muscles to produce efficient corrective responses. Furthermore, they show that by changing the hands orientation, the pattern of stretch reflexes at the elbow were reversed in a way that was appropriate for returning the hand to the target position. Even the fastest – and therefore less computationally intensive – reflexes were tuned to the hand’s displacement relative to the target and not simply the amount of stretch experienced by the triceps or biceps muscles.

Using an upper limb robotic exoskeleton to position the arm, the researchers recorded EMG activity from the triceps, biceps, flexor carpi ulnaris and extensor carpi radialis. Each experiment tested a different aspect of the spinal stretch reflex. In the first the exoskeleton simultaneously flexed the elbow and either flexed, extended, or did not alter the angle of the wrist. The subjects were told to counter this perturbation and to return to the original target position. For the second experiment this was repeated using a perturbation that caused elbow extension. For the third experiment, subjects were told for the first block to return their hand quickly to the target following perturbation and for the second not to intervene to determine if the stretch reflex was influenced by task goals. Next, to show that the functional modulation occurs because of the wrist flexor spindles, the fourth and fifth experiments involved applying a small load to the wrist to inhibit the spindles in the wrist. The purpose of which was to either decrease or eliminate functional tuning the first experiment was done to test inhibition of wrist flexion spindles and the second to test inhibition of wrist extension spindles. For the sixth experiment, they used the previous experimental paradigm, but with a neutral wrist placement and without weight to inhibit or excite the muscle spindles. For the seventh experiment, they ruled out a hardwired response by using the same setup, changing only the orientation of the thumb. Either with the thumb pointing upwards, and again with the thumb pointing downwards.

One of the most interesting things about this study was the level of thought the researchers put into mapping out the spinal stretch reflex and its relationship between the wrist and elbow location. Another interesting aspect of the study is the way they performed the tests. Instead of trying to determine what the spinal cord is computing directly, by attempting to record from the cord itself, they used methods that are commonly found in control systems design. By perturbing the system and recording the response, the underlying behavior, also called the transfer function of the system, can be elucidated. This distinct style of methodology is evident in the last experiment. When the wrist was rotated (thumb in the downward position instead of the upward position) the response of the system, returning to the target position, was not negatively impacted. They show that the stretch reflex was tuned to the hands displacement from the target rather than the elbows rotation. The researchers go on to conclude that the “arm’s orientation diametrically altered the pattern of the triceps spinal stretch reflex and did so in a way that was appropriate for returning the hand to its initial location.”

Because this is a reflexive pathway, the idea that there could be a non-linear and coordinated response between the wrist and elbow joint is very interesting. Traditionally reflexive pathways are thought to be very basic. This result highlights both the lack of understanding in how the spinal cord functions and demonstrates how the spinal cord circuitry is more complex than previously thought. Moreover, this pathway does not exist at each joint. Previous studies have shown that the same connections don’t exist between the elbow and shoulder. There are several possibilities for why this may be, such as anatomical arrangement. However, the lack of this pathway in the shoulder further highlights that a multiple joint relationship isn’t the default of the spinal cord, instead it is something that was developed specifically like the vestibulo-ocular reflex. Overall, this paper demonstrates the importance of a well thought out experiment and how the same experiment can be used with minor changes to develop a nuanced understanding of a single pathway.

This paper consists of seven different experiments, each experiment was done to test the modulating effect wrist placement has on the stretch reflex of the elbow when perturbed away from a desired target. The experiments show that the stretch reflex integrates inputs from multiple muscles to produce efficient corrective responses. Furthermore, they show that by changing the hands orientation, the pattern of stretch reflexes at the elbow were reversed in a way that was appropriate for returning the hand to the target position. Even the fastest – and therefore less computationally intensive – reflexes were tuned to the hand’s displacement relative to the target and not simply the amount of stretch experienced by the triceps or biceps muscles.

Using an upper limb robotic exoskeleton to position the arm, the researchers recorded EMG activity from the triceps, biceps, flexor carpi ulnaris and extensor carpi radialis. Each experiment tested a different aspect of the spinal stretch reflex. In the first the exoskeleton simultaneously flexed the elbow and either flexed, extended, or did not alter the angle of the wrist. The subjects were told to counter this perturbation and to return to the original target position. For the second experiment this was repeated using a perturbation that caused elbow extension. For the third experiment, subjects were told for the first block to return their hand quickly to the target following perturbation and for the second not to intervene to determine if the stretch reflex was influenced by task goals. Next, to show that the functional modulation occurs because of the wrist flexor spindles, the fourth and fifth experiments involved applying a small load to the wrist to inhibit the spindles in the wrist. The purpose of which was to either decrease or eliminate functional tuning the first experiment was done to test inhibition of wrist flexion spindles and the second to test inhibition of wrist extension spindles. For the sixth experiment, they used the previous experimental paradigm, but with a neutral wrist placement and without weight to inhibit or excite the muscle spindles. For the seventh experiment, they ruled out a hardwired response by using the same setup, changing only the orientation of the thumb. Either with the thumb pointing upwards, and again with the thumb pointing downwards.

One of the most interesting things about this study was the level of thought the researchers put into mapping out the spinal stretch reflex and its relationship between the wrist and elbow location. Another interesting aspect of the study is the way they performed the tests. Instead of trying to determine what the spinal cord is computing directly, by attempting to record from the cord itself, they used methods that are commonly found in control systems design. By perturbing the system and recording the response, the underlying behavior, also called the transfer function of the system, can be elucidated. This distinct style of methodology is evident in the last experiment. When the wrist was rotated (thumb in the downward position instead of the upward position) the response of the system, returning to the target position, was not negatively impacted. They show that the stretch reflex was tuned to the hands displacement from the target rather than the elbows rotation. The researchers go on to conclude that the “arm’s orientation diametrically altered the pattern of the triceps spinal stretch reflex and did so in a way that was appropriate for returning the hand to its initial location.”

Because this is a reflexive pathway, the idea that there could be a non-linear and coordinated response between the wrist and elbow joint is very interesting. Traditionally reflexive pathways are thought to be very basic. This result highlights both the lack of understanding in how the spinal cord functions and demonstrates how the spinal cord circuitry is more complex than previously thought. Moreover, this pathway does not exist at each joint. Previous studies have shown that the same connections don’t exist between the elbow and shoulder. There are several possibilities for why this may be, such as anatomical arrangement. However, the lack of this pathway in the shoulder further highlights that a multiple joint relationship isn’t the default of the spinal cord, instead it is something that was developed specifically like the vestibulo-ocular reflex. Overall, this paper demonstrates the importance of a well thought out experiment and how the same experiment can be used with minor changes to develop a nuanced understanding of a single pathway.

Source:

Weiler, J., Gribble, P.L. & Pruszynski, J.A. Spinal stretch reflexes support efficient hand control. Nat Neurosci 22, 529–533 (2019) doi:10.1038/s41593-019-0336-0