Neural Science and Vision – Part 8

Daniel M. Wolpert is a huge baseball fan. Not just for the love of the sport, but for the way in which the motor planning and act of hitting a baseball illustrates the principles of sensorimotor control.

Principles of Sensorimotor Control is the title of chapter 30 in the new (6th) edition of Principles of Neural Science, a chapter co-authored by Daniel Wolpert and Amy Bastian, and one that is at the core of what old OEP masters used to address regarding “vision is motor”. Wolpert and Bastian state at the outset that motor systems produce neural commands that act on muscles causing them to contract and generate movement. This is challenging for at least six essential reasons:

1. Motor systems have to contend with different forms of uncertainty. We’ve addressed this before with regard to Bayesian Probabilty, and if you check out a blog entry from eight years ago on Visual Thinking and the Bayesian Brain you’ll note the link to Wolpert’s TED talk in which he states that people have found out that studying vision in the absence of finding out why you have vision is a mistake.  You have to study vision with the realization of how the movement system is going to use vision. And in this context, one of the principle roles of vision is to reduce the cloud of uncertainty inherent in motor actions through guided feedback, as illustrated below in Figure 30-5.

2. The motor system reduces the degrees of freedom of the musculoskeletal system by controlling groups of muscles, termed synergies, to simplify control. On the one hand we’re used to thinking of patients developing degrees of freedom in motor systems such as accommodation and convergence. But a well-functioning visual system, specifically at nearpoint, is a synergistic interplay of groups of muscles that are neurotypically linked.

3. Unwanted disturbances, termed noise, corrupt many signals. Noise is present at all stages of sensorimotor control, from sensory processing, through planning, to the outputs of the motor system.

4. Time delays are present at all stages of the sensorimotor system. This includes the delays arising from receptor dynamics, conduction delays along nerves and synapses, and delays in contraction of muscles in response to motor commands. These delays, which can be on the order of 100ms, though longer for vision (up to 150ms) and proprioception, depend on the sensory modality and the complexity of processing.

5. The properties of the motor system are continually changing. This places a premium on our ability to use motor learning to adapt control appropriately. Discrepancies between predicted and actual sensory feedback, termed sensory prediction error, is essential in motor control.

6. Error-based learning is the driving force behind sensorimotor adaptation paradigms. An example provided is the relation between the visual and proprioceptive location altered by prism. The initial shift in visual input results in visual reach misdirection. Over repeated attempts the trajectories adjust to account for the discrepancy between vision and proprioception, reflecting the brain’s reorganizing or adjusting motor commands, a process termed visuomotor learning.

One other item in this chapter of note is Figure 30-9 and the concept of motor equivalence. That term was frequently used in the 1960s and 70s by perceptual-motor pioneers such as Kephart (we blogged about him here), and was a big deal in handwriting development. It is still pertinent with regard to the fine motor movements in keyboarding, as well as larger bilateral integrative movements such as chalkboard circles in the vision therapy room.