According to Reflex

A reflex, or reflex action, is an involuntary and nearly instantaneous movement in response to a stimulus. A reflex is made possible by neural pathways called reflex arcs which can act on an impulse before that impulse reaches the brain.

Is the magnitude, direction, etc. of this response evaluated somewhere in some type of neurons? Is there information processing in relation to, for this type of stimulus magnitude, it's evaluated which magnitude of response should be triggered by a neuron, or does the stimulus simple activate a neural pathway which activates a fixed response and the difference in the response is the result of how many neural pathways the stimulus activate?

To make an analogy, do reflexes work like turning several light switches for more or less light, or like inputing data into a program for different kind of responses?


1 Answer 1


Short answer
Reflex arcs can perform basic processing operations like inversion, summation and integration.

Reflexes include a wide variety of responses and in fact an all-inclusive definition seems to be lacking. Mann defines a reflex on their website as

[A r]elatively stereotyped movement or response elicited by a stimulus applied to the periphery, transmitted to the central nervous system and then transmitted back out to the periphery

In textbooks, reflex arcs are often depicted as simple on/off switches, and the one underlying the knee jerk reflex is one that is often cited (Fig. 1).The quadriceps is key here. When it contracts it extends the angle of the knee joint by raising the lower leg. Tapping the patellar tendon momentarily stretches the quadriceps muscle and causes the sensory receptor of the muscle, called a spindle fiber, to send a signal along the afferent neuron to the spinal cord. This causes the efferent neuron to return a signal to the quadriceps muscle to contract and lift the lower leg. This action resists the initial stretch and is a classic example of negative feedback.

The knee jerk reflex arc in Fig. 1 is a so-called monosynaptic arc; just a single synapse connects the sensory and motoneuron. As far as I know, this simple circuit is more like an on/off switch; tap the tendon and the neurons start firing. However, if you tap it softly, nothing may happen, either because the spindles' signal is too low to reach threshold of the afferent neuron, or too little afferent neurons are recruited for a visible jerk to occur.

There are also disynaptic and polysynaptic reflex arcs due to the presence of 1 or more interneurons within the spinal cord that connect the sensory and motoneuron.

The interneuron in a disynaptic arc is often inhibitory. In this case the excitatory input from the sensory neuron is 'reversed' into an inhibitory response, thereby for instance reducing muscle tone instead of increasing it. In terms of logical circuits the interneuron would act as a 'NOT' (inverter) port. In that sense, the reflex arc performs a type of very basic Boolean processing (Fig. 2).

It is also known that for instance clasping the hands together and pulling on one of them (the Jendrassik maneuver) the tendon tap reflex is increased in magnitude. So somehow the reflex arc is subject to integration of neuronal inputs, also akin to a processing step.

Likewise, reflex arcs can be subject to enhancement by repetitive stimuli, akin to a summation processing step somewhere in the arc (Eccles & Sherrington (1930); Johnson et al., (2017)).

reflex arc

Reflex arc of the knee jerk (patelllar reflex). source: Wikipedia

disynaptic reflex arc
Fig. 2. The same knee jerk circuit as Fig. 1, but now with the accompanying disynaptic reflex arc drawn as well. This arc releases muscle tone of the antagonistic flexor muscle, such that the extensor muscle activated by the knee jerk can freely lift the leg. source: Human Physiology Academy

- Eccles & Sherrington, J Physiol (1930); 69(1): 1–28
- Johnson et al., Front Neural Circuits (2017); 11: 83

  • 1
    $\begingroup$ Great answer. +1 I learned a while ago that for people who have trouble with the gag reflex (eg, trying to swallow pills, going to the dentist, etc), there are ways to desensitize or distract from the reflex, suggesting that some intervening processing may be possible. $\endgroup$
    – Arnon Weinberg
    Commented Dec 1, 2021 at 20:59
  • 1
    $\begingroup$ "Likewise, reflex arcs can be subject to enhancement by repetitive stimuli, akin to a summation processing step somewhere in the arc". So, in neural network terms (not human terms) do they learn? $\endgroup$
    – Pablo
    Commented Dec 1, 2021 at 21:09
  • $\begingroup$ @Pablo - Given the comment from Arnon the circuits can at least be desensitized. If they can get sensitized as well, you might call that learning. $\endgroup$
    – AliceD
    Commented Dec 2, 2021 at 8:45
  • $\begingroup$ @aliced this is all interesting because there are several concepts in NS which are ambiguous, and when you look for definitions to see what they can actually meaning in empirical terms, you reach things like this. Consciouness : awareness. Awareness: knowledge. Knowledge: information acquired through experience. When you look for things such as, where does consciouness begin, can you rule out reflex movements given these definitions? And if these definitions arent the correct ones, which ones are the correct ones? With other definitions you reach circular definitions. $\endgroup$
    – Pablo
    Commented Dec 2, 2021 at 10:52
  • 1
    $\begingroup$ @Pablo - Scientific terminology is messy at best $\endgroup$
    – AliceD
    Commented Dec 2, 2021 at 10:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.