28.
When the foot is in the dorsal flexion position, the soleus MLR corresponds directly with the clonus beats. However, during plantar flexion, the soleus MLR and clonus beats are lost.

29.
In this study we examined the similarities between clonus beats and the soleus MLR in terms of latency and amplitude. We emphasized the relationship between these two responses.
30.
The similarities between stretch induced ankle clonus and peroneal nerve stimulation induced soleus MLR give a chance for speculation about the afferents of the responses.
31.
Even if previous publications considered mainly Group I afferents as clonus afferents, we propose Group II afferents to be evaluated in the formation of clonus because of findings about soleus MLR afferents.
32.
FCR MLR is an example of a response in the upper extremity. Stimulation of an antagonist muscle nerve results in a medium latency response by the agonist muscle. Radial nerve stimulation causes the EDC muscle to contract and MLR is achieved by latency much later than the FCR H reflex.
33.
Clonus can be obtained through radial nerve stimulation during wrist extension, as with lower extremities ankle clonus.
34.
Latency of the response obtained from the FCR is about 42-43 ms. However, FCR -H or T reflex latency does not exceed 20 ms.
35.
In the lower extremities, Patellar clonus could establish correlation with MLR, such as ankle clonus. In this case, clonus can be obtained by femoral nerve stimulation.
36.
The medium latency response is about 50 ms at the antagonist Biceps femoris muscle during the patellar T reflex.
37.
However, latency of the H-reflex obtained by femoral nerve stimulation is about 15-20 ms. Clearly it is much shorter than the MLR latency.
38.
And bringing together all of these values​, an interesting comparison can be made with clonus latencies. We have already mentioned the values ​​of clonus frequencies.
39.
H / T reflex latencies and clonus periods were compared to MLR latency. The Group II afferents constitute a more convenient time for the clonus period as opposed to Group Ia afferents.
40.
When the measured clonus frequencies are compared to the calculated frequencies that are formed by Group Ia and Group II afferents, we can say that Group II afferents contribute to the clonus.
41.
We examined the effects of the afferent paths on the frequency of clonus. Returning to the formula for oscillation again; we can ask what the factors that determine p are. So, which factors determine dumping or building oscillation?
42.
Factors which tend to produce oscillation will increase the value of the p, while factors which tend to stabilize the system will decrease the value of p.
Sustained clonus has to have p=0 value because it is a maintained oscillation.
44.
We can calculate the damping ratio by dividing the amplitude of the first and second beats. In Ash 1 spasticity which mildly increase gain of the reflexes.
45.
In Ash 2 spasticity, the damping ratio is increased. The gain of the reflex is much higher than in Ash 1 spasticity.
46.
But even if the gain of the reflex is increased, the damping ratio is decreased in Ash 3 spasticity.
47.
The damping ratio decreases in Ash 3 spasticity.
48.
And in Ashworth 4, even if there is maximum increase gain, the damping ratio is not high because the damping ratio could not be calculated due to the second beats cancellation.
49.
When all the data is put together, what we see is quite interesting. The damping ratio is below normal in Ash 1 and 2, but it is up to the normal level in Ash 3. So, decaying of the oscillation is low in the situation increasing gain of the reflex but the reverse occurs in Ash 3 and 4.
50.
We can conclude that as the reflex gain increases, decaying of the oscillation is lowered.
51.
Similarly, if you look at the pendulum counts, they increase with high reflex gain but reverse in Ash 3.
52.
We can summarize by saying that; p ; Related to the gain of the stretch reflex and f ; Related to the delay of the stretch reflex afferents (paths).
So, in response to the question "Why do we have reflex oscillations?" we could say, "Because our reflexes have reflex gain and a delay in the reflex afferent path."
53.
I would like to come back to the old story.
Stein and Oguztöreli’s model can explain the oscillation phenomenon by gain and delay of the reflexes.
54.
With a very nice prediction in this model they showed that the afferent path of the stretch reflex is not one-way. If there are three different afferent paths, they change the situation relating to the oscillation of the system. Through different delays and different reflex gains, afferents lead up to stabilize the oscillation.
55.
Stein and Oguztöreli put forward the theory that sharing the rates of p and f between different paths can have stabilizing effects. As is graphed in this figure, each individual path is prone to oscillation with its own reflex gain, the cumulative effects of the different paths suppress oscillation and the system becomes more stable.
56.
Perhaps "Why do we have reflex oscillations?" is not the right question. It might be better to ask: "Why don't we have reflex oscillations in normal conditions? The answer is multiple delays and divided gain in multiple paths. This phenomenon may be referred to as "stabilizing factors".