Golgi Tendon Organs

Firstly congratulations on compiling a great site. It is no doubt of great use to many in the field. However, whilst I agree with much of what you say about programming for resistance training I have noted a few doubtful justifications. For example, you state...

"It is theorized that over use of forced repetitions on very heavy weight may teach the muscles to prematurely fail. Strength training involves a neurological adaptation (motor development, contraction efficiency), as well as a morphological adaptation. Repeated use of forced repetitions on very heavy weight may prematurely activate the golgi tendon organ. Activation of the golgi tendon organ inhibits muscular contraction to protect the muscle from perceived injury." See Forced Repetitions.

As far as I know there is no justification for these beliefs although I recognize that they are held by many practitioners. The golgi tendon organ is no where near as powerful an inhibitor of muscle activation as many in the fitness industry believe (see Crago et al., 1975). In fact it merely has the effect of slightly inhibiting the firing rates of some units in order that smooth increases in force can be produced when another unit is recruited.

The following is an extract from my PhD thesis that addresses this issue.

Partner Assisted Pullup

"In the strength training literature, the inhibitory influence of the Golgi tendon organ is frequently cited as a possible limiting factor to voluntary muscle activation (see for example; Determining factors of strength, (1985). National Strength and Conditioning Association Journal, 7(1): 10-23 and 7(2): 10-17). However, this view is inconsistent with a number of experimental observations. Firstly, there is evidence that type Ib afferents deliver only weak autogenic inhibition (Crago et al., 1975). Secondly, Ib afferents of limb extensor muscles deliver, in certain movement contexts, an excitatory influence to both the muscle that they innervate and that muscle's synergists (Pratt et al., 1995). Furthermore, during maximal voluntary contractions of the human tibialis anterior and hand muscles, afferents of muscular origin deliver a net excitation to homonymous motoneurones (Gandevia et al., 1990; Macefield et al., 1993). This is most clearly demonstrated by the 30 to 40% decline in the maximum motoneurone firing rate that occurs when the motor nerves are blocked, distal to the recording site, by anaesthetic (Gandevia et al., 1990; Macefield et al., 1993). These findings suggest that, during brief contractions, the excitatory influence of type Ia and II afferents from muscle spindles greatly exceeds any possible inhibitory effect of type Ib afferents from Golgi tendon organs. These observations also clearly indicate that afferent fibers from muscle spindles provide a significant proportion of the excitatory drive to the motoneurone pool during maximal voluntary contractions. "

Please note that I have written to you only because...

    1. you have obviously gone to a lot of trouble to compile a great resource and I feel that a few loose hypotheses reduce its value unnecessarily.
    2. On these sorts of issues your opinions are almost exactly the same as mine were before I started to study the field in more depth.

The GTO myth is just one of many in the area. The myths exist because too many in the exercise sciences are happy to hypothesise on the basis of a superficial knowledge of the underlying physiology.

By the way - I do not mean to suggest that forced reps are appropriate because I have no evidence of that and I know that many (myself included) make good progress without them.

Keep up the good work and best wishes

Tony Shield

  • Crago, P. E., Houk, J. C., and Rymer, W. Z., (1975). Influence of motor unit recruitment on tendon organ discharge. Neuroscience Abstracts, 1: 280.
  • Gandevia, S. C., Macefield, V. G., Burke, D., and McKenzie, D. K., (1990). Voluntary activation of moto axons in the absence of afferent feedback. Brain, 113: 1563-1581.
  • Macefield, V. G., Gandevia, S. C., Bigland-Ritchie, B., Gorman, R. B., and Burke, D., (1993). The firing rates of human motoneurones voluntarily activated in the absence of muscle afferent feedback. Journal of Physiology, 471: 420-443.
  • Pratt, C. A., (1995). Evidence of positive force feedback among hindlimb extensors in the intact standing cat. Journal of Neurophysiology, 73(6): 2578-2583.

Tony Shield,

I appreciate your bringing the the golgi tendon organ issue to my attention. Without referring to the studies you cited, I hope you do not mind me asking you questions, so I might make more sense of the implications of your findings.

James Griffing

Hi James

I will try to answer your questions as best I can by addressing the original statement with which I took issue - the idea that the GTO inhibits MVC force and that training with forced reps may enhance its influence and therefore inhibit or reduce strength gains.

  1. As stated previously, the GTO reflex is not strong enough to significantly inhibit MVC force (eg. Crago et al., 1975). GTOs start to fire at very low forces (the sorts of forces that are created by a single motor unit or even a single fibre) and they increase their output to maximum well before contractions reach 50% of maximum. Thus the maximum inhibitory effect of the GTO is evident in relatively weak contractions - good evidence that it is weak. Furthermore, the type Ib fibers from GTO's are not plentiful and they are vastly outnumbered by other potentially inhibitory fibers (type III and IV that come from various receptors in ligaments, joint capsules and the muscles themselves).
  2. In fatigue (induced by a sustained contraction) the negative feedback from the GTO's decreases (Zytnicki et al., 1990). Consequently, the GTO's influence is probably even smaller during the forced reps portion of the set than during the first few reps.
  3. There is no evidence that suggests the GTO exerts progressively more influence in response to any form of training (Hutton and Atwater, 1992).
  4. It is also important to consider that reflexes can be overriden by the voluntary commands that control movement. Otherwise we would be slaves to our reflexes and our movement control would be very limited and stereotyped. It is well known (outside S&C circles but perhaps not inside them) that many reflexes can be amplified or damped by the commands that accompany voluntary movement. Reflexes that are inhibitory tend to be inhibited themselves (by the process of pre-synaptic inhibition) during MVC's.

1) If you were handed a weight and you tried to prevent it from falling, but could not because it was too heavy, would not the golgi tendon organ's inhibitory responses be greater than the muscle spindles myotatic reflex?

I don't believe so because of the reasons outlined in points 1 and 4 above - if there is an inhibitory effect it almost certainly doesn't come from the GTO. However, my exact response would depend on how much weight I was given. Assuming that the load was just above my max for an isometric contraction the eccentric contraction would be slow. In this case the spindles would be strongly activated by the stretch and would deliver a powerful excitatory effect to the motoneurone pool.

You may remember the previously quoted studies that indicate that the sensory information from muscle is much more excitatory than inhibitory during attempted isometric MVCs (when this feedback is taken away the output from the spinal cord decreases by 30-40%). I would imagine that this would be the same during slow eccentric contractions with 105-130% of the maximum load.

If the load was extremely heavy (160-200% max) I would probably drop it based on the feedback obtained from a number of sensory organs, including the GTO. This information would arrive in the brain and I would decide that resistance was futile. In this case the GTO (along with other sensors) provides the feedback necessary for a safe decision to be made but does not actually inhibit effort during an MVC.

Whilst on the topic it must be said that muscle activation is incomplete during eccentric MVC's (Amiridis et al., 1996; Westing et al., 1990), but the reason why is a mystery. For the reasons stated above it seems unlikely that the GTO would exert a powerful enough effect to significantly limit muscle activation.

2) If you were performing a bench press and paused the bar at the chest, would you not inhibit the muscle spindles reflex. You have noticed that it is much more difficult to push the weight back up after a pause. It seems the muscle spindle reflex is greatest immediately at end of the eccentric contraction, or change in direction. If this is true, would there not be phases of a weight training exercise that the golgi tendon organ inhibitory reflex could be potentially greater than the muscle spindles excitatory reflex - for example, the middle of the concentric contraction?

The contribution of the stretch reflex would be smaller with the pause, however for the same reasons (1 & 4) it is doubtful that the inhibition provided by the GTO is powerful enough to provide much inhibition. Furthermore you do not eliminate the stretch reflex with a pause - the excitatory input from type Ia and type II fibers (from muscle spindles) is still high during maximal isometric and concentric contractions (eg. Hutton and Atwater, 1992). In these cases the spindles are stretched by the action of the intrafusal fibers which are powerfully activated (by the gamma motoneurones) during MVC's.

The pause interferes with performance mainly because the mechanical energy stored within the cross-bridges is allowed to dissipate.

3) Does your research negate the golgi organs role in Proprioceptive neuromuscular facilitation (PNF) stretches.

PNF techniques and the sometimes dodgy theories behind them are covered well by Hutton in chapter 2c of Strength and Power in Sport that you quoted Komi from. Hutton suggests that the GTO does not play the role that many people think and provides some convincing evidence for his arguments.

It is most everyone's experience who perform PNF stretches that after pushing against an immovable force, a muscle can be stretched significantly more through reciprocal innervation. Does this not demonstrate that the golgi tendon organ inhibitory reflex can be greater than the muscle spindles excitatory reflex?

No - it demonstrates that PNF stretches are a valuable means of increasing flexibility but it demonstrates nothing with regard to the GTO's involvement because numerous other factors could equally well explain these observations.

4) In accordance with your studies, is this still true? During running, don't we experience the effects of both the golgi tendon organ and the muscle spindle. On impact of the heel: golgi tendon organ's inhibition to the dorsal flexors and excitation of its antagonist (plantar flexors). As the body moves forward, weight is shifted to the forefoot stretching the plantar flexors: muscle spindles' excitation to the plantar flexors.

You are probably on the right track here because reflexes do take a significant role in gait which is automated to a significant extent. MVCs are not automated to the same extent.

5) In what movement contexts does the golgi tendon organ exhibit an excitatory influence to both the muscle that they innervate and that muscle's synergists (re: Pratt et al., 1995)

When load bearing - an increase in load is met by a reflex that excites the extensor muscles of the weight bearing limbs (this is aka the loading reflex). Thus when it is desirable for an increase in load to be met by an increase in extensor force the GTOs input is processed differently than when an increase in force is not desirable.

6) As reported by Kreighbaum's Biomechanics text: Komi (1992) proposed the decreasing the sensitivity of the Golgi organs through strength and power training... ...If decreasing the Golgi organs sensitivity is possible, would not increasing its sensitivity also be possible?

Komi made the proposal but this is not evidence that the adaptation actually occurred. I don't know of any research that can back-up his guess. Neither did Hutton in 1992 (Hutton and Atwater, 1992). Interesting that Hutton contributed to Komi's book just before he discredited Komi's claim in another article.

7) As far as proprioceptive functions, would you disagree that golgi tendon organs are sensitive detectors of tension in their host muscle? As sensors of degrees of tension, would they not be important contributors to fine motor control as well as protective agents?

No I would not disagree with the statement that the GTOs are detectors of tension. However, whether they protect us in the manner that you suggest is doubtful for reasons 1 & 4. They may, however, (along with a number of other receptors such as pressure receptors in the skin and joints) protect us in the manner outlined in answer to question 1. (ie. they may let us know when it is wise to jump out of the way).

It is important to consider that much of the neurophysiological research that influenced strength and conditioning (S&C) researchers and practitioners in the early days was performed on animal preparations in which certain parts of the nervous system were surgically removed. For example, much of Houk's work on the GTO was done with cats who had had their spinal cords cut or certain parts of the brain destroyed. Consequently, this work showed how the spinal reflexes could work in the absence of voluntary control and without the addition of longer loop reflexes that involve the brain. The picture of the GTO that emerged from these studies is significantly different from that which is emerging from more recent studies on intact animals and humans.

Hope this helps. Best regards.

Tony Shields

  • Amiridis, I. G., Martin, A., Morlon, B., Martin, L., Cometti, G., Pousson, M., and van Hoecke, J., (1996). Coactivation and tension regulating phenomena during isokinetic knee extension in sedentary and highly skilled humans. European Journal of Applied Physiology, 73:149-156.
  • Hutton and Atwater (1992) Sports Medicine 14(6): 406-
  • Westing, S.H., Seger, J. Y., and Thorstensson, A., (1990). Effects of electrical stimulation on eccentric and concentric torque-velocity relationships during knee extension in man. Acta Physiologica Scandinavica, 140: 17-22.
  • Zytnicki et al., (1990) J. Neurophysiol, 64: 1380-

Tony Shields

  • BHMS, Hons I
  • PhD Southern Cross University N.S.W Australia

Tony's undergraduate study (bachelor of human movement studies with honours) was completed at the University of Queensland. Shield's chief area of investigation during his PhD studies included 'neural adaptations to resistance training' and he was specifically interested in examining the extent to which increases in voluntary muscle activation contribute to increases in strength and the task-specificity of those increases.

Related Articles