Journal Logo

Clinical Sciences: Clinically Relevant

Shoulder proprioception: latent muscle reaction times


Author Information
Medicine & Science in Sports & Exercise: October 1999 - Volume 31 - Issue 10 - p 1394
  • Free


The neuromuscular reflexive arc consists of both afferent and efferent neural components and is greatly influenced by visual, auditory, and vestibular exteroception. Afferent components originate from musculotendinous, capsuloligamentous, and cutaneous mechanoreceptors and terminate in the dorsal spinal cord, while the efferent components originate from the ventral spinal cord and terminate at the myoneural junctions of numerous muscle fibers (19).

No previous studies to our knowledge have reported rotator cuff muscle latent muscle reaction timing (LMRT) for either normal untrained subjects or trained overhead throwing athletes. Throwing is a complex motor task requiring continuous afferent feedback and precise muscular activation to provide both accurate throws and dynamic shoulder stabilization during deceleration. When throwing a ball, the glenohumeral joint approximates 90° of abduction and external rotation at ball release (25). During the acceleration phase of throwing, glenohumeral joint internal rotation can achieve velocities of 90000·s −1 (28). Following ball release, sudden shoulder deceleration occurs via eccentric rotator cuff and posterior deltoid muscle activation (12,16,28).

Because of their slowly adapting nature, the contributions of capsuloligamentous mechanoreceptors to proprioception tends to be diminished during higher (functional) velocity movements, with this function being primarily provided by musculotendinous mechanoreceptors (26). Previous glenohumeral joint proprioception studies have relied on relatively low (nonfunctional) velocities to maximize capsuloligamentous proprioceptive contributions while minimizing musculotendinous contributions (1,31). Muscle spindles and golgi tendon organs (GTO) are rapid response mechanoreceptors responsible for sensing musculotendinous length and tension changes, respectively. During the deceleration phase of throwing, peak shoulder external rotator muscle torque of 300 inch-lbs (or 33.88 Nm) may occur at terminal range of motion, thereby stimulating the GTO (25). Houck et al. reported that GTO activation was minimal until the end range of motion when the passive tension of muscular antagonists were greatest (15).

This study compared differences in rotator cuff LMRT in response to a sudden internal rotation perturbation force between trained overhead throwers and untrained normal control subjects. The hypothesis was that the specific training of the overhead throwing group would produce rotator cuff LMRT differences (P ≤ 0.05) compared with control group subjects. Differences detected between trained overhead throwers and untrained control subjects are attributed to the proprioceptive influences of repetitive overhead throwing. The ability to detect the instant at which the rotator cuff muscles decelerate the glenohumeral joint would provide critical data to clinicians who design and implement conditioning programs for injury prevention and who rehabilitate overhead throwers following injury or surgery. These data would provide evidence of neuromuscular adaptations that occur as a direct result of overhead throwing training which would prove useful in designing injury prevention conditioning programs for the glenohumeral joint or in the rehabilitation of injured throwers.



Fifteen trained overhead throwers (male intercollegiate baseball players) and 15 untrained controls (male subjects not active in competitive throwing activities) participated in this study. Subjects from both groups were between 18 and 24 yr of age. A brief medical history was completed and a screening exam was performed by the primary investigator to eliminate subjects with rotator cuff injury history, deficits in dominant upper extremity range of motion/strength, or excessive glenohumeral joint laxity (20). Dominant glenohumeral joint external rotation was determined by the primary investigator with a handheld goniometer (24). Subjects were given a detailed explanation of the research procedures and possible risks. Following this, subjects read and signed a Medical Institutional Review Board approved informed consent form.


Subjects were seated (to enable normal scapulothoracic function) with their glenohumeral joint first abducted and externally rotated 90° and then horizontally adducted 30° (as verified by the primary investigator with a handheld goniometer). This “scapular plane” position was chosen to enhance rotator cuff muscle function and decrease tendon impingement (17,34). The elbow was flexed to 90°, and the forearm was placed in a foam sleeve. The sleeve was attached to a spring loaded perturbation device (Fig. 1A). During testing, subjects wore headphones to eliminate auditory feedback and were instructed to keep their eyes closed to eliminate visual feedback. The approximate axis of glenohumeral joint rotation was aligned with the mechanical axis of rotation of the perturbation device. The forearm sleeve provided uniform compression and support within the device while diminishing the contribution from cutaneous mechanoreceptors (1,31). A block attached to the wheel of the device limited maximal internal rotation to 45°. This study attempted to negate contributions from GTO by initiating the internal rotation perturbation forces within the mid-range of glenohumeral joint motion. Subjects were instructed to maintain their head in an upright and neutral position throughout testing to negate the influence of varying neck postures on reflex muscle activation (32).

Figure 1:
A. Start position for sudden shoulder internal rotation perturbation. B. Finish position for sudden shoulder internal rotation perturbation.

Electromyographic (EMG) data using fine wire techniques were collected from the: supraspinatus, infraspinatus, and teres minor muscles (5). Surface electrodes were used with the posterior deltoid muscle. Proper electrode placements were confirmed by manual muscle testing and EMG signal inspection on an oscilloscope (20). Before test trials, an EMG “quiet file” was collected from each muscle, stored on a PC type computer, and subsequently used as the baseline from which muscle activation onsets were determined. Quiet files were collected with subjects seated and with their arms resting in their laps to determine resting or baseline EMG activity.

During test trials a sudden internal rotation perturbation force was applied following a period of relative muscular inactivity (as determined by EMG signal inspection on an oscilloscope by the principle investigator). A manual switch enabled the release of a variably timed, sudden internal rotation force from the perturbation device to the dominant upper extremity at a velocity of approximately 2000·s−1 (Fig. 1B). This velocity was selected to enhance the proprioceptive effects of rotator cuff muscle spindle activation and negate the contributions of capsuloligamentous mechanoreceptors (because of their slow adaptation characteristics) (14). Subjects were instructed to react to the sudden internal rotation perturbation force by attempting to decelerate and stop the movement. Subjects performed three practice trials with visual and auditory feedback to ensure a proper understanding of verbal cues and to familiarize themselves with the device. Following this, 10 consecutive test trials were performed and recorded (without visual and auditory feedback) at varying time intervals between trials. Following each trial, subjects were instructed to relax and not resist the device as it was manually returned to the test initiation position.

EMG analysis.

EMG data sampling (2000 Hz, 2-s duration) was initiated immediately before perturbation device activation. Following sampling, EMG data underwent an analog to digital conversion and was stored on a PC-type computer. Muscle activation onsets for each trial were determined using Asyst v.2.1 software (Macmillan Software Company, New York, NY) and the following protocol: First, a threshold voltage for each muscle onset was calculated from an EMG “quiet file” (baseline muscle activity at rest); the mean and SD of the rectified signal was then determined; following this, the threshold voltage (Vo) required for muscle onset was calculated from the following equation: VO = Mean + 3 * SD . A 3 standard deviation activation threshold was used to minimize the probability of incorrectly describing an inactive muscle as being active (Type I error). Excellent reliability is reported with this computer assisted analysis with simultaneous visual verification to improve the validity of muscle onsets (9,14).

The onset of muscle activity was evaluated by comparing discrete data point values (Vi) in a point-by-point fashion to the threshold voltage. When the mean voltage of a 25-ms window of data immediately adjacent to the point (50 ms total) exceeded Vo, the initial data point value was considered to represent the onset of muscle activity. The window of active EMG that most temporally corresponded to the initiation of the sudden internal rotation perturbation force was determined for each muscle. The mean values were determined for each muscle. Duration of EMG activity (time between EMG onset and cessation using the above criterion) was also assessed to determine the temporal characteristics of the muscle activation in response to the sudden perturbation.

Statistical analysis.

The response variables were the LMRT and EMG duration of the supraspinatus, infraspinatus, teres minor, and posterior deltoid muscles of the dominant upper extremity following perturbation. A one-way ANOVA was used to evaluate mean differences between groups (P < 0.05).


Mean active external rotation range of motion for the dominant glenohumeral joint was (mean ± SD) 94 ± 10°, and 97 ± 10° for the control and trained thrower groups, respectively (P > 0.05). Mean active internal rotation range of motion for the dominant glenohumeral joint was (mean ± SD) 41.5 ± 14°, and 42.1 ± 12° for the control and trained thrower groups, respectively (P > 0.05). The LMRT for the trained throwers (mean ± SD) was as follows: supraspinatus, 65.42 ± 28.75 ms; infraspinatus, 55.82 ± 27.17 ms; teres minor, 52.69 ± 28.06 ms; and posterior deltoid, 60.22 ± 25.25 ms. The LMRT for the control group subjects (mean ± SD) was as follows: supraspinatus, 63.25 ± 27.92 ms; infraspinatus, 44.94 ± 19.98 ms; teres minor, 42.44 ± 17.55 ms; and posterior deltoid, 56.86 ± 26.47 ms. Trained throwers had slower LMRT than control group subjects for the infraspinatus (P = 0.011) and teres minor (P = 0.024) muscles (Fig. 2).

Figure 2:
Latent muscle reaction times following perturbation (ms). *P < 0.05.

The muscle activation duration following perturbation for the trained throwers (mean ± SD) was as follows: supraspinatus, 821.84 ± 586.92 ms; infraspinatus, 849.54 ± 624.79 ms; teres minor, 718.23 ± 599.52 ms; and posterior deltoid, 522.74 ± 379.13 ms. The muscle activation duration following perturbation for the control group subjects (mean ± SD) was as follows: supraspinatus, 1177.68 ± 486.09 ms; infraspinatus, 754.84 ± 602.42 ms; teres minor, 676.27 ± 571.4 ms; and posterior deltoid, 711.86 ± 501.29 ms. The duration of muscle activation for the supraspinatus (P = 0.001) and the posterior deltoid (P = 0.0001) were less for trained throwers than for control group subjects (Fig. 3).

Figure 3:
EMG duration following perturbation (ms). *P < 0.05.


The results of this study suggest that trained overhead throwers have a unique rotator cuff muscle activation pattern compared with control group subjects. The delayed LMRT in the infraspinatus and teres minor muscles of the trained overhead throwers in response to sudden internal rotation perturbation may enable a prolonged acceleration phase during throwing, probably enabling a greater pitching velocity before the rotator cuff muscles are suddenly activated to decelerate the arm following ball release. The trained throwers also had a shorter EMG activation duration for the supraspinatus and posterior deltoid muscles, suggesting altered deceleratory control to enable a greater pitching velocity.

Osseous and capsuloligamentous glenohumeral joint structures that enable the necessary mobility to complete the overhead throwing task also contribute to its relative instability (2). Saha (30) and others (18,27) reported that the upper half of the teres minor and the subscapularis, supraspinatus, and the infraspinatus muscles provide the main impetus for dynamically stabilizing the humeral head within the glenoid fossa. Although static glenohumeral joint stabilization is enhanced by the inferior portion of the glenohumeral ligament when abducted and externally rotated, glenohumeral joint stabilization is greatly dependent upon dynamic rotator cuff muscle activation (27,30).

Numerous researchers have used LMRT to assess neuromuscular reflex arc responsiveness to sudden perturbation. Lynch et al. (23) studied sudden ankle inversion perturbations during standing and reported muscle activation latency ranges of 79–107 ms for the tibialis anterior, peroneus longus, and peroneus brevis muscles. Konradsen and Raven (22) reported LMRT for the peroneus longus and brevis muscles of approximately 69 ms in normal ankles and 84 ms in unstable ankles in response to a sudden 30° inversion perturbation while standing, suggesting that injury to the afferent reflex arc component directly affects the responsiveness of the efferent component. Bennett reported normal biceps brachii and brachialis muscle activation latencies between 25–60 ms and concluded that responses to elbow flexion/extension perturbations differed depending on the speed and direction of the perturbation (4). Goodin and Aminoff reported LMRT of 20–25 ms for normal wrist flexor and extensor carpi radialis LMRT of 20–25 ms as subjects resisted or released static loads, suggesting that subjects had partial volitional control over their reflexive activity (13).

The rotator cuff muscles are commonly injured from prolonged overhead throwing, and the relatively avascular musculotendinous junction of the supraspinatus muscle is the most commonly injured component (29,35). The supraspinatus muscle is maximally activated (12,16) and most commonly injured (25) during throwing deceleration, following ball release. Although the exact nature of this feedback mechanism during the throwing task is not well established, it is believed to be of considerable functional significance (1,3,7,11,13,15,22,31,33).

Reports of decreased dominant glenohumeral joint internal rotation secondary to decreased infraspinatus and teres minor muscle extensibility have been reported as a potentially injurious functional adaptation to repetitious overhead activities such as baseball (6) and tennis (21). These findings relate well with the neuromuscular adaptations in rotator cuff LMRT that we observed. We propose that these neuromuscular adaptations may precede changes in infraspinatus and teres minor muscle stiffness with their preinjury presentation, possibly serving as a valuable window for clinical intervention. Further investigations regarding this relationship is certainly warranted.

During the entire muscle activation duration, trained overhead throwers had greater amplitude signals than the control group subjects. Lack of standardization of the EMG data to a maximal voluntary isometric muscle activation renders these results difficult to interpret, although worthy of further investigation. Future investigations also need to correlate these muscle activation data with joint kinetic analysis during overhead throwing and relate these data to performance. Studies using anesthetic injection of the glenohumeral joint to enable further muscle spindle and GTO isolation may further elucidate the multifactorial control of the upper extremity during athletic performance.

Traditional rotator cuff muscle rehabilitation and conditioning has relied on relatively low velocity movements in hopes of simultaneously improving their dynamic stabilization capacity during functional tasks. Unfortunately, these methods do not adequately address the functional demands of overhead throwing (10,28). The recognition of trained rotator cuff muscle responses to sudden internal rotation perturbation provides support for the concept of task specific, velocity specific, and functionally relevant rehabilitation and conditioning of overhead throwing athletes. Cordasco (8) recommended using a medicine ball plyometric training program to train the stretch-shortening cycle of muscles that surround the glenohumeral joint. Although this method of rehabilitation trains the upper extremity in positions and at velocities that are more functionally relevant than traditional methods, considerable work remains in designing these applications for the overhead throwing athlete.

This research study was funded by American College of Sports Medicine’s Doctoral Student Award, Reebok Inc.


1. Allegrucci, M., S. L. Whitney, S. M. Lephart, J. J. Irrgang, and F. H. Fu. Shoulder kinesthesia in healthy unilateral athletes participating in upper extremity sports. J. Orthop. Sports Phys. Ther. 21:229–227, 1995.
2. Andrews, J. R., W. G. Carson, and K. Ortega. Arthroscopy of the shoulder: technique and normal anatomy. Am. J. Sports Med. 12:1–7, 1984.
3. Barrack, R. L., H. B. Skinner, M. E. Brunet, and S. D. Cook. Joint laxity and proprioception in the knee. Physician Sportsmed. 11:130–135, 1983.
4. Bennett, D. J. Electromographic responses to constant position errors imposed during voluntary elbow joint movements in humans. Exp. Brain Res. 95:499–508, 1993.
5. Basmajian, J. V. and G. Stecko. A bipolar indwelling electrode for electromyography. J. Appl. Physiol. 17:849, 1962.
6. Brown, L. P., S. L. Niehues, A. Harrah, P. Yavorsky, and H. P. Hirshman. Upper extremity range of motion and isokinetic strength of the internal and external shoulder rotators in major league baseball players. Am. J. Sports Med. 16:577–585, 1988.
7. Burgess, P. R., J. Y. Wei, F. J. Clark, and J. Simon. Signaling of kinesthetic information by peripheral sensory receptors. Ann. Rev. Neurosci. 5:171–187, 1982.
8. Cordasco, F. A., I. N. Wolfe, M. E. Wooten, and L. U. Bigliani. An electromyographic analysis of the shoulder during a medicine ball rehabilitation program. Am. J. Sports Med. 24:386–392, 1996.
9. Difabio, R. P. Reliability of computerized surface electromyography for determining the onset of muscle activity. Phys. Ther. 67:43–48, 1987.
10. Fleisig, G. S., J. R. Andrews, C. J. Dillman, and R. F. Escamilla. Kinetics of baseball pitching with implications about injury mechanisms. Am. J. Sports Med. 23:233–239, 1995.
11. Glencross, D. and E. Thornton. Position sense following joint injury. J. Sports Med. 21:23–27, 1981.
12. Glousman, R., F. Jobe, J. Tibone, D. Moynes, D. Antonelli, and J. Perry. Dynamic electromyographic analysis of the throwing shoulder with glenohumeral instability. J. Bone Joint Surg. 70-A:220–226, 1988.
13. Goodin, D. S. and M. J. Aminoff. The basis of functional role of the late EMG activity in human forearm muscles following wrist displacement. Brain Res. 58:39–47, 1992.
14. Hodges, P. W. and B. H. Bui. A comparison of computer-based methods for the determination of onset of muscle contraction using electromyography. Electroencephalo. Clin. Neurophysiol. 101:511–519, 1996.
15. Houck, J. C., J. J. Singer, and E. Henneman. Adequate stimulus for tendon organs with observations on mechanics of the ankle joint. J. Neurophysiol. 34:1051–1065, 1971.
16. Jobe, F., J. Tibone, J. Perry, and D. Moynes. An EMG analysis of the shoulder throwing and pitching. Am. J. Sports Med. 11:3–5, 1983.
17. Johnston, T. B. The movements of the shoulder joint: a plea for the use of the “plane of the scapula” as the plane of reference for movements occurring at the humero- scapular joint. J. Bone Joint Surg. 25-B:252–260, 1937.
18. Jones, D. The role of shoulder muscles in the control of humeral position (an electromyographic study). Thesis. Cleveland Ohio: Case Western Reserve University, 1970, pp. 1–202.
19. Kandel, E. R. and J. H. Schwartz. Principles of Neural Science. New York: Elsevier Science, 1985, pp. 443–468.
20. Kendall, F. P. and E. K. McCreary. Muscles Testing and Function. Baltimore: Williams & Wilkins, 1983, pp. 108–114.
21. Kibler, W. B., T. J. Chandler, B. P. Livingston, and E. P. Roetert. Shoulder range of motion of elite tennis players: effect of age and years of tournament play. Am. J. Sports Med. 24:279–285, 1996.
22. Konradsen, L. and J. G. Ravin. Ankle instability caused by prolonged peroneal reaction time. Acta Orthop. Scand. 61:388–390, 1990.
23. Lynch, S. A., U. Eklund, D. Gottlieb, P. A. Renstrom, and B. Beynnon. Electromyographic latency changes in the ankle musculature during inversion moments. Am. J. Sports Med. 24:362–369, 1996.
24. Maday, M. C., C. D. Harner, and J. P. Warner. Shoulder Injuries. In: Sports Injuries, Mechanisms, Prevention, Treatment, F. H. Fu and D. A. Stone (Eds.). Baltimore: Williams & Wilkins, 1994, p. 900.
25. McLeod, W. D. The pitching mechanism. In: Injuries in the Throwing Arm, B. Zarins, J. R. Andrews, and W. G. Carson (Eds.). Philadelphia: W. B. Saunders, 1985, pp. 22–29.
26. Newton, R. A. Joint receptor contributions to reflexive and kinesthetic response. Phys. Ther. 62:22–28, 1982.
27. O’Brien, S., M. Neves, and S. Arnoczky. The anatomy and histology of the inferior glenohumeral ligament complex of the shoulder. Am. J. Sports Med. 18:449–456, 1990.
28. Pappas, A. M., R. M. Zawacki, and T. J. Sullivan. Biomechanics of baseball pitching: a preliminary report. Am. J. Sports Med. 13:216–222, 1985.
29. Rathbun, J. B. and I. MacNab. The microvasculature pattern of the rotator cuff. J. Bone Joint Surg. 52-B:540–543, 1970.
30. Saha, A. K. Dynamic stability of the glenohumeral joint. Acta Orthop. Scand. 42:491–505, 1971.
31. Smith, R. L. and J. Brunolli. Shoulder kinesthesia after anterior glenohumeral joint dislocation. Phys. Ther. 69:106–112, 1989.
32. Vakos, J. P., A. J. Nitz, A. J. Threkeld, R. Shapiro, and T. Horn. Electromyographic activity of selected trunk and hip muscles during a squat lift. Spine 19:687–694, 1994.
33. Vallbo, A. B. Afferent discharge from human muscles spindles in noncontracting muscle: steady-state impulse frequency as function of joint angle. Acta Physiol. Scand. 90:303–318, 1974.
34. Wilk, K. E. and C. Arrigo. Current concepts in the rehabilitation of the athletic shoulder. J. Orthop. Sports Phys. Ther. 18:365–375, 1993.
35. Worrell, T. W., T. Corey, S. York, and J. Santiestaban. An analysis of supraspinatus EMG activity and shoulder isometric force development. Med. Sci. Sport Exerc. 24:774–748, 1992.


© 1999 Lippincott Williams & Wilkins, Inc.