Journal of Applied Physiology
Vol. 81, No. 5, pp. 2046-2052, November 1996
EXERCISE AND MUSCLE

Coordination, the determinant of velocity specificity?

Bjørn Almåsbakk and Jan Hoff

Department of Sport Sciences, Norwegian University of Science and Technology, 7055 Dragvoll, Norway

ABSTRACT

INTRODUCTION

METHODS

RESULTS

DISCUSSION

FOOTNOTES

REFERENCES

ABSTRACT

Almåsbakk, Bjørn, and Jan Hoff. Coordination, the determinant of velocity specificity? J. Appl. Physiol. 80(5): 2046-2052, 1996.Initial strength gains were examined in the context of learning a new skill. Forty female volunteers were randomly assigned to one of four groups: a bench-press training group utilizing heavy loads in its training, a bench-press training group utilizing almost no load, an alternative training group using different exercises, or a control group that did not train. Training period was 6 wk, with three training sessions per week. Emphasis was put on keeping the coordination and muscular adaptation demands in the bench-press groups as invariant as possible. Bench-press training with a light or with a heavy weight was shown to be equally effective in improving the maximal velocity of contraction for a given absolute resistance. Mean velocity with loads of 0.37, 6.6, 16.6, and 20 kg improved by 21.1, 15.8, 16.9, and 19.5%, respectively. No significant differences in the percent improvement at the four different loads were apparent, indicating that no significant velocity-specific adaptations were present. The bench-press training group, utilizing heavy loads in its training, was the only group with improved one repetition maximum. Overall, findings point to the development of coordination as the determining factor in early velocity-specific strength gains.

strength training; force-velocity relationship; maximum dynamic strength; women

INTRODUCTION

STUDIES OF RESISTANCE TRAINING have shown specific adaptations of muscular force depending on the training program applied (26). Velocity-specific adaptations to resistance training are a major subject of interest to the researcher, the coach, and the athlete alike. One consensus of opinion is that velocity specificity (despite a lack of agreement over what constitutes high and low velocity) is the way to produce optimal strength and power improvement at a given test or performance speed and that the further the velocity of a movement deviates from the trained velocity, the less effective the training will be (5, 8, 14, 15, 19).

That this hypothesis is not universal and that even within that consensus opinions differ as to the mechanisms underlying velocity-specific adaptations is attested to by the literature. A number of studies, for example, have shown that an increase in maximal dynamic muscle strength [one repetition maximum (1 RM)] in a well-defined movement is followed by an increase in the velocity of the same movement (4, 6, 7, 27, 31). Although these studies point to the possibility of increasing movement speeds by using heavy-resistance training, they do not discuss this possibility from a movement or velocity specificity of training point of view. Voigt and Klausen (30) showed that even though heavy-resistance training by itself does not improve the speed of a skilled unloaded movement, it does enhance the gain in movement speed if it is combined with specific training.

Training with heavy loads, which, in turn, implies low velocity, primarily affects the high-force part of the force-velocity curve, whereas training with light weights and high velocities primarily affects the high-velocity portion of the curve (15, 16). Petersen et al. (22) found, in contrast, that resistance training at high velocity seems to be no more effective in increasing high-velocity gains in performance than does training at low velocity. Similarly, Doherty and Campagna (9) found that previously untrained subjects exhibited similar increases in maximal force production at both high and low velocities regardless of the velocity at which the training was carried out. Two studies (5, 14) showed that training at high velocity improves performance only at high velocities, whereas training at low velocity results in an improvement at all test velocities. This is not, however, consistent with the results of Moffroid and Whipple (19) and Coyle et al. (8), who showed an improvement at all test velocities after training at high velocities, whereas training at low velocities only improved performance at the same or a slightly higher test velocity. Behm and Sale (2) demonstrated the importance of the instruction given to subjects. It was the intended rather than the actual movement speed that created a high-velocity adaptation. Their findings point to this as a possible explanation for the divergent findings of the previously mentioned studies. The two studies showing general velocity improvements (9, 22) specifically state that their subjects were instructed to carry out the contractions as fast as possible.

Although the majority of studies support some kind of velocity-specific adaptation, there is enough controversy to warrant approaching the question from another angle. A resistance-training program will increase the generation of force within a muscle or muscle group (1). A mechanism responsible for a velocity-specific effect is unknown. However, one can limit the number of possibilities to two broad velocity-specific adaptations: 1) within the muscle, altering its force-velocity characteristics, or 2) within the nervous system, altering the recruitment pattern (23).

In strength training, the increase in voluntary neural drive accounts for the largest proportion of the initial strength increment, and, thereafter, both neural adaptation and hypertrophy take place with further increases in strength, with hypertrophy becoming the dominant factor (12, 20, 21). A velocity-specific adaptation within the muscle seems an unlikely explanation for the results of the previously discussed studies.

In a movement requiring a large amount of force, the timing of the muscle contraction is important. The ability of the nervous system to activate agonists, synergists, and antagonists in synergy is fundamental to the force produced. During the first weeks of a training program, an improvement in the ability to activate and coordinate contraction of the muscles involved in the movement trained has been shown to be the important factor (24). A possible explanation for velocity-specific adaptations might, therefore, be that training programs reflect the acquisition of skill and that training improves such activation at the trained velocity (13, 23). Siegel and Davis (28) demonstrated that when learning a novel skill, the greatest improvement is at the velocity used in training. Further support for this concept is the finding that control groups also show small improvements at all tested speeds (19, 28), effects of mental practice on strength (32), and strength-training effects on the contralateral limb (11, 20, 21).

No conclusive evidence has been reported from any study of velocity- or load-specific adaptations, which would exclude adaptations within the muscle. It is probable that an increase in voluntary neural drive is the mechanism behind this initial strength improvement. On this basis, the postulate of Rutherford (23) and Jones et al. (13) that training improves the subject's ability to activate muscle groups at the trained velocity would appear to be a reasonable explanation for velocity- and load-specific adaptations. Thus the results of studies on velocity-specific adaptations have to be seen in the context of the learning of a new skill. If, therefore, performance demands in terms of coordination and activation of muscles are kept as invariant as possible, it should be possible to produce equivalent effects on the force-velocity curve with the same training program but with loads from different zones of the curve. This postulate is explored in the experiment to be reported.

METHODS

Table 1. Physical characteristics Group n Age, yr Height, cm Weight, kg Lifting height, cm Con 9 21.6 ± 3.0  166 ± 4  64.5 ± 15.0  45 ± 3  Alt 8 22.0 ± 3.0  170 ± 8  66.8 ± 3.2  46 ± 4  BPH 10 20.8 ± 1.0  168 ± 4  64.9 ± 10.5  46 ± 3  BPL 9 22.2 ± 2.5  166 ± 5  63.2 ± 5.0  44 ± 3 Values are means ± SD; n, no. of subjects. Con, control group; Alt, alternative training of involved muscles; BPH, bench press with heavy load; BPL, bench press with light load.

RESULTS

Table 2. Mean velocity for concentric part of bench press and 1-RM pre- and posttest scores for BPH, BPL, Alt, and Con groups together with combined score for BPH and BPL groups Group n v0 v6 v16 v20 1 RM Pre Post Pre Post Pre Post Pre Post Pre Post BPH 10 1.39 ± 0.25  1.64 ± 0.20  1.01 ± 0.17  1.17 ± 0.19  0.73 ± 0.13  0.82 ± 0.09  0.63 ± 0.13  0.72 ± 0.10  31.0 ± 5.7  39.0 ± 6.0  BPL 9 1.37 ± 0.20  1.66 ± 0.26  0.94 ± 0.18  1.07 ± 0.12  0.58 ± 0.14  0.69 ± 0.14  0.47 ± 0.13  0.56 ± 0.11  23.5 ± 4.7  28.3 ± 4.0  Alt 8 1.61 ± 0.32  1.65 ± 0.28  1.01 ± 0.09  1.07 ± 0.11  0.71 ± 0.10  0.74 ± 0.07  0.58 ± 0.12  0.63 ± 0.09  27.5 ± 6.0  33.1 ± 4.7  Con 9 1.51 ± 0.19  1.48 ± 0.21  1.04 ± 0.11  1.01 ± 0.06  0.71 ± 0.14  0.76 ± 0.12  0.62 ± 0.15  0.64 ± 0.10  31.1 ± 9.3  35.0 ± 9.3  BPC 19 1.38 ± 0.23  1.62 ± 0.19  0.98 ± 0.18  1.12 ± 0.17  0.66 ± 0.15  0.75 ± 0.13  0.55 ± 0.15  0.64 ± 0.13  27.5 ± 6.4  34.0 ± 7.4 Values are means ± SD; n, no. of subjects. v0, v6, v16, and v20, mean velocity (in m/s) for concentric part of bench press at 0.37-, 6.6-, 16.6-, and 20-kg load, respectively; 1 RM, 1 repetition maximum (in kg); Pre and Post, before and after testing; BPC, combined BPH and BPL groups.

DISCUSSION

The two bench-press groups, BPH group training with progressive loads at 80-85% of 1 RM and BPL group training with a wooden stick weighing 0.37 kg, showed equivalent effects (no significant group × time interaction; P = 0.671) on the force-velocity curve. This finding points to coordination as being the determining factor in early velocity-specific strength gains. This finding is in line with Rutherford and Jones (24) and provides empirical support for the suggestions of Rutherford (23) and Jones et al. (13) that training programs reflect the acquisition of skill and that training improves such activation at the trained velocity. The improvements in movement speed are also consistent with previous reports (4, 6, 7, 27, 31). If the velocity of training were to be defined in terms of the actual velocity of the training movement, then the BPL group would be classified as a high-velocity training group, whereas the BPH group would be classified as a low-velocity training group. Mean velocity with loads of 0.37, 6.6, 16.6, and 20 kg improved by 21.1, 15.8, 16.9, and 19.5%, respectively. No velocity-specific training response was found. This is counter to some of the previously reported results (e.g., Refs. 2, 15, 19) but is supportive of others (9, 22).

The absence of a velocity-specific training effect can possibly be explained in the terms of the invariance in coordination and muscular activation demanded by the two bench-press training groups despite the variance in the loads to be lifted. In typical training regimens with heavy loads or with an isokinetic apparatus, little or no emphasis is put on a rapid rate of force development or explosive movements. One performance demand with which the subjects had to cope was that of carrying out the concentric part of the movement as fast as possible. This was a constraint previously imposed by Behm and Sale (2), who found, in contrast to the present finding, a high-velocity-specific training effect. A possible explanation of the divergent findings is to be found in the instructions given to the subjects. Whereas, in the present experiment, emphasis was placed on the learning of the whole movement (the bench press), in their study, Behm and Sale placed emphasis mainly on the initiation phase of the movement. By using a dynamic exercise such as the bench press, subjects were able to accelerate the bar for most of the movement duration, whereas in isokinetic exercises, the movement velocity is kept constant. In this way, subjects were able to explore and exploit the coordination dynamics implicit in the subject-environment interaction, thereby discovering a more optimal solution to the problem. This exploration could lead to the development of more efficient coordination and activation patterns within the nervous system, which could be a more dominating factor than any adaptations due to the speeds at which the movement was carried out. Furthermore, the present study was a between-subjects design that included a control group, whereas Behm and Sale trained the legs of their subjects under two different training regimens, thereby precluding the control of a previously reported transfer of training across limbs (21). Support for the interpretation of the findings presented here is provided in two studies that also failed to find a velocity-specific training effect, namely those of Petersen et al. (22) and Doherty and Campagna (9). Both of these studies put emphasis on the fact that the subjects were required to carry out the exercise demands of the training with the highest possible intensity. Moreover, the study of Petersen et al. (22) used isokinetic training, which would seem to indicate that the muscle action explanation for a non-velocity-specific finding is not a valid one.

In contrast to the bench-press training groups, the Alt group did not show any significant improvement on the force-velocity curve (see Fig. 4). Given the interpretation of the results of the bench-press groups being presented here, this is not too surprising because the subjects in this group trained with a very different movement pattern (25, 26). An increase in voluntary neural drive accounts for the best part of initial strength gains (12, 20, 21). An increase in the voluntary neural drive in the Alt group does not lead to improvement in the bench-press tests. This could strengthen the importance of neural adaptations, and specifically the learning of coordination patterns, in strength training. Improvement in the bench-press tests in the Alt group would probably have been due to morphological adaptations within the muscle that, in principle, could be transferred to different movement patterns.

Although the percent improvement in rate of force development variables for the BPC group was rather large, it was not significant. Previous work had reported increases in the rate of force development after strength training (2, 12). The rate of force development data were gathered by an isometric test, whereas the training involved concentric and eccentric movements. Thorstensson et al. (29) found that this way of training led to relatively small improvements in isometric force. A lack of increase in rate of force development isometrically after dynamic training has been found in a number of studies and has been attributed to training-mode specificity (e.g., Ref. 10). A variable training effect on the rate of force development within the training groups was also found in the present study, which is a returning problem in many strength-training studies. The results indicate that repeating the test by itself gives better performance, with an improvement in the Con group of 10 and 18% (Figs. 6 and 7). However, no definitive explanation for the lack of significant improvement in the rate of force development can be stated.

The difference in improvement between the BPH (8.0 kg) and BPL (4.8 kg) groups in maximal dynamic strength shows that even in the early stages of a strength-training program, the use of heavy resistance (in this case 80-85% of 1 RM) is necessary to develop maximal strength. The BPH group showed the expected development of maximal dynamic strength (3). A number of studies have studied the effect on maximal strength when different loads were utilized with the same training program and have shown differing improvements [see Ref. 1 for a review]. The results of the present study indicate that with regard to improvements in 1 RM, the use of heavy resistance is a more dominant factor than coordination improvements alone, even in the early stages of training.

In the introduction to this study, it was stressed that the results of studies on velocity adaptation had to be seen in the context of the learning of a new skill, i.e., velocity-specific adaptations have to do with the learning of a velocity-specific skill. Siegel and Davis (28) demonstrated that, when learning a new skill, the greatest training improvement is brought about at the velocity used in training. The novelty introduced in the present experiment to explore this proposition further was the use of two groups utilizing the same coordination patterns but training with markedly different loads ranging from negligible to high. The results confirmed that there was no difference between the two groups at submaximal performance (the velocity variables), but there was a significant difference at maximal performance. This provides positive support for the contention here that initial strength gains are brought about by the developing coordinative structure, whereas the heavier training weight utilized by the BPH group gave it the necessary advantage at maximal performance. This would be in line with the reported improvements in control groups in the present study and in the studies by Moffroid and Whipple (19) and Siegel and Davies (28).

The findings presented here have significant implications for both the coach and athlete. Maximal advantage would be gained if movements were to be trained with a high resistance and rapid action. This would, given time, have the included advantage (or in some sports perhaps disadvantage) of muscular growth (17, 18). Hypertrophy would also enhance performance at a range of velocities.

The results clearly show the importance of the instruction given to the subjects and clearly underline the proposal that nervous adaptations are critically dependent on the way the muscles are activated by the nervous system. The proposition put forward here that improvement in activation and coordination is the dominant factor in velocity-specific strength gains needs to be explored in other contexts.

FOOTNOTES

Address reprint requests to B. Almåsbakk (E-mail: balm{at}alfa.avh.unit.no).

Received 7 June 1995; accepted in final form 24 June 1996.

REFERENCES