Brent S. Rushall, in answer to questions from Dr. Larry Weisenthal (1997).

  1. Physiological changes in one activity are not necessarily beneficial to another. The aerobic adaptations derived from running generally cannot be used effectively in swimming. Only those derived from swimming training can be effectively used in swimming. The specificity of aerobic adaptation is so specific that the great South African physiologist and endurance running guru, Dr. Tim Noakes of Cape Town University, has said:

    "In aerobic endurance training, mitochondrial adaptations occur only in muscles stimulated by activity. The response is further limited to those fibers which are activated in the activity. Thus, white fibers are very unlikely to be stimulated to produce a training response in work that is consistently at or below anaerobic threshold. These adaptations are only specific and do not generalize to other forms of activity that may use the muscle, and therefore muscle fibers, differently. For example, endurance gained from flat-track running does not generalize or facilitate hill running." [This is from an abstract in Coaching Science Abstracts Vol 1(2).]

  2. It also has been clearly shown, but in the strength literature, that excess strength does not produce any more muscular endurance than a training limited amount. In fact, in an abstract posted to the Coaching Science Abstracts excess strength training actually decreased muscular endurance. I use this example to argue by analogy. Having excess in a capacity in one activity does not appear to enhance performance above that level which is specifically needed in another activity.

    Strength training advocates have proposed that excess strength provides a reserve that can be tapped. The research evidence does not support this nor does anecdotal evidence. It is not appropriate (possible?) to create a capacity above that which can be maximally trained in a particular activity.

  3. The value of cross training diminishes proportionally with the increase in performance level of the athlete.

    I could go on at great length about the pitfalls of the human being if it was a generalizing animal instead of one that discriminates. A great biological argument could be made about the value of specificity for species survival which is contradictory to the generalization characteristics proposed in "cross-training." In the history of sport science, the current bent on cross-training is at least the third time that the "generality" of training principle has emerged. Each time (the last was in the early 1960s) it has arisen, research has been regenerated and demonstrated that there are neither general capacities nor general exercise responses. Specificity is one of the strongest response characteristics of exercise adaptation.

  4. The neural representation that exists in the cortex that dominates the training response. Patterns of movements are what is represented in the brain as a result of training. If there are no patterns then an individual has to respond consciously to exercise, that is, think every movement (the cognitive control of movement). That is what happens with beginners. However, for automated movements, a movement pattern has to be represented so that the pattern is evoked without any cognitive interference. Thus, it is not possible for the highest level of performance to select characteristics from other movements and employ them. This is one reason why the "excess capacity" argument does not work.

    For very high levels of performance to occur, movements have to be strongly and specifically represented in the cerebrum. Without sufficient training to produce an overlearned state, automated and smooth movements will not result. This means that only kicking in swimming in an appropriate manner and with lots of repetition will there develop a good kicking pattern. The problem is do the training activities produce sufficient repetitions of "race-type" kicking?

The central feature of all performance is that all movement and movement capacities are neurally controlled. To argue theoretical postulations without including a neural qualifier is a dangerous path.

Return to Table of Contents for this issue.