Studying the psychological impact of exercise deprivation: are experimental studies hopeless?

Szabo, Attila

Understanding the physical and psychological consequences of deprivation from regular exercise, in committed exercisers, is an important issue because today exercise is a not only a personal but also a social striving. Researchers who venture in this field of study may become quickly discouraged when they realize how limited is the literature in this area of research. Citation indexes and well compiled computerized data bases are virtually void of studies on the effects of exercise deprivation. What may be the reason for the shortage of work in this field? The answer is simple: People exercise for some sort of benefit(s). No matter what that benefit may be, from a personal perspective it is most often greater than the benefit derived from participation in an exercise-deprivation study. Therefore, highly devoted exercisers are unlikely to enroll in deprivation studies even if there were some alluring incentives involved (Baekeland, 1970).

To date, there are less than ten experimental studies that examined the psychological impact of exercise deprivation and only a little more than ten opportunistic or survey-type studies (Szabo, 1995). But even more disappointing is the fact that most of these studies looked to the problem of exercise deprivation indirectly. More precisely, they examined the concept in relation to "exercise addiction" based on the presumption that negative emotions during episodes of exercise deprivation are reflections of addiction to exercise (e.g., Anshel, 1991). Accordingly, the primary objective in these studies was not the understanding of how or what people feel during intervals of exercise deprivation, but rather whether they are or are not addicted to their exercise (e.g., Anshel, 1991; Gauvin, 1990; Sachs & Pargman, 1979). Therefore, the examination of the effects of exercise deprivation was most often secondary to the analysis of issues related to exercise addiction. The possibility of isolating exercise deprivation from the concept of addiction was stressed only recently (Szabo, 1995; Wingate, 1993). This isolation is necessary in the conceptual foundation of experimental inquiries aimed to the understanding of how regular (not addicted) exercisers feel when they are deprived of their exercise. Once this foundation is established, the realization of experimental studies may present major difficulties.

The Subject-Recruitment Dilemma

As noted above, few people volunteer for exercise deprivation studies. Those who are willing to take part in these studies may be different in some way from those who do not take part in similar investigations for any incentive (Baekeland, 1970). How different? In fact volunteers for exercise deprivation inquiries may represent a totally different population than the population from which they are thought to be. A sturdy example for this assertion stems from a recent experience. Recruitment leaflets calling for participation in an exercise deprivation study were distributed to more than 5000 participants in a large marathon (Gauvin & Szabo, 1996). Only 24 people ([less than]0.5%!) showed interest in participation by replying to the call. In terms of probability, as based on this observation, it may be estimated that less than five marathon runners in a thousand may be interested in volunteering for exercise deprivation studies. But who are those five people? How are they different from the others? Do they possess some unique personality characteristics? Or it is their relationship with their exercise that is unique? Regrettably no answers are available to these questions at this time. One can speculate and hypothesize, but only the empirical testing of these hypotheses will shed light on the questioned differences. While it may be argued that marathoners do not represent the general exercising population and that the above figure may be different in subjects committed to other forms of exercise, the key point that merits attention is commitment. Other exercisers may be as committed to their form of exercise as runners (Gauvin & Szabo, 1992), but they were not studied as intensively as runners (Szabo, 1995). When deprived from their favorite activity, all committed exercisers may experience hardship (Anshel, 1991; Baekeland, 1970; Crossman, Jamieson, & Henderson, 1987; Gauvin, 1990; Gauvin & Szabo, 1992; Wingate, 1993). Therefore, the unwillingness to take part in deprivation studies, observed in runners, may be comparable to the general "exercise-committed" population at least until more exercise-specific data will be available.

To elucidate further on why people are reluctant to take part in exercise deprivation studies, the attitudes of a runners' group were sought on the information highway. A message, asking runners whether or not they would take part in an exercise deprivation study, was posted on the Internet news-group "rec.running". (The procedure was not intended to represent a "scientific" inquiry, but rather to probe into runners' viewpoint and to use their opinion as an aid in the current discussion.) A brief summary of the answers given by the 10 runners who replied to the message is presented in Table 1. Clearly, none of the runners expressed unconditional will for taking part in a deprivation study. One vivid reason for this resistance was phrased by one of the runners: "Life is too short not to do the things you like to do. . ."(1)

Incentives for Participation

Incentives in deprivation studies may play a role. Some people may be attracted with some incentives, especially monetary offers. However, the experimenter(s) goes against the value of basic needs when such incentives are offered. A "hooked" runner (Lewis, 1984; Mickel, 1979) or exerciser may be compared to the "hooked" smoker. These people may be allured by certain incentives and believe that they can comply with the requirements of a behavioral deprivation protocol, but once started their willpower may be shattered by the strong craving for the behavior to which they are hooked. At that point the researcher may be faced with the chance of noncompliance and, unless there are some control measures, the reliability of the data may be shattered. To assign weight to this dilemma a statement made by one of the runners ought to be reproduced. "If by chance, I did say yes, (to participation in a deprivation study) it would be very difficult not to cheat."

It may be argued that people who are not offered an incentive for participation may also become non-compliant. However, these people have almost nothing (i.e., an important incentive) to lose if they are not conforming to the requirements of an experimental protocol. Therefore, they may be less likely to become dishonest, because the incentive-based temptation is not present. Monetary incentives are not as critical as incentives that may bound more strongly, such as grade points offered to university students (e.g., Gauvin & Szabo, 1992) in exchange for their participation. Such incentives may become crucial for some people and assume the a role of a basic need as the behavior for which they are studied. When two basic needs are in conflict the subjects' compliance is extremely unpredictable which shatters the data-reliability.

The Length of Deprivation

As noted in Table 1, apart from monetary reward, the length or the duration of the deprivation period was another condition stated by two runners who showed interest in participating in an exercise deprivation study. Beyond doubt, the longer the deprivation period the harder may be the adherence to the requirements of the inquiry. However, to estimate the consequences of exercise deprivation in real life settings, the deprivation period in experimental studies should mimic the naturally occurring deprivation periods. Clearly, short periods of deprivation, lasting one (Thaxton, 1882) or two days (Tooman, Harris, & Mutrie, 1985), may coincide either with a planned or an opportunistically assimilated rest period. An example for this assertion stems from an inquiry by Crossman et al. (1987). In this study the authors concluded that exercise deprivation had no negative effect on teenagers who had a break from running for one day or from swimming for five days. However, the subjects knew in advance about their training suspension. When the interruption is planned it can no longer be classified as exercise deprivation. Quoting another runner may illustrate this point: "I often schedule time off from running and experience little discomfort from that time off. In fact, it puts running into perspective and keeps other aspects of my life healthier". Therefore, short-duration inquiries have limited value in exercise deprivation literature because they are biased by diurnal variations in mood/affect (Clark, Watson, & Leeka, 1989), planned time-off, and by the arising opportunities to fulfil other obligations or to participate in other activities. But the status of exercise deprivation literature is very deficient if short-duration studies are overlooked. To date, only three studies have adopted an exercise deprivation period that lasted one week or longer (Baekeland, 1970; Gauvin & Szabo, 1992; Morris, Steinberg, Sykes & Salmon, 1990). While it is clear that there is a need for similar inquires, their realization may be hopeless.

Table 1.

Summary of Runners, Feelings About Participation in an
Exercise-deprivation Study.

Runner Gender Willingness
 to participate

1 male definitely not

2 female most likely no

3 male yes, depending on the
 length of deprivation

4 male yes, depending on the
 length of deprivation

5 male yes, depending on the
 season of the year

6 female yes, motivated by
 monetary incentive

7 male yes, depending on the
 season of the year

8 male probably not

9 female not even for money

10 female yes, for monetary
 incentive


Note: The runners were asked whether they would be willing to take
part in an exercise deprivation study.

The Season

In some forms of exercise the season plays an important role. Running is a good example. Season was the condition put forward by two interrogated runners. It goes without explanation that it is easier to commit oneself to a training interruption when the season does not favor an adopted activity. However, the season, chiefly in case of runners, is not only tied to weather but also to the time of public races. In northern regions running races are less frequent in the winter than in the summer. Some people may even take a couple of weeks or months of break from their exercise when both weather and lack of opportunity for challenge decline. These people could become candidates for exercise deprivation studies in that season. However, their participation may be biased by the previously discussed notion of planned versus unplanned training-interruption. Opportunistic studies may take advantage of planned exercise layoff, but these inquiries cannot answer key questions regarding the effects of coercive exercise deprivation.

Mission Impossible?

lt is obvious that recruitment of subjects for experimental studies is a difficult problem. Further, volunteers are most likely different from the non-volunteers for reasons that remain to be tested. Moreover, the difficulties are topped by the fact that the offering of incentives for participation and/or the acceptance of certain conditions put forward by the subjects can distort the answer to the posed question. Currently, there are three possible solutions that may be thought to solve the problem of reliability. However, two of these solutions harden further the recruitment of subjects and one of them may be answering a different question.

The first possible solution for gathering reliable data in exercise deprivation studies is the direct observation of subjects in an isolated environment. However, this method gives rise to a new confound which is the impact of isolation. In institutional settings (i.e., prisons, army, readaptation centers, etc.), where people are habituated to confinement, an overt control for exercise deprivation is possible, but people in these settings do not always represent the general population (Szabo, 1995). Therefore, direct observation strategy remains a theoretical, rather than practical, solution in the area of exercise deprivation.

The second conceivable solution is the employment of Caltrac equipment that gauges caloric expenditure by vertical displacement. The monitoring of caloric expenditure could also control for the substitution of one activity with another, that may be a confound in deprivation studies. However, the obstacle is that the use of such apparatus is costly and these instruments are inefficient in certain activities, such as stationary bicycling, weight lifting, and other types of exercises without vertical displacement. Thus, the use of such device has some but not utter value in assuring the reliability of the experimental data.

Finally, a third possible solution is the granting of certain flexibility (as opposed to rigidity) for subjects with regard to the adherence to the protocol (e.g., Morris et al., 1990; Gauvin & Szabo, 1992). Accordingly, subjects are told that they should try (instead of must) to stop exercising, but if they have an urge for exercising they may do so without negative consequences. In these conditions the subjects record their exercise so that the volume of the actual deprivation can be contrasted with the target volume. Such a "flexible" approach may generate a less coercive atmosphere and also stimulate subjects, adherence, but it will most likely result in the gauging of a reduction in training as opposed to exercise deprivation. Therefore, this "soft" method may be useful to some extent, but it is not the optimal method for answering the question whether forced or involuntary exercise deprivation has an impact on the exercising individual.

It appears that thorough control in experimental studies addressing the issue of exercise deprivation may be futile. This is not to say that experimental studies seem to be useless, but simply to emphasize that most control measures may not resolve the problem of subject-related confounds. These confounds should be recognized and dealt with in the most appropriate manner but their elimination cannot be conceived at this time. Therefore, future experimental studies should be developed in such a way that these confounds are accounted for in advance and highlighted in the data interpretation of the results. Perhaps one of the most puzzling problems, begging for scrutiny, is the description of the participants, in contrast to nonparticipants, in exercise deprivation studies.

The Extra Weight of Descriptive and Opportunistic Research

Descriptive data may not yield accurate results, because of distortions in memory and lack of proper comparison standards, but they are certainly helpful in dealing with the key problem: subject recruitment. Although they are often appraised as being limited in value, descriptive inquiries are and will be important resources of knowledge in exercise deprivation studies. Their utility should not be undermined in this particular field of study. However, carefully formulated opportunistic research may be even more helpful. Chan and Grossman (1988) examined exercise deprivation by using such an approach. These authors studied a group of injured runners whose injuries did not interfere with any other routine activity than running. The latter is of paramount importance in the separation of the observed effects due to injury and due to deprivation from exercise. Similar type of opportunistic studies may be carried out more easily than experimental studies in future. In this specific field of study, in which the performance of experimental research is almost impossible, the value of descriptive and opportunistic research merits some extra weight.

Conclusion

To date, the psychological effect of exercise deprivation has been studied sparsely. The difficulty in subject recruitment is the main cause for the lack of experimental inquiries. Further, participants are likely different, in yet unknown way(s), from nonparticipant and consequently the results of the experimental studies cannot be generalized. Existing control measures are weak for the elimination of confounds in experimental studies. While it sounds pessimistic, non-experimental studies may be the main sources of knowledge for psychologists interested in the effects of exercise deprivation.

Footnote

1 All the quotations are reproduced with permission from runners.

References

Anshel, M. H. (1991). A psycho-behavioral analysis of addicted versus non-addicted male and female exercisers. Journal of Sport Behavior, 14, 145-154.

Baekeland, F. (1970). Exercise deprivation: Sleep and psychological reactions. Archives of General Psychiatry, 22, 365-369.

Chan, C. S., & Grossman, H. Y. (1988). Psychological effects of running loss on consistent runners. Perceptual and Motor Skills, 66, 875-883.

Clark, L. A., Watson, D., & Leeka, J. (1989). Diurnal variations in the positive affect. Motivation and Emotion, 13, 205-234.

Crossman, J., Jamieson, J., & Henderson, L. (1987). Responses of competitive athletes to lay-offs in training: Exercise addiction or psychological relief? Journal of Sport Behavior, 10, 28-38.

Gauvin, L. (1990). An experiential perspective on the motivational features of exercise and lifestyle. Canadian Journal of Sport Sciences, 15, 51-58.

Gauvin, L. & Szabo, A. (1992). Application of the experience sampling method to the study of the effects of exercise withdrawal on well-being. Journal of Sport Exercise Psychology, 14, 361-374.

Gauvin, L. & Szabo, A. (1996). The impact of deprivation from running on the psychological states of marathoners: A daily monitoring study. Manuscript submitted for publication.

Lewis, R. (1984, January). Some athletes let their "fix" take precedence over everything else. Women's Sports, 14-17.

Mickel, H. (1979, April) The true runner becomes hooked on a feeling. Runner's World, 30-31.

Morris, M., Steinberg, H., Sykes, B. A., & Salmon, P. (1990). Effects of temporary withdrawal from regular running. Journal of Psychosomatic Research, 34, 493-500.

Sachs, M. L., & Pargman, D. (1979). Running addiction: A depth interview examination. Journal of Sport Behavior, 2, 143-155.

Szabo, A. (1995). The impact of exercise deprivation on well-being of habitual exercisers. The Australian Journal of Science and Medicine in Sports, 27(3). 68-75.

Thaxton, L. (1982). Physiological and psychological effects of short-term exercise addiction on habitual runners. Journal of Sport Psychology, 4, 73-80.

Tooman, M. E., Harris, D. V., & Mutrie, N. (1985). The effect of running and its deprivation on muscle tension, mood and anxiety. Paper presented at the congress of the International Society for sports Psychology, Copenhagen, Denmark.

Wingate, C. F. (1993). Exploring the karate way of life: Coping, commitment, and psychological well-being among traditional karate participants. Unpublished doctoral dissertation, Temple University, Philadelphia.