4.2: Ok, New Theories. Or Is that Difficult?

According to the philosopher of science Thomas Kuhn, the development of science can be described as a recurring sequence that unfolds from a phase of so-called normal science in to a phase of extraordinary science (i.e., a novel conceptual or methodological approach, leading to new insights into the phenomena being investigated) into a phase of new normal science (Kuhn,1962/1996). Times of normal science produce research "firmly based upon one or more past scientific achievements, achievements that some particular scientific community acknowledges for a time as supplying the foundation for its further practice" (p. 10). A crisis occurs when existing theories and scientific practices fail to solve the problems that have been defined by these theories and these scientific practices. When a growing number of scientists acknowledge that existing theories provide a poor fit to empirical observations, this marks the beginning of the transition from normal to extraordinary science. While scientific progress during normal phases is entirely possible and can result in evolutionary advances, crises and extraordinary phases can lead to leaps in knowledge that have been characterized as scientific revolutions. It is difficult to achieve this kind of progress,however, because one characteristic of normal science is that it tends to be highly resistant to change.

We claim that exercise psychology today is in a phase of normal science,with clear emerging signs of a crisis, both with regard to theories (Ekkekakis, 2017) and with regard to research practices (Caldwell et al., 2020).While change is afoot and the crisis is brewing, a scientific revolution in a Kuhnian sense is still at an early stage and gaining momentum.

We have already explained in various texts that the continuing insistence on the social cognitive and the humanistic/organismic frameworks(see Figure 4.2) entails a limiting, narrow cognitivist perspective on PA and exercise behavior change (e.g., Brand & Cheval, 2019; Ekkekakis et al., 2019; Ekkekakis & Brand, 2021). In particular, we have argued that the contribution of affective processes has been trivialized or marginalized (e.g., Ekkekakis et al., 2018; Ekkekakis & Brand, 2019). These two points will be reiterated as part of our description of the affective-reflective theory(ART)later in this chapter. Before we do this, however, we will focus on two other points.First, we will highlight the common misconception that results from studies testing theories are directly useful for informing interventions. Second, we will point to a very unfavorable development in the practice of research as a reason why it is so difficult for theories outside of the current conceptual "mainstream"to appear on the research community's radar.

4.2.1. Research Programs and Types of Knowledge Generated

According to the epistemology of science,there are different ways of doing research (Bunge, 1983; Lakatos, 1978).Psychology was described as a discipline in research programs that can be characterized by how the psychologists involved would usually direct their research (Figure 4.3; Herrmann, 1994).Those interested in psychological basic research programs aim to explain, understand, and conceptualize reality as accurately as possible(e.g., decipher the mechanisms governing human thinking or motivation). Psychologists who involve themselves in applied research programs instead aim to provide practitioners with background knowledge about why things are the way they are, and how they may intervene to bring about positive change (e.g., treat a disorder or motivate people to be more physically active).

Figure 4.3. Research Programs and How They Relate to Each Other

Note. Black arrows represent greater impact.

Within basic research programs, theory programs can be distinguished from more phenomenon-oriented programs. Theory programs are aimed to test the validity of a theory. Researchers immersed in this kind of research tend to look for domains in which a defined explanation (theory) can be applied and tested. Researchers immersed in phenomenon-oriented programs, on the other hand, tend to follow research questions such as "What should a theory postulate in order to explain the problem of interest as thoroughly as possible?" Research of this kind focuses on phenomena for which there is presently no adequate theoretical explanation.

In applied research programs, studies are designed to yield a foundation for action. Where one type of applied program primarily aims to provide background knowledge for interventions ("What should you know to better understand what needs to be done in practice?"), another type, which can be described as action programs, tests the intervention techniques themselves ("In order to achieve the result 'x', do 'y' because that is more effective than doing 'z'"). Both forms of applied research can relate to theory, of course. But, unlike basic research, in both types of applied research, the explanation for why something works is less important than the feasibility and efficacy of the solution.

Now, why is it important to know about these distinctions?The reason is that knowledge generated from basic research programs that study the relationships between variables in theories is not sufficient for the scientific substantiation of interventions. This is the kind of knowledge that the vast majority of exercise psychology studies have accumulated thus far,however. What is now required is direct evidence from applied research programs consisting of studies that test the effectiveness of intervention techniques derived from variables defined in the theories.⁹ Knowing that two variables are correlated or make a difference in reality does not automatically imply evidence about the means of changing these variables.

4.2.1.1. In a Nutshell

Although exercise psychology aims to be an applied research discipline, most of its empirical studies have been designed (perhaps unwittingly) as basic research aimed to test relations between variables,as defined in theories.The mantra-like assertion(which can be found in all meta-analyses cited earlier) that empirical evidence on the relations between theoretically defined concepts ensures that actions derived from them will be effective is,in fact,baseless. This is exactly what recent research from applied intervention programs shows.According to meta-analytic findings, whether one of the described mainstream theories is used as a basis makes no difference in the effectiveness of interventions (e.g., McEwan et al., 2019). Reasons that may, at least in part, explain why the handful of theories that have been found to be generally insufficient to achieve meaningful and sustained PA behavior change still dominate the research literature are described in the next section.

4.2.2. Publish or Perish

An increasing number of research articles are being published worldwide each year.Researchers have never before been under so much pressure to demonstrate and continuously prove their talent and performance through publications, with career preservation and advancement dependent upon it (Grimes et al., 2018). According to another quick search of the Web of Science database,the number of published articles per year worldwide first exceeded 1 million in 1991. In the year 2008, the 2 million mark was passed, and in 2016, there were 3 million published research articles per year. Exercise psychology has not escaped this trend. As depicted in Figure 4.2, the number of published articles on exercise and PA motivation has increased exponentially since about 2000. This may be related to the fact that the topic has become increasingly important over the past several years (precisely because of growing societal pressure to develop interventions to effectively promote PA and exercise; Guthold et al., 2018). Another possible reason is that the number of exercise psychologists in academia, who have to publish in order to promote their careers, has increased in a similar way. The problems emanating from this publish-or-perish culture are well described (e.g., Everett & Earp, 2015; Lilienfeld, 2017; Zou et al., 2018).

The remarkable increase in the volume of the research output in exercise psychology should not be interpreted as an indication that most studies are of low quality.Instead,we believe that there are now enough studies demonstrating that the correlations between variables defined in SCT, TPB, SDT,etc.are significantly different from zero,and that study participants with higher values do tend to exhibit more PA.

From our point of view, one reason why researchers have continued to conduct correlational studies in large numbers for too long is that doing intervention studies instead is expensive and time-consuming: intervention content must be defined (and manualized), then delivered by practitioners (researchers are generally trained for different things), and then the possible effects must be monitored over months and years (at least when it comes to providing evidence for sustained behavior change). How much easier(and more rational in terms of promoting one's career)is it to only ask study participants to complete some questionnaires or tests for a few minutes!Researchers are incentivized to do what "works,"especially in times of normal science (see above).

4.2.2.1. In a Nutshell

The increasing number of studies supporting a handful of theories is probably due to the fact that researchers are required to publish a lot in order to meet the constantly increasing (and often outrageous) performance expectations of modern-day academia. It seems quite rational that,under these circumstances,researchers tend to focus on studies that are "quick and easy" to implement and produce articles that can be published just as quickly and easily thanks to the predictably positive results.

4.2.3. Interim Conclusion: Yes, It Is Difficult Indeed.

It is not easy to understand why the few theories of behavior change that are currently preferred in exercise psychology, despite meta-analyses having shown that they are of limited value as the basis of interventions, continue to be the focus of most studies. We interpret this as a sign of exercise psychology being in a state of Kuhnian crisis,perhaps approaching the farewell of a phase of normal science (Kuhn 1962/1996). Only in this way¹⁰ can we rationalize why the same handful of theories continue to be emphasized in review articles and textbooks, leading to assertions that these few theories should be used as the foundation of public-health intervention programs. As we emphasized, such assertions disregard fundamental epistemological principles (recall that knowledge from theory testing in basic research programs cannot be assumed to entail that interventions would be effective).

In such times, with exercise psychology in crisis,it is not easy to convince researchers that it might be necessary to look beyond the well-known "successful"theories (in the paradoxical sense described above) and to seek new ones that offer a broader perspective. Nevertheless, here we present our alternative view. The affective-reflective theory (ART), published only a few years ago (Brand & Ekkekakis, 2018), is already receiving some attention according to citation numbers.¹¹

---

References

⁹ The logical proof behind this claim may be more than what you need to know. Nevertheless, it is still cool to know it, and it can certainly impress your professor: Basic research programs and applied background knowledge programs aim to develop and test nomological statements of the form "if x, then y". For example, with regard to an evidence-based claim of SDT, if there is autonomous motivation (x), then there will be exercising (y). Nomological statements only hint at eventual actions and consequences. Actions have to meet additional conditions for the anticipated consequence to occur:actions must represent operationalizations (x*) of x, and actions x* must cause the effect y. Furthermore, y must be operationalized (y*). Nomopragmatic statements of the form "if x* is done, then y* occurs"are closer to application than nomological statements. However, a nomopragmatic statement never excludes the possibility that there are other actions with which y* can be achieved,in the same way or even better (e.g., "if x° is done, then y* occurs"). This means, that from "if x* is done, then y* occurs,"it can only be concluded that "x* is one possible way of reaching y*." Empirical studies that examine nomological statements in no way provide the scientific evidence that interventions derived from them will be effective.

¹⁰ In addition, fundamental cognitive effects also contribute to the situation(e.g., confirmation bias, or the tendency to interpret new evidence as confirmation of one's existing theories or beliefs).

¹¹ If researchers cite their colleagues' publications, this may imply that they considered material in these texts to be useful or valuable. Citations may reflect—with significant limitations—aspects of scientific impact and relevance.But there is no evidence that citations reflect other important dimensions of research quality(e.g., plausibility and soundness, scientific and societal value,originality; Aksnes et al., 2019).