Speaker
Description
For decades, sports and exercise research has advanced by isolating single variables or physiological signals to study behavior and performance. While informative, it has become increasingly clear that such reductionist methods struggle to capture the complex, dynamic, and context-dependent nature of sports and exercise behavior. Although advances in data analysis, artificial intelligence, and open data standards now enable unprecedented insights into complex datasets, comparatively little attention has been paid to the place where data originate: the laboratory. Here, devices are often siloed, proprietary, or limited to aggregated outputs, thereby constraining the questions a laboratory can address. For laboratory infrastructure to keep pace with advances toward complex systems research, it is helpful to first clarify what a measure represents in sports and exercise psychology. Rooted in the classic work on construct validity and measurement (Cronbach & Meehl, 1955; Van Ede et al., 2014), we think of measurements in three epistemic layers: the surface layer (raw numeric outputs), the proxy layer (physiological or behavioral subsystems), and the target layer (emergent constructs such as arousal, effort, or performance). These layers are epistemic in that they describe how meaning is inferred from signals and the type of losses and mismatches that can occur at or between them. For example, when a researcher seeks to study effort as a target construct, inferential precision hinges on how well the chosen proxy layer covaries with changes in actual effort and how accurately that proxy is digitized into an interpretable numerical output, such as produced force in Newton. Importantly, this framework offers practical implications for how laboratory devices can be selected and evaluated. From it, we derive six practical desiderata to guide infrastructure decisions: acquisition, data access, interoperability, event marking, data transparency, and flexibility. For each category, we propose criteria that differentiate excellent, medium or poor fulfilment of each category’s demands. We believe that adopting (and improving) such a ranking scheme will help communication and decision-making for setting up the best possible laboratory infrastructure to study of complex, dynamic behavior.