Missing the Function: How Experimental Practices Undermine Theory by Obscuring Functional Relationships Between Independent and Dependent Variables

Experimental psychology often aims to identify causal effects of metric constructs—such as affect, motivation, or cognitive processing styles—by manipulating them at only two or three levels. This practice discretizes inherently continuous constructs, hindering the ability to test functional relationships between independent and dependent variables. It reflects the dominant mode of verbal theorizing in psychology, which identifies relevant variables and predicts effect direction but rarely specifies the functional form of those relationships. We argue that this limitation is sustained by a vicious circle between imprecise theory and impoverished experimental methods: experiments rarely test functional forms because theories rarely specify them, and theories rarely specify them because experiments include too few levels of the independent variable to reveal them. To break this circle, we propose that experimental designs systematically vary manipulation intensity to uncover functional relationships—an approach that may also help address challenges in replication and effect heterogeneity. We emphasize the importance of manipulation checks and outline analytic tools for investigating functional relationships. Embedding functional considerations into experimental design can advance the development of more testable and generalizable theories.