In light of the limitations of nomothetic research, efforts to investigate within-, rather, than between-person, variation have increased. This shift has motivated the collection for intensive longitudinal data (ILD). Through ecological momentary assessment (EMA), a method for collecting ILD, researcher’s hope to gain better insight into how psychological processes interact and evolve over time. However, we argue that the intensive longitudinal framework, like many other psychological methods, suffers from a fundamental weakness: a lack of formal theory. 

In our study, we present a formal model of psychopathology as a bistable dynamical system. Through simulation, we use the model to demonstrate that knowledge of the quantitative structure of the relationships between psychological variables is necessary to gain meaningful insight from the quantitative methods used to investigate psychological phenomena. We do not argue that a perfect model of a psychological process is necessary, but that even a general one is valuable because it can reveal gaps in our understanding of what participants, sampling schedules, and statistical models are required for our studies. We conclude that, as long as we allow methodological conventions to determine what we can observe, we cannot expect to improve our theoretical understanding of psychological processes.