Continuous Time Models

Continuous Time Models

While most psychological processes develop continuously over time, we need to rely on discrete measurement occasions to infer them. The goal is thus to reconstruct the mechanisms underlying a continuously unfolding process, such as human development, based on few discrete snapshots in time. Continuous-time models are well suited to achieve this goal. In contrast to popular discrete-time methods, continuous-time models link discrete-time observations to underlying continuous-time parameters by stochastic differential equations. This may not only remove bias due to variability in sampling time and improve comparability across different research designs, it also yields valuable information about the nature of change.

Predictions based on previous observations

Observations (red circles) are error prone, our knowledge of the underlying systems is often highly uncertain, so predictions of the future (solid red and shading) must take both elements of uncertainty into account.


Integrating Within-Person and Between-Person Information in the Search for Causal Mechanisms

Most empirical research in psychology is based on analyzing between-person variation. In contrast, most applied psychology is concerned with variation within individuals. In addition, the mechanisms specified by psychological theories generally operate within, rather than across, individuals. This disconnect between research practice, applied demands, and psychological theories constitutes a major threat to the conceptual integrity of the field. We are working on reconciling these extreme positions, both conceptually and methodologically.


COGITO
In the COGITO study, 101 younger adults (20–31 years of age) and 103 older adults (65–80 years of age) participated in 100 daily sessions in which they worked on cognitive tasks measuring perceptual speed, episodic memory, and working memory, as well as various self-report measures. You can find detailed information on this here. more

Selected Publications

Brandmaier, A. M., Driver, C. C., & Voelkle, M. C. (2018). Recursive partitioning in continuous time analysis. In K. van Montfort, J. H. L. Oud, & M. C. Voelkle (Eds.), Continuous time modeling in the behavioral and related sciences (pp. 259–282). Springer.
Driver, C. C., & Voelkle, M. C. (2018). Hierarchical Bayesian continuous time dynamic modeling. Psychological Methods, 23(4), 774–799. https://doi.org/10.1037/met0000168
Driver, C. C., & Voelkle, M. C. (2018). Understanding the time course of interventions with continuous time dynamic models. In K. van Montfort, J. H. L. Oud, & M. C. Voelkle (Eds.), Continuous time modeling in the behavioral and related sciences (pp. 79–109). Springer.
Voelkle, M. C., Gische, C., Driver, C. C., & Lindenberger, U. (2018). The role of time in the quest for understanding psychological mechanisms. Multivariate Behavioral Research, 53(6), 782–805. https://doi.org/10.1080/00273171.2018.1496813
Driver, C. C., Oud, J. H. L., & Voelkle, M. C. (2017). Continuous time structural equation modeling with R Package ctsem. Journal of Statistical Software, 77(5). https://doi.org/10.18637/jss.v077.i05
Go to Editor View