Abstract:  Fixed-links modeling (FLM) is theory-guided statistical modeling for the purpose of analyzing experimental data in the framework of structural equation modeling (SEM). The main characteristic of a fixed-links model is that the loadings of the manifest variables on the latent ones are pre-fixed according to theory-based expectations. The fixation of loadings enables the decomposition of variances and covariances originating from experimental data into several parts each representing a unique source of variances. The loadings are frequently constrained according to functions that reflect the relationship between treatment levels and underlying cognitive processes. FLM has been widely used in representing processes underlying working memory, attention, and learning. It also plays an important part in investigating the relationships among these abilities and their sources.

Key words:  fixed-links modeling, experimental manipulation, loading setting, cognitive process