Longitudinal assessment is a type of assessment involving repeated measures over a period to evaluate whether and when an attribute (e.g., ability, skill) changes. Thus, change detection is of central interest in longitudinal assessment. In the assessment setting, change in the desired direction (typically upward) is often referred to as "learning". While many methods have been proposed to detect change, they all require a prior definition of a changing point. However, this information is often unknown in practice. As an alternative, we focus on a family of tests based on stochastic processes of case-wise derivatives of the likelihood function (i.e., scores). These score-based tests could detect "learning" without prior information of a changing point and signal the changing point to the users. In this article, we will illustrate what the score-based tests are and the novel application by using the data of two physicians participating in longitudinal assessment for a medical specialty certifying board's continuous certification program.