Laboratory courses represent a unique and potentially important component of the undergraduate physics curriculum, which can be designed to allow students to authentically engage with the process of experimental physics. Among other possible benefits, participation in these courses throughout the undergraduate physics curriculum presents an opportunity to develop students' understanding of the nature and importance of experimental physics within the discipline as a whole. Here, we present and compare both a longitudinal and pseudolongitudinal analysis of students' responses to a research-based assessment targeting students' views about experimental physics--the Colorado Learning Attitudes about Science Survey for Experimental Physics (E-CLASS)--across multiple, required lab courses at a single institution. We find that, while pseudolongitudinal averages showed increases in students' E-CLASS scores in each consecutive course, analysis of longitudinal data indicates that this increase was not driven by a cumulative impact of laboratory instruction. Rather, the increase was driven by a selection effect in which students who persisted into higher-level lab courses already had more expertlike beliefs, attitudes, and expectations than their peers when they started the lower-level courses.