Abstract
Feature models are commonly used to capture and communicate the commonality and variability of features in a Software Product Line. The core component of Feature models is feature diagrams, which graphically depict features in a hierarchical form. In previous work we have proposed a new notation that aims to improve the cognitive effectiveness of feature diagrams.
The objective of this paper is to empirically validate the cognitive effectiveness of the new feature diagrams notation in comparison to its original form.
We use two distinct empirical user-studies to validate the new notation. The first empirical study uses the survey approach while the second study is a subject-based experiment. The survey study investigates the semantic transparency of the new notation while the second study investigates the speed and accuracy of reading the notation.
The results of the studies indicate that the proposed changes have significantly improved its cognitive effectiveness.
The cognitive effectiveness of feature diagrams has been improved, however there remains further research for full acceptance of the new notation by its potential user community.