The impact of specification structure on human memory performance - Experiences from a first experiment

Abstract

Context and motivation: The major workload in Requirements Engineering lies with those people who create requirements specifications. Inevitably, in doing so people use their memory to store and process related information. Question/problem: This paper examines the question: does the underlying structure of a requirements specification (template vs. prose) have an impact on the memory performance of requirements engineers? Principal ideas/results: We present results from cognitive psychology that support the assumption that template-based specifications lead to better memory performance and present an experiment to test this assumption. Contribution: An initial run of our experiment did not provide sufficient results to support or refute our assumption. In this research preview, we report on the design of experiment, our initial results, and conclusions for future research.