screenshot: upassoc.org
Laura Faulkner has published early results of her unpublished doctoral dissertation “Structured Software Usability Evaluation: An Experiment in Evaluation Design”. Thus it appears that breaks improve the results of heuristic evaluation.
“Early results of a study being performed by the Applied Research Laboratories, The University of Texas at Austin (ARL:UT), indicate that simply taking breaks may be the most significant contributor to improved performance of the usability evaluator (3). The study took apart the two aspects of the sHEM study approach, looking at performance using variations that introduce structure alone (smaller sets of criteria coupled with taking breaks), detailed heuristics alone, and the two in combination. With two-thirds of the data in, the clear front-runner so far for number of known usability problems found (4) is the one in which the evaluators simply took breaks, otherwise using their regular approach to a heuristic evaluation.
The moral of this story? You may find yourself more effective in performing heuristic or expert evaluations if you divide them into sections and take breaks between to refresh your thinking. The approach used in the studies was to divide the heuristics themselves into sets. Using those smaller sets of criteria per session, work for an hour, then get away for 30 minutes or more, whether on a complete break or simply to another type of task. You may find that each time you begin it is with the freshness of mind of a whole new evaluation.”
