THE IMPACT OF EXPLANATION FACILITIES ON USER ACCEPTANCE OF EXPERT-SYSTEMS ADVICE

Providing explanations for recommended actions is deemed one of the mo st important capabilities of expert systems (ES). There is little empi rical evidence, however, that explanation facilities indeed influence user confidence in, and acceptance of, ES-based decisions and recommen dations. This paper investigates the impact of ES explanations on chan ges in user beliefs toward ES-generated conclusions. Grounded on a the oretical model of argument, three alternative types of ES explanations -trace, justification, and strategy-were provided in a simulated diagn ostic expert system performing auditing tasks. Twenty practicing audit ors evaluated the outputs of the system in a laboratory setting. The r esults indicate that explanation facilities can make ES-generated advi ce more acceptable to users and that justification is the most effecti ve type of explanation to bring about changes in user attitudes toward the system. These findings are expected to be generalizable to applic ation domains that exhibit similar characteristics to those of auditin g: domains in which decision making tends to be judgmental and yet hig hly consequential, and the correctness or validity of such decisions c annot be readily verified.