Adaptive design for systems factorial technology experiments

摘要

Systems factorial technology (SFT) is a powerful framework for examining how people use multiple sources of information together. Unfortunately, it is often difficult to apply. Appropriate manipulation of the salience of each source of information is critical to assessing processing characteristics so a significant amount of time can be spent piloting to determine the correct levels. Even with piloting, some participants' data ends up unusable due to individual differences. We first examine the use of an accuracy-only adaptation for stimulus levels, based on the Psi method. In some cases a focus entirely on accuracy may be insufficient, particularly given that response time (RT) is the primary measure with SFT. Hence, we also introduce an approach to adapting stimulus levels for each individual participant's joint accuracy and RT. This will increase the likelihood that salience manipulations will be effective and that a participant's data will be usable. (C) 2019 Elsevier Inc. All rights reserved.