Adaptive optimal stimulus selection in cognitive models using a model averaging approach

Stimulus selection based on the maximum Fisher information (MFI) principle enables efficient estimation of participant parameters cognitive models with elaborated experimental tasks. However, in typical applications, a single model is assumed to be true, whereby uncertainty ignored, causing poor performance MFI-based stimulus selection. To address this problem, study proposes model-averaged MFI stimulus-selection method that simultaneously considers multiple averaging framework. Three simulation studies were conducted investigate proposed method. The results indicate performed as well ideal case which true known, conventional under misspecified performing worse. Thus, they demonstrate superiority approach, because unknown reality. robust selection, leading high efficiency parameter estimation.