視覚探索処理のWeibull-Gaussianモデルによる推定

抄録

In this article, we provided a mathematical model for analyzing the distribution of reaction time and estimated which types of visual search processes would be observed, such as memory-driven or memory-free types mentioned in Horowitz and Wolfe (2003). Our model was a convolution of the Weibull and Gaussian distributions (Weibull-Gaussian model). We conducted visual search experiments and model fittings for the obtained data. The model comparison revealed that the Weibull-Gaussian model was more appropriate than the ex-Gaussian model. Applying the model to the data revealed that the parameter indicating the type of process changed within a session. These results indicated that the visual search processes were not fixed in one type.