Maddox W Todd, Ing A David, Lauritzen J Scott
Department of Psychology, University of Texas, 1 University Station A8000, Austin, TX 78712, USA.
Percept Psychophys. 2006 Oct;68(7):1176-90. doi: 10.3758/bf03193719.
Two experiments were conducted that examined information integration and rule-based category learning, using stimuli that contained auditory and visual information. The results suggest that it is easier to perceptually integrate information within these sensory modalities than across modalities. Conversely, it is easier to perform a disjunctive rule-based task when information comes from different sensory modalities, rather than from the same modality. Quantitative model-based analyses suggested that the information integration deficit for across-modality stimulus dimensions was due to an increase in the use of hypothesis-testing strategies to solve the task and to an increase in random responding. The modeling also suggested that the across-modality advantage for disjunctive, rule-based category learning was due to a greater reliance on disjunctive hypothesis-testing strategies, as opposed to unidimensional hypothesis-testing strategies and random responding.
进行了两项实验，这些实验使用包含听觉和视觉信息的刺激来检验信息整合和基于规则的类别学习。结果表明，在这些感觉模态内进行信息的感知整合比跨模态整合更容易。相反，当信息来自不同的感觉模态而非相同模态时，执行基于析取规则的任务会更容易。基于定量模型的分析表明，跨模态刺激维度的信息整合缺陷是由于为解决任务而增加使用假设检验策略以及随机反应增加所致。建模还表明，基于析取规则的类别学习的跨模态优势是由于更依赖析取假设检验策略，而非一维假设检验策略和随机反应。
