Department of Cognitive Sciences, University of California, Irvine, 92697-5100, USA.
Psychol Rev. 2013 Jan;120(1):293-6. doi: 10.1037/a0030971.
The scale-invariant memory, perception, and learning (SIMPLE) model developed by Brown, Neath, and Chater (2007) formalizes the theoretical idea that scale invariance is an important organizing principle across numerous cognitive domains and has made an influential contribution to the literature dealing with modeling human memory. In the context of free recall data, however, there is a previously unreported conceptual error in the specification of the SIMPLE model. We show that the error matters not only in theory but also in practice by reapplying the corrected SIMPLE model to the benchmark data reported by Murdock (1962). The corrected model makes different predictions about serial position curves, shows better fit to the Murdock (1962) data, and infers different parameters that require substantively different psychological interpretation.
布朗、尼思和恰特（2007 年）提出的不变量记忆、感知和学习（SIMPLE）模型形式化了一个理论观点，即不变量是许多认知领域的重要组织原则，并对处理人类记忆建模的文献做出了有影响力的贡献。然而，在自由回忆数据的背景下，SIMPLE 模型的规范中存在一个以前未报告的概念性错误。我们通过将修正后的 SIMPLE 模型重新应用于默多克（1962 年）报告的基准数据，表明该错误不仅在理论上而且在实践中都很重要。修正后的模型对序列位置曲线做出了不同的预测，对默多克（1962 年）的数据拟合得更好，并推断出需要实质性不同心理解释的不同参数。
