School of Psychology, University of Wollongong, Wollongong, New South Wales, Australia.
J Exp Psychol Learn Mem Cogn. 2012 Nov;38(6):1731-40. doi: 10.1037/a0028470. Epub 2012 May 14.
Studies of the effect of word frequency in the serial recall task show that lists of high-frequency words are better recalled than lists of low-frequency words; however, when high- and low-frequency words are alternated within a list, there is no difference in the level of recall for the two types of words, and recall is intermediate between lists of pure frequency. This pattern has been argued to arise from the development of a network of activated long-term representations of list items that support the redintegration of all list items in a nondirectional and nonspecific way. More recently, it has been proposed that the frequency effect might be a product of the coarticulation of items at word boundaries and their influence on rehearsal rather than a consequence of memory representations. The current work examines recall performance in mixed lists of an equal number of high- and low-frequency items arranged in contiguous segments (i.e., HHHLLL and LLLHHH), under quiet and articulatory suppression conditions, to test whether the effect is (a) nondirectional and (b) dependent on articulatory processes. These experiments demonstrate that neither explanation is satisfactory, although the results suggest that the effect is mnemonic. A language-based approach to short-term memory is favored with emphasis on the role of speech production processes at output.
在序列回忆任务中，对单词频率的影响进行的研究表明，高频词列表比低频词列表更容易被回忆；然而，当高、低频词在列表中交替出现时，两种类型的词的回忆水平没有差异，并且回忆处于纯频率列表之间的中间位置。这种模式被认为是由于激活的长期项目网络的发展而产生的，该网络以非定向和非特定的方式支持所有项目的重新整合。最近，有人提出，频率效应可能是词边界处的项目协同发音及其对复述的影响的结果，而不是记忆表征的结果。当前的工作在安静和发音抑制条件下，检查了由相等数量的连续分段（即 HHHLLL 和 LLLHHH）排列的高、低频项目的混合列表中的回忆表现，以测试该效应是否（a）无方向和（b）依赖于发音过程。这些实验表明，这两种解释都不理想，尽管结果表明该效应是记忆的。一种基于语言的短期记忆方法受到青睐，重点是输出处言语产生过程的作用。
