Stróżak Paweł, Leynes P Andrew, Taurogiński Kamil
Department of Experimental Psychology, The John Paul II Catholic University of Lublin, Al. Racławickie 14, 20-950 Lublin, Poland.
Department of Psychology, The College of New Jersey, P.O. Box 7718, Ewing, NJ 08628-0718, USA.
Int J Psychophysiol. 2025 Feb;208:112506. doi: 10.1016/j.ijpsycho.2025.112506. Epub 2025 Jan 6.
Perceptual fluency can increase familiarity of some of the items in recognition tests and enhance attributions of these items to the past. It is not clear, however, whether perceptual fluency can influence recognition under conditions promoting recollection-based memory. To this end, we performed a systematic replication of a study by Lucas and Paller (2013) using a letter-segregated method. We recorded ERPs while participants performed recognition task in letter segregated (LS) blocks, in which new words were always composed of different letters than old words, and in letter non-segregated (LNS) blocks, in which half of the new words came from the same letter pool as the studied words (new related words), and the other half came from the other pool (new unrelated words). Unlike the Lucas and Paller study, deep encoding promoted more recollection-based memory. In the LNS blocks, the comparison between old and new unrelated words revealed early (180-260 ms) P200 old/new effect, showing that recognition can be supported by an early discrimination of perceptual differences between studied and unstudied test probes. The relatively large hit rates and relatively high sensitivity measures, as well as the late (500-700 ms) LPC old/new effects in both blocks, indicated high levels of recollection for old words. Still, recognition memory was more accurate in the LS blocks, whereas in the LNS blocks there were more false alarms for new related than for new unrelated words. This suggests that perceptual fluency derived from low-level information may influence not only familiarity, but also recollection-based memory judgments.
知觉流畅性可以提高识别测试中某些项目的熟悉度，并增强对这些项目属于过去的归因。然而，尚不清楚在促进基于回忆的记忆的条件下，知觉流畅性是否会影响识别。为此，我们使用字母分离法对卢卡斯和帕勒（2013年）的一项研究进行了系统的重复。我们记录了参与者在字母分离（LS）块中执行识别任务时的事件相关电位，在该块中，新单词总是由与旧单词不同的字母组成；以及在字母非分离（LNS）块中，一半新单词与学习过的单词来自相同的字母库（新相关单词），另一半来自另一个字母库（新不相关单词）。与卢卡斯和帕勒的研究不同，深度编码促进了更多基于回忆的记忆。在LNS块中，旧单词与新不相关单词之间的比较揭示了早期（180 - 260毫秒）的P200新旧效应，表明对学习过的和未学习过的测试探针之间的知觉差异进行早期辨别可以支持识别。两个块中相对较高的命中率和相对较高的敏感性指标，以及晚期（500 - 700毫秒）的LPC新旧效应，表明对旧单词有较高水平的回忆。尽管如此，在LS块中的识别记忆更准确，而在LNS块中，新相关单词的错误警报比新不相关单词更多。这表明从低水平信息中获得的知觉流畅性可能不仅影响熟悉度，还影响基于回忆的记忆判断。
