Why does the phonological similarity effect reverse with nonwords?

Abstract

The detrimental phonological similarity effect (PSE), a robust finding in serial recall of words, sometimes reverses with nonwords. The current study tested the hypothesis that nonwords benefit from phonological similarity because they are harder to retrieve. In two experiments serial recall and serial reconstruction of visually presented words and nonwords were compared. Phonological similarity is known to have a positive effect on item memory and a negative effect on position accuracy in serial recall, and the demands on item retrieval were greatly reduced in the latter task. PSE occurred for words in both tasks and was reversed for nonwords in serial recall, but not in serial reconstruction-a new finding in the literature. The following conclusions can be made: (1) the detrimental PSE on order retrieval occurs irrespective of lexicality, in accordance with prominent short-term memory models; and (2), the positive PSE on item retrieval is crucially affected by lexicality, a finding less well explained by the existing models.

Short article: Visual similarity at encoding and retrieval in an item recognition task

“…However, when retrieval is held constant, the two conditions in which similar items are presented at encoding have obtained higher scores than those in which items are dissimilar. Therefore, the results are partly consistent with those studies that have focused on item memory regardless of order in the verbal domain, in which the similarity effect reverses and thus facilitates performance (Fallon et al, 1999;Fournet et al, 2003;Gathercole et al, 1982;Karlsen et al, 2007). In the verbal domain, the most widely accepted explanation is that phonological similarity in terms of rhyme or shared phonemes provides an efficient retrieval cue (Fallon et al, 1999).…”

“…With regard to similarity in the verbal domain, phonological similarity refers to the robust, replicated finding that similar-sounding items are less accurately recalled in immediate serial recall tasks than are dissimilar-sounding items (Conrad & Hull, 1964). However, several further studies have indicated that the effect reverses for item identity (as opposed to order) in immediate serial recall tasks when the words of the similar list rhyme (Fallon, Groves, & Tehan, 1999;Gathercole, Gardiner, & Gregg, 1982) and also when nonwords are used (Karlsen, Gravir, Johannessen, Endestad, & Lian, 2007). A positive similarity effect has also been found in free recall of words (Fournet, Juphard, Monnier, & Roulin, 2003).…”

Interference between maintenance and processing in working memory: The effect of item–distractor similarity in complex span.

“…When memory items all share the same rhyme, recall of the correct items (regardless of order) is typically better than item recall for dissimilar lists (Fallón, Groves, & Tehan, 1999;Gupta, Lipinski, & Aktunc, 2005;Karlsen, Imenes, Johannessen, Endestad, & Lian, 2007). However, a potentially problematic consequence of manipulating similarity by rhymes is that this could invite specific coding strategies.…”

“…Experiments manipulating interitem similarity in serial recall have uncovered qualitative differences between rhyming stimuli (e.g., “mat, hat, fat, rat”) and stimuli that are phonologically similar without all sharing the same rhyme (e.g., “mat, man, cat, can”). When memory items all share the same rhyme, recall of the correct items (regardless of order) is typically better than item recall for dissimilar lists (Fallon, Groves, & Tehan, 1999; Gupta, Lipinski, & Aktunc, 2005; Karlsen, Imenes, Johannessen, Endestad, & Lian, 2007). Under some conditions, even recall in correct order has been found to be superior in rhyming lists compared to dissimilar lists (Gupta et al, 2005; Karlsen et al, 2007).…”

“…When memory items all share the same rhyme, recall of the correct items (regardless of order) is typically better than item recall for dissimilar lists (Fallon, Groves, & Tehan, 1999; Gupta, Lipinski, & Aktunc, 2005; Karlsen, Imenes, Johannessen, Endestad, & Lian, 2007). Under some conditions, even recall in correct order has been found to be superior in rhyming lists compared to dissimilar lists (Gupta et al, 2005; Karlsen et al, 2007). No such advantages of similar over dissimilar lists on recall in correct position have been observed when similarity has been manipulated in a more varied fashion.…”