The similarities (and familiarities) of pseudowords and extremely high-frequency words: Examining a familiarity-based explanation of the pseudoword effect.

Abstract

The pseudoword effect is the finding that pseudowords (i.e., rare words or pronounceable nonwords) give rise to more hits and false alarms than words. Using the retrieving effectively from memory (REM) model of recognition memory, we tested a familiarity-based account of the pseudoword effect: Specifically, the pseudoword effect arises because pseudowords lack distinctive semantic meanings. Because semantics can differentiate orthographically similar words (e.g., horse vs. house), by lacking distinctive semantics, pseudowords have greater interitem similarity than words, and hence more familiarity, which gives rise to the pseudoword effect. Across two sets of simulations, we demonstrate that this account explains the pseudoword effect in addition to accounting for why the pseudoword effect is absent when irregular nonwords are compared with words. Furthermore, our modeling efforts suggest a novel experiment that leads us to the discovery of a new concordant effect. Namely, extremely high-frequency words behave like pseudowords (giving rise to more hits and false alarms than high-frequency words) and also have less distinctive semantics than high-frequency words. We conclude that our work provides strong evidence in favor of the familiarity-based accounts of the pseudoword effect. We discuss the implications of our research with regard to various issues surrounding the pseudoword effect and REM model.

Behavioral and ERP evidence of word and pseudoword superiority effects in 7- and 11-year-olds

“…In this paradigm and across variants, participants are more accurate in identifying the correct letter when the briefly presented string is a word (e.g., DARK ) than if it is a nonword (e.g., RDKA, Adams, 1979; Estes and Brunn, 1987; Ferraro and Chastain, 1993; Johnston and McClelland, 1974; Juola et al, 1974; Krueger, 1992; Prinzmetal, 1992; Williams et al, 1985). A similar pseudoword superiority effect, such that accuracy is higher for letters embedded in pseudowords (e.g., DARL ) than in nonwords, has also been reported in adults and may be due to the word-likeness of pseudowords (e.g., Estes and Brunn, 1987; Grainger and Jacobs, 1994; Massol et al, 2011; Ozubko and Joordens, 2011). Because the letter strings in Reicher-Wheeler-type paradigms are presented briefly and masked, these sorts of paradigms can be used to index automatic orthographic processing.…”

Overdistribution illusions: Categorical judgments produce them, confidence ratings reduce them.

“…Concerning word frequency, it is well established that recognition memory is better for low- than for high-frequency words (Hall, 1979). This appears to be a semantic-processing effect that occurs because the semantic content of low-frequency words receives more processing attention than that of high-frequency words (e.g., Estes & Maddox, 2002; Ozubko & Joordens, 2011). The implication for our research is that subjects will be more likely to access verbatim traces of List 2 targets than List 1 target, and they will be more likely to access gist traces of low- than of high-frequency targets.…”

Elicited Imitation: A Test for All Learners?

“…The motivation for this was that pseudowords are by definition unknown to all participants, irrespective of their background (Keuleers & Brysbaert, 2010b). While it has been observed that pronounceable pseudowords exhibit similar recall qualities as very rare real-word items (Ozubko & Joordens, 2011), it is possible that EI repetition of unknown lexical items that may yet carry meaning associations differs from the repetition of pseudowords.…”