Semantic and phonological schema influence spoken word learning and overnight consolidation

Abstract

We studied the initial acquisition and overnight consolidation of new spoken words that resemble words in the native language (L1) or in an unfamiliar, non-native language (L2). Spanish-speaking participants learned the spoken forms of novel words in their native language (Spanish) or in a different language (Hungarian), which were paired with pictures of familiar or unfamiliar objects, or no picture. We thereby assessed, in a factorial way, the impact of existing knowledge (schema) on word learning by manipulating both semantic (familiar vs unfamiliar objects) and phonological (L1- vs L2-like novel words) familiarity. Participants were trained and tested with a 12-hr intervening period that included overnight sleep or daytime awake. Our results showed (1) benefits of sleep to recognition memory that were greater for words with L2-like phonology and (2) that learned associations with familiar but not unfamiliar pictures enhanced recognition memory for novel words. Implications for complementary systems accounts of word learning are discussed.

Offline consolidation supersedes prior knowledge benefits in children's (but not adults’) word learning

“…However, it has been demonstrated that overnight recall improvements occur in the absence of repeated retrieval practice (Henderson et al ., ). Furthermore, Havas, Taylor, Vaquero, Diego‐Balaguer, Rodríguez‐Fornells, and Davis () found similar changes in schema benefit to occur over a 12‐hr period containing sleep and not over an equivalent period of wake. This suggests that reductions in schema benefit can be at least partially attributed to offline consolidation processes, given that the wake group will have had identical amounts of retrieval practice.…”

The effect of overnight consolidation in the perceptual learning of non-native tonal contrasts

“…Converging evidence indicates that sleep supports various aspects of language learning by facilitating the memory consolidation of newly learned knowledge [see 1,2 for review]. Sleepmediated memory consolidation (i.e., overnight consolidation) plays an important role in novel word learning by facilitating the learning process through an overnight interval [3][4][5][6][7]. For instance, one study found that only the set of novel words learned in the evening and consolidated during the overnight interval, compared with another set of novel words learned during the daytime without overnight consolidation, showed a lexical competition with existing words and elicited faster responses than those learned during the daytime in the set of assessment tasks [6].…”

The Role of Prior Knowledge in Morphological Learning in an Artificial Second Language

“…It was hypothesized that following consolidation the newly learned words should be able to prime existing words. Havas et al (2018) failed to find evidence of priming for either the L1-similar or L1-dissimilar words. While the results of this study are inconsistent with the hypothesis that similarity between L1 and L2 linguistic forms will facilitates consolidation, the lack of evidence for consolidation 12 hours after training is in line with prior work demonstrating that it can take several days or weeks for newly learned words to show evidence of consolidation (e.g., Gaskell & Dumay, 2003;Merkx et al, 2011).…”

“…At the lowest level, prior knowledge may manifest through shared linguistic forms. While the role of this type of prior knowledge has not been investigated within the context of morphological learning, studies have explored the role of shared phonological forms in L2 word learning (Havas et al, 2018;Nora et al, 2012;Nora et al, 2015). These studies have overall demonstrated better learning for novel words composed of L1-similar phonological forms as compared to L1-dissimilar phonological forms.…”

“…These studies have overall demonstrated better learning for novel words composed of L1-similar phonological forms as compared to L1-dissimilar phonological forms. However, they have either not explored (Nora et al, 2012;Nora et al, 2015) or failed to find evidence (Havas et al, 2018) that this type of L1-L2 similarity facilitates the consolidation process. Havas et al (2018) found that native Spanish speakers learning novel words that were either phonologically similar to Spanish or to an unfamiliar language (Hungarian) were more successful at acquiring the novel words consistent with their L1.…”