Prediction is Production: The missing link between language production and comprehension

Abstract

Language comprehension often involves the generation of predictions. It has been hypothesized that such prediction-for-comprehension entails actual language production. Recent studies provided evidence that the production system is recruited during language comprehension, but the link between production and prediction during comprehension remains hypothetical. Here, we tested this hypothesis by comparing prediction during sentence comprehension (primary task) in participants having the production system either available or not (non-verbal versus verbal secondary task). In the primary task, sentences containing an expected or unexpected target noun-phrase were presented during electroencephalography recording. Prediction, measured as the magnitude of the N400 effect elicited by the article (expected versus unexpected), was hindered only when the production system was taxed during sentence context reading. The present study provides the first direct evidence that the availability of the speech production system is necessary for generating lexical prediction during sentence comprehension. Furthermore, these important results provide an explanation for the recruitment of language production during comprehension.Recent studies have provided evidence for a potential role of production processes in language comprehension [1][2][3][4] , but what exactly is the link between production and comprehension is a central topic in language sciences and remains to be determined 5 . Based on several recent frameworks, this missing link could be prediction. Listeners constantly predict upcoming information during language comprehension (to facilitate comprehension and dialogue), and such predictions are accompanied by covert production [6][7][8][9] . Except some indirect support in the literature showing that language production skills (category fluency task, production vocabulary) and prediction are related 10-14 , there is no direct evidence so far that the production system is necessary for prediction during comprehension (see 15 for review). Providing experimental evidence for this claim would generate important knowledge regarding the production-comprehension link and consequently improve our understanding of the neurocognitive foundations of human communication.In the present study, we measured lexical prediction during sentence comprehension when taxing the production system. The rationale was that if production is mandatory for prediction, then prediction should vanish when the availability of the production system is reduced. This was done by preventing inner speech through articulatory suppression (AS; e.g., uttering a certain syllable repeatedly while completing the primary task 16 ). If performance on the primary verbal task relies on inner speech, it should be significantly impaired by AS because articulation of irrelevant information prevents subvocal rehearsal of the verbal input 17 . Lexical prediction can be measured through event-related potential (ERP) responses derived from electrophysiological recording during...

The Role of Awareness and Cognitive Aptitudes in L2 Predictive Language Processing

“…Prediction as a feature of language processing is not uncontroversial; its precise nature and its role in language processing has been debated extensively (for excellent discussions, see Huettig, ; Kaan ; Kuperberg & Jaeger, ; Van Petten & Luka, ). One of the issues concerns the mechanisms that generate predictions (e.g., Foucart, Ruiz‐Tada, & Costa, ; Martin, Branzi, & Bar, ), which, according to Huettig (), may actually result from several different mechanisms that may include association and priming, the engagement of the production system in comprehension, and event‐simulation processes. Another issue concerns the inevitability of prediction in information processing (Kuperberg & Jaeger, ; Van Petten & Luka, ).…”

Definitely saw it coming? An ERP study on the role of article gender and definiteness in predictive processing

“…We defined additional dependent variables for anterior electrodes and for the subsequent time-window (500-700 ms) to capture later activity like extended N400 effects, the PNP or P600 (DeLong, Quante, & Kutas, 2014;Nieuwland et al, 2019;Van Petten & Luka, 2012). We predicted that article gender-mismatch would elicit enhanced N400 amplitude compared to gender-match, like the patterns observed in Spanish and Italian Foucart et al, 2014;Martin et al, 2018;Molinaro et al). This would be consistent with preactivation of the noun, but would not suffice to conclude participants predicted article form.…”

“…This variability may signal something meaningful like crosslinguistic differences or differences associated with specific methodological choices, it may signal random fluctuations (noise), or an unknown mix of the above (for discussion, see Ito et al, 2017c). Almost all the studies with Romance languages such as Spanish, Catalan or Italian report N400 effects (e.g., Foucart et al, 2014;Martin et al, 2018;Molinaro et al, 2014). There is only one Spanish study reporting a P600 effect (Wicha et al, 2004), which, to our knowledge, has not yet been replicated.…”

“…Arguably the strongest evidence for word anticipation comes from pre-nominal manipulations, which measure behavioral or neural responses to an article or adjective appearing before a noun (for review, see Kutas, DeLong & Smith, 2011;Van Berkum, 2009). Most studies of this type use gender-marking of pre-nominal articles, such as in Spanish and Dutch, and report differential event-related potential (ERP) responses to articles that mismatch the gender of a highly predictable noun, compared to gender-matching articles (e.g., for Dutch, Otten and Van Berkum, 2009;Van Berkum et al, 2005; for Spanish, Foucart, Martin, Moreno & Costa, 2014;Gianelli & Molinaro, 2018;Martin, Branzi & Bar, 2018;Molinaro, Gianelle, Caffarra & Martin, 2017;Wicha, Bates, Moreno & Kutas, 2003;, 2004. Of particular relevance is a study by Otten and Van Berkum (2009), wherein participants read twosentence mini-stories that contained an article-adjectives-noun combination of which the noun was either predictable (e.g., "de verfijnde maar toch opvallende ketting", thecom sophisticated yet striking necklacecom) or not predictable and of a different gender than the predictable noun (e.g., "het verfijnde maar toch opvallende collier", theneu sophisticated yet striking collarneu).…”

“…Recent studies with written Spanish sentences show predominantly N400-like effects in relation to gender-mismatching articles, i.e. enhanced negativities in the typical N400 time window 1 (Foucart et al, 2014;Martin et al, 2018;Molinaro et al, 2014).…”