Reading between the (head)lines: A processing account of article omissions in newspaper headlines and child speech

Register affects language comprehension: ERP evidence from article omission in newspaper headlines

“…e.g., for language acquisition data De Lange, Vasic, & Avrutin, 2009;Gerken, 1991, and for impaired production de Roo, Baauw, Avrutin, & Hofstede, 2002;Ruigendijk, 2002). 1 We sought to close this gap by investigating the comprehension patterns of newspaper headlines -a register that allows article omission to a certain extent -through the recording of event-related brain potentials (ERPs).…”

“…In fact, offline experiments have shown that readers prefer headlines with (some) articles omitted, although such omission is clearly unacceptable in normal, non-headline sentences (cf. De Lange, 2008;De Lange et al, 2009).…”

“…Interestingly, cross-linguistic comparisons of article omission (in production) have revealed a strong correlation between the (morphosyntactic) richness of the article paradigm and the informativity of the article on the one hand and the amount of omissions found in a particular language on the other hand (Caramazza, Miozzo, Costa, Schiller, & Alario, 2001;De Lange et al, 2009;Guasti, Gavarro, de, & Caprin, 2008;Janssen & Caramazza, 2003;Miozzo & Caramazza, 1999). De Lange (De Lange, 2008;De Lange et al, 2009) has coached her findings form newspaper headlines and acquisition data within an information-theoretic approach, where information load and uncertainty within the system (i.e. entropy) determine the amount of omissions.…”

“…Yet, many factors contribute to the richness/informativity of the article paradigm, such as the number of members in a set and the number of different morphosyntactic features represented by the article (gender, number, Case, etc.) (De Lange, 2008, De Lange et al, 2009, phonological features (Caramazza et al, 2001;Gerken, 1991), or semantic features associated with the nominal system (Chierchia, 1998). In the current investigation, we seek to address the more general question of how uncertainty caused by article omission impacts sentence comprehension and whether task-specific expectations (i.e.…”

Optimal encoding! - Information Theory constrains article omission in newspaper headlines

“…to account for the structural asymmetry in article omission observed in Stowell 1996), and on specific constructions (like article omission in the complement of a preposition; see Kiss 2010), but less so on the question why in a given utterance token in a specific context an article is or is not realized. A notable exception is the work by De Lange and colleagues (see for example De Lange 2008, De Lange et al 2009). De Lange and colleagues, however, investigate article omission in newspaper headlines primarily from a typological perspective and relate omission frequencies (on the basis of Information Theory) to the overall complexity of the respective article systems along the following lines: The more complex an article system is, the less predictable is a given article (like German der, die or das, for example); and the less predictable a given article is, the more pressure there is to overtly realize the article.…”