Interference effects from grammatically unavailable constituents during sentence processing.

Abstract

Evidence from 3 experiments reveals interference effects from structural relationships that are inconsistent with any grammatical parse of the perceived input. Processing disruption was observed when items occurring between a head and a dependent overlapped with either (or both) syntactic or semantic features of the dependent. Effects of syntactic interference occur in the earliest online measures in the region where the retrieval of a long-distance dependent occurs. Semantic interference effects occur in later online measures at the end of the sentence. Both effects endure in offline comprehension measures, suggesting that interfering items participate in incorrect interpretations that resist reanalysis. The data are discussed in terms of a cue-based retrieval account of parsing, which reconciles the fact that the parser must violate the grammar in order for these interference effects to occur. Broader implications of this research indicate a need for a precise specification of the interface between the parsing mechanism and the memory system that supports language comprehension.Keywords cue-based memory retrieval; interference; complexity effects; sentence processing; constraint-based parsing Interference effects in sentence processing are beginning to be recognized, but the conditions that give rise to these effects are not well understood (Gordon, Hendrick, & Johnson, 2001, 2004Van Dyke & Lewis, 2003;Van Dyke & McElree, 2006). colleagues (2001, 2004) observed that the classic processing advantage for subject-relative clauses over objectrelative clauses was reduced or eliminated when the second noun phrase (NP) in sentences such as 1a and 1b was either a pronoun (you or everyone) or a proper name (Joe). They attributed their result to a reduction in similarity-based interference that occurs when the two NPs have different referential characteristics (i.e., common NPs refer via their description; pronouns and proper names refer directly to objects previously established in the discourse). Thus, on this account, interference effects are due to the presence of NPs with shared referential characteristics.(1a) The banker that praised [the barber/a barber/Joe/you/everyone] climbed the mountain.(1b) The banker that [the barber/a barber/Joe/you/everyone] praised climbed the mountain.Several researchers have suggested an alternative account, one that implicates retrieval as the source of interference effects (Lewis, Vasishth, & Van Dyke, 2006;McElree, Foraker, & Dyer, 2003;Van Dyke, 2002;Van Dyke & Lewis, 2003;Van Dyke & McElree, 2006). Although these are not the first sentence-processing theories to include a retrieval component, previous theories (e.g., Gibson, 1998Gibson, ,2000 have emphasized decay as the source of processing complexity, on Correspondence concerning this article should be addressed to Julie A. Van NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author Manuscript the basis of hypothesized memory demands that certain structures present for comprehenders. In contrast, approaches...

Event-related brain potentials index cue-based retrieval interference during sentence comprehension

“…An effect of the gender of the structurally unavailable noun on the retrieval of the antecedent of otra would constitute evidence for cue-based interference during sentence comprehension. This would be consistent with results from basic memory research that show retrieval interference as a major determinant of retrieval failure (Anderson and Neely, 1996;Keppel and Underwood, 1962;see Nairne, 2002a for a review; Waugh and Norman, 1965), and provide further insights to the major determinants of processing complexity during language comprehension (e.g., Gordon et al, 2001;Lewis et al, 2006;McElree, 2006;McElree et al, 2003;Van Dyke, 2007).…”

“…Here, the syntactic cues at the gap site generate interference in (c) compared to (d) because in (c) both the resident and the warehouse are subjects, while in (d) the warehouse is the object of a preposition. Indeed, example (c) showed reliably slower reading times at the verb following the gap site than (d), suggesting that shared syntactic features of the nouns within the dependency affected processing at the verb (Van Dyke and Lewis, 2003;Van Dyke, 2007).…”

“…Eguren (2010) notes that although agreeing gender morphology on a remnant of ellipsis is not a crucial licensing condition for empty nominals (because some remnants, due to their particular lexical features, do not bare gender, e.g., grande), "nominal ellipsis allows a mismatch in number between the elided element and its antecedent, but requires a match in gender features." Van Dyke, 2007;Van Dyke and Lewis, 2003;Van Dyke and McElree, 2006). To illustrate this paradigm from the example in Table 1, the intervening attractors the feminine falda "skirt" in (1) and the masculine vestido "dress" in (2) have recently been processed when otra is encountered.…”

“…Previous work on cue-based retrieval interference has solely relied on behavioral measures (Lewis et al, 2006;McElree, 2006;Van Dyke, 2007;Van Dyke and McElree, 2011; but see Xiang et al, 2009, for a related ERP study). In our study, we used ERPs to index retrieval operations while participants read sentences for comprehension.…”

Semantic indeterminacy in object relative clauses

“…Our results also provide a way to reconceptualize memory-based approaches that explain comprehension difficulty in terms of distance between noun-verb relations (Gibson, 1998, Grodner et al, 2002Grodner & Gibson, 2005) or similarity-based retrieval at integration points (Lewis and Vasishth, 2005;Van Dyke and Lewis, 2003;Van Dyke, 2007;Gordon et al 2001Gordon et al , 2004. These researchers have viewed long reading times and/or poor comprehension accuracy in object relatives as evidence of memory costs during processing, while we have interpreted these data as stemming from the activation of alternative interpretations during processing.…”