Interdisziplinäres Zentrum für Kognitive Sprachforschung
print

Links und Funktionen
Sprachumschaltung

Navigationspfad


Inhaltsbereich

Abstract Zeschel

Processing collocations: probing pattern competition in comprehension

In usage-based approaches, linguistic knowledge is conceived as a large stock of (more or less complex) stored exemplars of previous linguistic experience and (more or less schematic) abstractions over these that have formed as a by-product of countless individual categorizing events (Langacker 2000, Goldberg 2005, Bybee 2006; see also Bod 2006). Categorization itself is modelled as a process of pattern capture in which several attractors – concrete exemplars as well as more schematic units – compete for activation as the categorizing structure to be invoked for a particular target. Specifically, the selection of categorizing structures is assumed to be driven by three types of factors (Langacker 2000): relative degree of entrenchment (resting activation), contextual priming (context-induced co-activation) and degree of similarity (amount of overlap with the target).

The present paper investigates aspects of the interaction between these factors by studying competition between concrete exemplars (collocational chunks) and entrenched abstractions (constructional schemas) in local syntactic ambiguity resolution. The paper focuses on the widely studied so-called ‘DO/SC-ambiguity’ in which a postverbal NP following strings such as e.g. The scientist knew… is ambiguous between a direct object (= DO, as in The scientist knew the answer to the problem) and the subject of a sentential complement (= SC, as in The scientist knew the answer to the problem was difficult). Previous research has produced a rich body of experimental evidence that speakers use a range of different information sources for resolving such kinds of ambiguity, and there are also connectionist models of the way that the proposed constraints interact in this specific case (Elman et al. 2004). However, previous studies have largely focused on issues of entrenchment and contextual priming alone and devoted little or no attention at all to the ‘overlap’-parameter, i.e. possible influences of lexicogrammatical chunking on speakers’ parsing performance. The present study addresses this issue by investigating the processing of direct object collocations and DO-collocating ‘compound lexical items’ (in the sense of Sinclair 1996) for SC-biased verbs (e.g. She didn’t believe a single word of it was true). From an exemplar-based perspective, such expressions are predicted to cause local processing difficulty because of competition effects (to be investigated in e.g. self-paced reading paradigms) between entrenched collocational chunks and similarly entrenched conflicting higher level schemas associated with the verb lemma (to be read off from corpus distributions). By combining these two empirical perspectives, the study contributes to the discussion of how psycholinguistic designs can be fruitfully combined with corpus-linguistic methodologies in order to approximate the representations that speakers actually work with in on-line comprehension (cf. Roland and Jurafsky 2002, Hare et al. 2004).

 

References

Bod, R. (2006). Exemplar-based syntax: how to get productivity from examples. To appear in The Linguistic Review, 23, Special issue on exemplar-based models of language.

Bybee, J. (2006). From usage to grammar: the mind's response to repetition. To appear in Language 62.

Elman, J., M. Hare and K. McRae. (2004). Cues, constraints and competition in sentence processing. In: M. Tomasello and D. Slobin (eds.), Beyond nature-nurture: Essays in Honor of Elizabeth Bates. Mahwah: Lawrence Erlbaum. 111-138

Goldberg, A. (2005). Constructions at work. The nature of generalization in language. Oxford. OUP.

Hare, M., K. McRae and J. Elman. (2004). Admitting that admitting verb sense into corpus analyses makes sense. Language and cognitive processes 19, 2: 181-224

Langacker, R. W. (2000). A dynamic usage-based model. In: S. Kemmer and M. Barlow (eds), Usage-based models of language. Stanford: CSLI. 1-63

Roland, D. and D. Jurafsky. (2002). Verb sense and verb subcategorization probabilities. In: P. Merlo and S. Stevenson (ed.), The lexical basis of sentence processing: Formal, computational, and experimental issues. Amsterdam: John Benjamins. 303-324.

Sinclair, J. (1996). The search for units of meaning. Textus: English studies in Italy, 9: 75-106