Constructing Alternatives: Evidence for the Early Availability of Contextually Relevant Focus Alternatives

The successful interpretation of focus requires a comprehender to infer the set of alternatives intended by the speaker (Rooth in Association with focus (montague grammar, semantics, only, even). PhD thesis, University of Massachusetts Amherst, 1985; Nat Lang Semant 1:75–116, 1992). Prior studies have endorsed a two-stage model of this process (Gotzner in Alternative sets in language processing: How focus alternatives are represented in the mind. Springer, 2017; Gotzner and Spalek in Lang Linguist Compass 13:e12310, 2019; Husband and Ferreira in Lang Cogn Neurosci 31:217–235, 2016). Under this view, an initial context-insensitive stage of semantic priming provides a second context-sensitive stage with the lexical activation necessary to represent focus alternatives as such. We discuss results from a cross-modal probe recognition task study in English which present a challenge to the two-stage approach. We found that alternative status influenced the speed of recognition for probe words that were not semantically primed by their focus. Counter to the predictions of the two-stage model, this effect was observed immediately after focus was encountered. We take these results to support a single-stage model of alternative selection under which discourseDiscourse representations drive the availability of alternatives independently from semantic priming.