2018 Friday Poster 6626

Friday, November 2, 2018 | Poster Session I, Metcalf Small | 3pm

Quantifier Relationships in the Lexicon: Scalar Competence and Performance
J. Grinstead, M. Oates, M. Nieves-Rivera, R. Padilla-Reyes

Children are delayed in rejecting distributive quantifiers in collective contexts and collective quantifiers in distributive contexts until roughly 10 (e.g. Hanlon 1986, Brooks & Braine 1996, Brooks et al. 1998, Pagliarini et al. 2012, Syrett & Musolino 2013, de Koster et al. 2016, 2017). Dotlačil (2010) proposes to account for this gradual development as a delayed emergence of a pragmatic scale of collectivity-distributivity, which we refer to as the Pragmatic Scale Hypothesis. Pagliarini et al. (2012) show that children’s decreasing acceptance of the Italian distributive quantifier ciascun as collective predicts children’s decreasing judgments of the plural definite article i as distributive, as expected on the Pragmatic Scale account, wherein ciascun’s distributive entailment anchors a pragmatic collective-distributive scale, and i derives its collective meaning via implicature. Padilla-Reyes et al. (2017) demonstrate a similar relationship between the distributive quantifier cada and the plural collective quantifiers los and unos. It is possible that executive function abilities, including inhibition, are relevant to drawing pragmatic inferences, including scalar implicatures (e.g. Kapa & Colombo 2014, though cf. Janssens et al. 2014).

Research Questions
Given these findings, the current study seeks to determine: 1) whether similar predictive relationships among distributive and collective quantifiers exist in child English, 2) whether the development of executive function abilities (attention, inhibition and non-verbal working memory, as in Miyake et al. 2000) helps explain the development of the scalar component of collective interpretations in child English and Spanish and 3) whether general lexical development indexes the distributive entailment of cada/each.

Methods
To answer these questions, an SES- and age-matched sample of monolingual Spanish-speaking children in Puerto Rico (n=31, mean age=95.25 months, SD=7.43) and English-speaking children in the Midwestern US (n=29, mean age=95.38 months, SD=7.65) were given a video- recorded Truth Value Judgment Task (TVJT – Crain & McKee 1985). Children were asked to affirm or reject the critical final sentences of each TVJT, which included predicates that occurred with quantifiers that varied by distributivity (distributive: each/cada; collective: the/los, some/unos), crossed with distributive/collective visual video contexts. Children were independently given 3 subtests of the EXAMINER executive function battery (Kramer et al. 2014) and the TVIP/PPVT (Dunn & Lugo 1986) tests of lexical development.

Analysis, Results & Implications
Results showed that: 1) Acceptance of distributive each in collective pragmatic contexts in child English predicted acceptance of both the and some in distributive pragmatic contexts (Figures 1 & 2). 2) Inhibition scores from the Flanker task, possibly relevant to drawing inferences, predicted Spanish-English child judgments of unos/some and los/the, but not cada/each, consistent with Dotlačil’s claim that the collectives derive their meanings via implicatures, while the distributive (cada/each) derives its meaning by entailment. 3) TVIP and PPVT scores were predictive of the acceptance of cada/each. We discuss cada/each as a mediator between general lexical development and the interpretations of unos/some and los/the, which constitute ontologically distinct components of the distributive-collective pragmatic scale.