Poster #117 - Novel Verb Learning Via Overheard Speech: A study of Dative Verbs

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Language, Communication

Integrative Statement

Exciting recent studies suggest children can do syntactic bootstrapping from overhearing speech, in the absence of an immediate referent scene. Children distinguish between novel transitive and intransitive verbs via implicit eye gaze measures (Arunachalam & Waxman, 2010; Messenger, Yuan, & Fisher, 2015; Yuan & Fisher, 2009). using the argument structure to hone in on the potential meaning of a novel verb with no referent scene present. The current line of work asks whether children are able to map dative verbs as well, when asked to make an explicit-selection via a touchscreen. The goal was to adapt the paradigm for use in an early language assessment.

Four studies presented a monologue that repeatedly used a novel verb – e.g. “swoof” – in the target syntactic frame across several different sentences (see Table 1). Intransitive frames should be linked to an independent action, transitive frames a causal action, and dative frames a transfer action. Three studies tested N=93 two-year-olds, varying the number of participants in the intransitive scene, whether an avatar spoke or just an abstract display during the monologues. Following the monologue, two separate novel animations of actions appeared on the screen and the children were asked e.g. “Find swoofing”. There were 4 of each novel verb type for a total of 8 items, and the particular novel verb assigned to each frame was balanced across child participants.

In three experiments, children identified transitive novel verbs at a rate at or just above chance. However, matching the intransitive novel verbs to a independent action was either at chance or in some cases below chance (see Figure 1), i.e. the same as transitives. Children slightly preferred transitive causal actions for each verb type regardless of syntactic frame. The mixed results in other work have suggested that children understand that the intransitive frame could represent either an independent action or be a reduced transitive. But the present data suggest that children show a pattern of preference for a causal event for an intransitive.

A fourth study tested with 3-year-olds (N=19), contrasting novel transitive verbs with novel dative verbs for each of which two participants interacted with one another though the dative was distinguished by the transfer of an object. The children were at chance in distinguishing between transitive and dative novel verbs. The results of the four studies together are consistent: young children seem to be attracted to scenes in which two participants are interacting and ignore the syntax.

The current studies using the overhearing paradigm do make higher demands than eyegaze: not only was an explicit choice required, but children were tested on more than one type of verb, rather than being assigned to a particular frame condition. It seems that argument structure alone is not enough for children to pick up the contrastive meaning of different novel verbs from overhearing, at least not sufficiently to make an explicit and consistent response. This may limit the paradigm from being used in language assessments.

Authors

• Emily Jackson, Smith College
  Presenting Author

• Jill de Villiers, Smith College
  Non-Presenting Author