Paying attention to attention: Perceptual priming effects on word order

ABSTRACT Two experiments are reported which examine how manipulations of visual attention affect adult speakers' linguistic choices regarding word order and verb use when describing simple visual scenes. Participants in Experiment 1 were presented with scenes designed to elicit the use of one of two perspective verbs (e.g., "A dog is chasing a man"/"A man is running from a dog"). Speakers' visual attention was manipulated by preceding the display with a crosshair positioned on one or the other character. Cross-hair position affected word order and verb choice in the expected direction. Experiment 2 replicated this effect with a subliminal attention-capture cue, and results were further extended to the order within conjoined noun phrases in sentential subjects ("A cat and dog are growling…"). The findings have important implications for incremental theories of sentence planning and suggest some specifics for how joint-attention might serve as a useful cue to children learning verbs.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Thesis (Ph. D.)--University of Rochester. Dept. of Brain and Cognitive Sciences, 2008. This dissertation investigates predictions from the Givenness Hierarchy and similar theories that associate different levels of discourse entities’ salience with different anaphoric forms. The experimental methodology utilizes the visual world paradigm to assess referent preferences of the three chosen anaphoric forms (indefinite, ‘a(n) N’; definite, ‘the N’; and demonstrative, ‘that N’) under conditions of developing discourse. The discourse entities’ salience is manipulated through the introduction of varying degrees of linguistic material between the antecedent and the subsequent reference. Participants followed auditory instructions to click with a computer mouse on various objects on the screen, allowing us to analyze both the online reference resolution process (eye gaze fixation proportions) as well as their final referent choices. We show that only discourse-relevant information affects discourse entities’ salience, that different referential forms react differently to such salience changes, and that the initial level of salience affects this sensitivity to change. Overall, our findings are problematic for approaches that use a single dimension, such as salience, to associate different referring expressions with their preferred referents. The framework that best accommodates our results is the form-specific multiple-constraints approach advocated by Kaiser and colleagues, which proposes that while an entity’s salience is important for determining the appropriate reference, individual anaphoric forms carry their own specific constraints on the conditions of their felicitous use. Experiments 1 and 2 establish the methodology and examine the change in the preferred referent for different referential forms under changing discourse conditions. We also examine how the initial salience of a discourse entity affects reference resolution by having the discourse introduce a potential referent as either the Theme (Experiment 1) or the Goal (Experiment 2). The results demonstrate that different anaphoric forms are affected differently by the dynamically changing discourse conditions. The initial preference for all anaphors was for the salient discourse mentioned entity, with increasing intervening material facilitating acceptance of the low-salience unmentioned entity as the intended referent. This effect was the strongest for indefinites and the weakest for demonstratives. Experiment 3 investigates possible reasons for the preference of indefinites for the mentioned referents. The experiment evaluated the hypothesis that an indefinite prefers to refer to one of several low-salience potential referents by creating a situation where the set of possible referents includes two identical low-salience items. The results of this experiment largely replicate the results from Experiments 1 and 2, which suggests that the indefinite form is interpreted in line with an ‘any’ interpretation rather than in line with the Givenness Hierarchy predictions. In Experiment 4 we investigate the same questions with more structured, schematic scenes, a narrative discourse, and a scene verification task. We use dynamic scene changes during the narrative description to manipulate the alternatives that participants had to consider in making their response. The results confirm earlier findings – the indefinite form resolved equally well to mentioned and non-mentioned referents, while the demonstrative reference did not accept a non-mentioned entity as a possible referent. We also investigate the possibility of evaluating the relative salience of discourse entities directly by using a change blindness paradigm. This manipulation suggests the feasibility of monitoring discourse representations through non-linguistic means.
  • Source
    Article: Hard words.
    [Show abstract] [Hide abstract]
    ABSTRACT: How do children acquire the meaning of words? And why are words such as know harder for learners to acquire than words such as dog or jump? We suggest that the chief limiting factor in acquiring the vocabulary of natural languages consists not in over- coming conceptual difficulties with abstract word meanings but rather in mapping these meanings onto their corresponding lexical forms. This opening premise of our position, while controversial, is shared with some prior approaches. The present dis- cussion moves forward from there to a detailed proposal for how the mapping problem for the lexicon is solved, as well as a presentation of experimental findings that support this account. We describe an overlapping series of steps through which novices move in representing the lexical forms and phrase structures of the exposure language, a proba- bilistic multiple-cue learning process known as syntactic bootstrapping. The machin- ery is set in motion by word-to-world pairing, a procedure available to novices from the
    British dental journal 07/1976; 140(11):387-8. · 0.81 Impact Factor
  • Source

Full-text (2 Sources)

Available from
May 20, 2014