Children's syntactic parsing and sentence comprehension with a degraded auditory signal

Abstract

During sentence comprehension, young children anticipate syntactic structures using early-arriving words and have difficulties revising incorrect predictions using late-arriving words. However, nearly all work to date has focused on syntactic parsing in idealized speech environments, and little is known about how children's strategies for predicting and revising meanings are affected by signal degradation. This study compares comprehension of active and passive sentences in natural and vocoded speech. In a word-interpretation task, 5-year-olds inferred the meanings of novel words in sentences that (1) encouraged agent-first predictions (e.g., The blicket is eating the seal implies The blicket is the agent), (2) required revising predictions (e.g., The blicket is eaten by the seal implies The blicket is the theme), or (3) weakened predictions by placing familiar nouns in sentence-initial position (e.g., The seal is eating/eaten by the blicket). When novel words promoted agent-first predictions, children misinterpreted passives as actives, and errors increased with vocoded compared to natural speech. However, when familiar words were sentence-initial that weakened agent-first predictions, children accurately interpreted passives, with no signal-degradation effects. This demonstrates that signal quality interacts with interpretive processes during sentence comprehension, and the impacts of speech degradation are greatest when late-arriving information conflicts with predictions.

FIG. 1.

(Color online): In the word-interpretation task, a sample display featuring the familiar object (e.g., seal), likely agent (e.g., large, novel creature), and likely theme (e.g., small, novel creature) in critical trials. In sentences, the identity of the novel word is disambiguated at the late-arriving verb morphology, which is underlined in sentences. The correct target refers to the identity of the novel word based on disambiguating cues in the construction.

FIG. 2.

(Color online): Eye-movements after the disambiguating sentence cue in the natural (panel A) and vocoded speech (panel B) conditions. Correct referents are likely agents in the novel-NP1/active condition (upper left of each panel A and B), likely themes in the novel-NP1/passive condition (lower left of each panel A and B), likely themes in the familiar-NP1/active condition (upper right of each panel A and B), and likely agents in the familiar-NP1/passive condition (lower right of each panel A and B).

FIG. 3.

(Color online): Preference scores of eye-movements from the disambiguating sentence cue to sentence offset. Positive scores indicate correct fixations in passive trials and negative scores indicate correct fixations in active trials. Dashed lines highlight the fixation differences that result from hearing active and passive sentences. Steeper lines indicate greater sensitivity to disambiguating sentence cues.

FIG. 4.

(Color online): For actions after sentence completion, responses coded based on accuracy and preference scores. For accuracy (panel A), higher values correspond to more correct actions. Since there are two novel objects, the dash line highlights chance performance in interpreting the meaning of the novel word. For preference score (panel B), positive values correspond to correct actions in passive trials and negative scores correspond to correct actions in active trials. Dashed lines highlight the differences in fixations that result from hearing active and passive sentences. Steeper lines indicate greater sensitivity to disambiguating cues.