Poster #108 - Does Word Order Constrain Cross-situational Semantic Learning?

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Language, Communication

Integrative Statement

Children develop their phonological, semantic, and syntactic knowledge simultaneously. However, there are few examinations of the potential interdependencies between these learning tasks. The present study examined whether the development of semantic knowledge in the context of cross-situational learning is influenced by the organization of the words. Specifically, we examined 2.5-years-old English-learning toddlers’ learning of nouns and verbs involving dynamic scenes of novel objects performing novel actions.

To examine the role of word order we included three learning conditions. In the noun-verb condition, as in English, the object label preceded the action label. In the verb-noun condition, the order of the object and action labels was reversed. Finally, in the flexible word order condition, word order did not provide a cue to the category of the referent. Children may learn particularly well when the word order corresponds to the one they have experience with, i.e., noun-verb. If not experience but the additional regularity available with word order supports learning, children should perform equally well in the noun-verb and verb-noun conditions and better than in the flexible word order condition.

Two novel objects and two novel actions were used as stimuli. They were combined to create four animations, each showing one of the objects doing one of the actions. In familiarization, participants watched a video which consisted of the four animations repeated eight times in random order. The action and object labels were presented along with each animation. Two languages were used by exchanging the words referring to objects and actions.

Testing was administered 5 min after familiarization. Two animations were simultaneously presented during each test trial along with a test word. When the word was a noun, children saw the two objects perform the same action. When the word was a verb, they saw identical objects perform the same action. There were two blocks of test trials with each word presented once.

The looking data were examined separately for noun and verb test trials given their different structure (n = 76). On verb trials, children looked at the target above chance and there was no difference between conditions. On noun trials, children in the noun-verb and flexible word order condition oriented to the target before children in the verb-noun condition.

Altogether, these findings suggest that verb learning is largely unaffected by the organization of words. Word order may impact noun learning but neither the experience nor the regularity hypothesis accounts for the better performance in the flexible than the verb-noun condition here. Further research is required to examine whether learning is affected by the amount and complexity of the input. Further research could also clarify whether the abstraction of word order information (rather than just its presence) matters.

Authors

• Ge Li, Jilin University
  Presenting Author

• Maya Albin, Queen's University
  Non-Presenting Author

• Stanka A. Fitneva, Queen's University
  Non-Presenting Author