2018 Sat Poster 6462

Saturday, November 3, 2018 | Poster Session II, Metcalf Small | 3:15pm

Learning words from context in ASL: Evidence from a Human Simulation Paradigm
A. Fitch, S. Arunachalam, A. Lieberman

Across languages, children map arbitrary words to meaning with great efficiency, despite a seemingly unconstrained space of potential mappings. Human Simulation Paradigms (HSP), where adults are asked to map words to meaning from the extralinguistic context alone, have contributed and confirmed some of what we know about how this space is constrained in spoken language. For example, concrete nouns are more easily observed and therefore easier to acquire than words for abstract concepts [1,2]; this is mediated by the caregiver’s input language (e.g. Chinese mothers provide more informative contexts for verb learning than American mothers, even for American participants [3]). Additionally, child eye gaze and caregiver pointing towards word referents makes for more informative language learning episodes [4,5]. Although there are likely to be similarities, it is not clear that these findings translate to sign languages. Modality differences lead to differences in the dynamics of social cues and allow for a degree of iconicity that may enhance the “observability” of particular signs’ meanings. In order to explore potential differences, we conducted an HSP in American Sign Language (ASL).

Twenty-five hearing adults with no background in ASL participated in the study. Participants viewed 48 40-second vignettes extracted from a corpus of video-recorded play sessions between Deaf mothers and their Deaf children (7 dyads, child age range: 21-39 months), interacting in ASL (Figure 1). Thirty seconds through each vignette the participant heard a ‘beep’ sound, which corresponded to a target sign. Target signs were the 24 most frequent nouns and 24 most frequent verbs in the corpus, one vignette per target. After each vignette, participants typed in the English translation of the target sign. Responses were coded for correctness (same lemma), and if incorrect, whether their response was semantically related to and/or from the same syntactic category as the target.

Overall, subjects were correct just 14.86% of the time, in line with prior findings from spoken language. Performance was somewhat better if semantically related incorrect guesses are included—20% of incorrect responses were related words. Responses were more likely to be verbs than nouns (t(24) = 8.41, p <.001), although participants were no more likely to successfully map noun items (M = .11, SD = .16) than verb items (M = .18, SD = .15), p = .13 (Figure 2). Item-wise, performance was positively correlated with iconicity ratings [6], d = .57, p<.001. Noun and verb items did not differ in iconicity (p = .59).

Findings demonstrated that as with spoken language, participants were overall not very successful at mapping word meanings through observation of a single event. However, the successful mappings they did make were just as frequent on noun trials as verb trials. This differs from English [1,2], but is in line with findings from Mandarin [3], implying that caregivers using ASL (and Mandarin) provide a more supportive context for mapping verbs than those using English. We are exploring the mechanism for this in an ongoing study, but the analyses presented here suggest a potential role of iconicity.

References

  1. Gillette, J., Gleitman, H., Gleitman, L., & Lederer, A. (1999). Human simulations of vocabulary learning. Cognition, 73, 135-176.
  2. Piccin, T. B. & Waxman, S. R. (2007). Why nouns trump verbs in word learning: New evidence from children and adults in the Human Simulation Paradigm. Language Learning and Development, 3(4), 295-323.
  3. Snedeker, J., Li, P., & Yuan, S. (2003). Cross-cultural differenes in the input to early word learning. Proceedings of the Annual Meeting of the Cognitive Science Society, 25, 1094-1099.
  4. Frank, M. C., Tenenbaum, J. B., & Fernald, A. (2013). Social and discourse contributions to the determination of reference in cross-situational word learning. Language Learning and Development, 9(1), 1-24.
  5. Medina, T. N., Snedeker, J., Trueswell, J. C., & Gleitman, L. R. (2011). How words can and cannot be learned by observation. Proceedings of the National Academy of Sciences, 108(22), 9014-9019.
  6. Caselli, N. K., Sehyr, Z. S., Cohen-Goldberg, A. M., & Emmorey, K. (2017). ASL-LEX: A lexical database of American Sign Language. Behavior Research Methods, 49(2), 784- 801.