2018 Friday Poster 6463

Friday, November 2, 2018 | Poster Session I, Metcalf Small | 3pm

Do spatial metaphors facilitate word learning?
A. Starr, M. Srinivasan

Across languages, spatial words are commonly used metaphorically to describe more abstract domains, like time (long movie) or pitch (high note), but the reverse is relatively rare. By some accounts, this linguistic asymmetry reflects an asymmetry in cognition: we co-opt spatial language to describe other domains because we rely asymmetrically on space to reason about these other, less tangible domains (1). Alternatively, this asymmetry in language need not reflect a conceptual asymmetry, but could instead be explained by the demands of word learning. For example, it might be more difficult to learn a new adjective that refers to the pitch of a sound— as opposed to the spatial properties of an object—because pitch is more fleeting in experience and more difficult to indicate ostensively. However, if a child has first learned the spatial meaning of a new word, this could help them guess the word’s meaning when it is used in more abstract contexts. For example, a child could guess that a word that has labeled high (or low) spatial positions is also likely to label high (or low) sound frequencies, on the basis of pre- existing and symmetric mental associations between pitch and space (2). Thus, spatial metaphor may be prevalent across languages because it facilitates the acquisition of abstract word meanings, making the lexicon more learnable.

To evaluate these proposals, we taught 3- to 6-year-old English-learning children (N=152) a novel adjective in the domain of space or pitch, evaluated children’s comprehension of the novel adjective in the trained dimension, and then tested their ability to extend the novel adjective to the other, untrained dimension (Figure 1). Through this approach, we assessed potential asymmetries in both word learning (if it is easier to learn words for space than pitch in training) and cross-domain extension (if extending from space-to-sound is easier than the reverse).

Additionally, to test whether children’s cross-domain extension depends on their experience with metaphorical language, we manipulated whether the novel adjective mapped onto a familiar English space-pitch metaphor (daxy = “high” or “low”; Height condition) or an unfamiliar metaphor (daxy = “thick” or “thin”, which label space and pitch in Turkish and Farsi; Thickness condition).

In training, children demonstrated an advantage for learning adjectives that apply to space relative to pitch, and the ability to learn adjectives for pitch improved with age (Figure 2). However, among children who successfully learned the novel adjectives, we found no asymmetry in cross-domain extension: children were equally proficient in extending from space to pitch and from pitch to space, and extension consistency improved with age. We also found that metaphor familiarity did not affect children’s performance. These results suggest that constraints on word learning, namely the relative difficulty of learning words for abstract concepts such as pitch, may be sufficient to explain asymmetries in metaphorical language, without the need to appeal to asymmetries in the mental representations of these concepts. Our results suggest that spatial metaphor may facilitate word learning (3) by scaffolding the acquisition of abstract word meanings.

References

1. Lakoff G. & Johnson M. (1980). Metaphors we live by. Chicago: The University of Chicago Press.
2. Walker, P., Bremner, J. G., Mason, U., Spring, J., Mattock, K., Slater, A., & Johnson, S. (2010). Preverbal infants’ sensitivity to synaesthetic cross-modality correspondences. Psychological Science, 21(1), 21-25.
3. Srinivasan M, Rabagliati H (2015) How concepts and conventions structure the lexicon: Cross-linguistic evidence from polysemy. Lingua, 157, 124-152.