Sign Language Promotes Object Categorization in Young Hearing Infants

• In Event: Infants’ learning about object properties and categories in diverse environments

Abstract

Language exerts a powerful effect on cognition from infancy. By 3 months, listening to spoken language boosts object categorization, implicating the important role of language in early concept learning (e.g., Ferry, Hespos & Waxman, 2010). Here we ask: How broad is this initial link between language and cognition? Are the effects of language specific to the auditory modality, or might it be sufficiently broad to include language presented in the visual modality as well? Specifically, we ask observing sign language, like listening to spoken language, promotes object categorization in hearing infants.
113 hearing infants (4.0-6.99 months) participated. Participants were video-recorded while sitting on their care-givers’ laps and shown images projected onto a large screen. See Figure 1. During Familiarization, infants watched as a woman presented 8 distinct members of an object category (e.g., 8 fish), in sequence. In the Sign Language condition, she looked and pointed to the object while signing the phrase “Look at the Modi, do you see the Modi?” in infant-directed American Sign Language (ASL). In the Non-Linguistic Control condition, she looked and pointed to the object but provided no ASL. Next, during Test, infants viewed 2 images simultaneously: a novel member of the now-familiar category (e.g., a new fish) and a member of a novel category (e.g., a dinosaur). Infants’ gaze was coded. We predicted that if infants formed an object category, they would exhibit a preference for the novel exemplar at test (novelty preference).
Results reveal that infants’ earliest link to cognition is indeed sufficiently abstract to include sign language, but that this link is short-lived (see Figure 2). We observed a significant interaction between age and condition (β = .08, SE=0.03, p=.01). Four-month-olds in the Sign Language condition displayed a significant novelty preference (t(19) = 3.01, p =0.007). This demonstrates that infants’ earliest link between language and core cognitive capacities is sufficiently broad to include not only spoken, but also signed language. Moreover, we find that the cognitive advantages of observing sign language rapidly wanes: Five- and six- month- in the Sign Language Condition performed at chance levels (ts<1, ps>.36). This implicates the critical role of the language environment in shaping the range of signals infants link to cognition. Importantly, the advantages conferred by observing ASL at 4-months are driven by its linguistic properties: 4-month-olds in the Non-Linguistic control condition performed at chance (t(19) = .37, p=.71), significantly lower than those in the Sign Language condition t(38) = 1.99, p= 0.05). Finally, although the effect of age within the Non-Linguistic Control condition was not significant, we did observe that the oldest infants (6-month-olds) performed above chance (t(15) = 2.27, p= 0.04). This raises the possibility that infants’ responses to extralinguistic communicative cues like pointing and eye gaze emerges within this developmental period, although more research with older infants is needed. Together, these results provide new insight into the breadth and specificity of infants’ precocious language-cognition link.

Authors

• Miriam A Novack, Northwestern University
  Presenting Author

• Susan Goldin-Meadow, University of Chicago
  Non-Presenting Author

• Diane Brentari, University of Chicago
  Non-Presenting Author

• Sandra R Waxman, Northwestern University
  Non-Presenting Author