You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster A41, Wednesday, November 8, 10:30 – 11:45 am, Harborview and Loch Raven Ballrooms

Furry hippos & scaly sharks: blind individuals’ knowledge of animal appearance

Judy Sein Kim1, Giulia Elli1, Marina Bedny1;1Johns Hopkins University

We learn about the world through many channels, including observing with our eyes and hearing linguistic descriptions. For example, we might learn about giraffes by seeing them or hearing people talk about them. Studying the concepts of blind individuals provides insight into the role of vision in knowing what we know. Previous studies suggest that blind individuals have surprisingly rich knowledge about categories which at first glance seem ‘visual’, such as verbs of perception (e.g., look and see) and colors (Landau & Gleitman, 1985; Shepard & Cooper, 1992). One hypothesis is that such information is gained through language. Are there types of knowledge that are uniquely or preferentially acquired through vision? To investigate this question, we tested blind people’s knowledge about animals. Prior neuropsychological and imaging work suggests that visual information may be particularly important for the category of living things (Martin & Caramazza, 2003; Warrington & Shallice, 1984; Farah et al., 1989). 20 congenitally blind and 20 matched sighted control participants made judgments about the physical properties of common animals. For shape, texture, and color, participants sorted animals into groups based on each dimension separately (n=30 animals, Braille or print). Similarity matrices were generated for each sorting rule and participant. For shape, participants additionally performed an odd-one-out task with triplets of animals. For texture, participants judged whether each animal had feathers, scales, skin, or fur (n=33 animals). Finally, participants performed a rank-ordering task with 15 animals based on their size and height. As controls, participants sorted objects based on where they are stored and animals based on where they live. Similarity matrices from objects and animal habitat sortings were indistinguishable across groups (sighted-subject-to-sighted-group (S-S) correlations vs. blind-subject-to-sighted-group (B-S) correlations, t-tests over rho’s: p>0.3). For shape, despite significantly correlated group average matrices (rho=0.83, p<0.0001), blind participants made less distinctions in the sorting (S-S vs. B-S correlations, t(37)=1.71, p=0.09) and triplets tasks (blind vs. sighted accuracy: t(37)=6.06, p<0.0001). With texture, group differences were even more pronounced both for sorting (S-S vs. B-S, t(36)=4.16, p=0.0002) and feature choice tasks (e.g., 42% of blind chose fur for hippo). Blind individuals tended to rely on taxonomy for shape and texture. For color, blind participants’ ratings were entirely uncorrelated with the sighted (group matrices correlation: rho=0.34, p=0.25; S-S vs. B-S, t(37)=11.66, p<0.0001). Unlike shape and texture, color may not be well predicted by taxonomy. When ordering animals based on their size or height, blind participants were less reliable in fine-grained size distinctions. Notably, a few blind individuals’ answers came close to those of the average sighted in all tasks except color. We find that unlike some other types of visual information, knowledge about the appearance of animals is different in blindness, and certain features are more affected than others. These results suggest that some knowledge about physical attributes are learned predominantly through vision and are not captured in linguistic communication. One possibility is that language is less able to capture continuous as opposed to categorical information (Pinker 2007).

Topic Area: Meaning: Lexical Semantics

Back to Poster Schedule