Viewpoint Preferences in Signing Children’s Spatial Descriptions

In many sign languages, spatial relations are primarily expressed from signer-viewpoint, in which a signer places a sign to her right, for example, for an entity on her right. The use of addressee-viewpoint – a signer placing a sign to her left for an entity on her right, such that the addressee's view of the hands corresponds to the spatial configuration – is attested but found to be less frequent in comparison. How do signing children deal with viewpoint-dependent spatial relations? Previous studies have shown that signing children, even at the age of 10, lag behind their speaking peers in comprehending spatial relations that require imposing a viewpoint – mostly due to the difficulties in acquiring mental rotation skills and cognitive reversal of perspective. However, there are no production studies with signing children on the acquisition of viewpoint preferences in encoding viewpoint-dependent spatial relations. In this study, we elicited spatial descriptions from preschool-age and school-age deaf children acquiring Turkish Sign Language (TİD) natively, and from deaf TİD-signing adults. Our results revealed a split in adult preferences for viewpoint: for encoding lateral-axis spatial relationships, adult signers of TİD preferred signer-viewpoint over addressee-viewpoint, but for encoding spatial relations on the sagittal axis, they used addressee-viewpoint more frequently than signer-viewpoint. TİD-acquiring children in both age groups, on the other hand, described spatial configurations (both on the lateral and sagittal axis) by using signer-viewpoint. Thus, they used signer-viewpoint as frequently as adults for lateral axis encodings, but addressee-viewpoint less frequently than adults. These results suggest that while learning to encode spatial relations that require imposing a viewpoint, TİD-acquiring children start with signer-viewpoint, even when adults might prefer addressee-viewpoint for certain relations. This is in line with previous studies showing that children acquiring a spoken language describe spatial relations from their own viewpoint. Thus, regardless of the modality of language being acquired, children may learn to express spatial relations from their own viewpoint earlier than the viewpoint of their addressee. It also shows that, at least for objects placed on the lateral axis, children even in the preschool-age group reached adult patterns earlier in production than reported for comprehension of signed expressions. The results are also discussed in relation to why addressee-viewpoint is preferred by TİD-signing adults for objects placed on the sagittal axis, which might have led to non-adult-like patterns by TİD-acquiring children.