Changes in Visual-spatial Cognition when Adults Learn American Sign Language and How Pre-existing Visual-spatial Cognition Predict Success in Learning

Abstract

Visual space has unique importance in signed languages because, unlike spoken language, signed languages use space to encode multiple linguistic features. For example, space is used for grammar, where spatial locations, and movements between them, are used to mark grammatical features such as subject and object. This implies that visual-spatial cognition is involved in processing signed languages, and indeed, compared to non-signers, native signers have been shown to have superior visual-spatial short-term memory (STM), working memory (Wilson et al., 1997; Lauro et al., 2014), and mental imagery abilities (Emmorey et al., 1993; Emmorey and Kosslyn, 1996). Fluent, non-native signers also have demonstrated enhanced visual-spatial working memory (Keehner and Gathercole, 2007), and hearing children learning sign language show visual-spatial cognitive benefits after one year of experience (Capirci et al., 1998). It is not known, however, how visual-spatial cognition changes in adults as they first begin to learn sign language. Furthermore, beyond improvements in visual-spatial skills with sign language learning, it is not clear if pre-existing individual differences in visual-spatial cognition can predict success in ASL learning. In the present study, adult English speakers with no prior sign language experience were recruited from first level ASL courses and performed tasks that assessed their visual-spatial cognition, verbal memory, and ASL proficiency before and after one academic semester of ASL instruction. To determine whether changes in visual-spatial cognition are specific to learning a visual-manual language, I also recruited adults learning any first level spoken language to serve as controls. Verbal STM/working memory was assessed with Digit Span Tasks and the OSPAN task. Visual-spatial STM/working memory was assessed by the Movespan task, which evaluated working memory for human actions, and by the Corsi block-tapping task (regular and rotated). The ability to mentally generate images was assessed by the mental clock imagery task, while mental rotation was assessed with the 3D mental rotation task and the rotated Corsi block- tapping tasks. The ASL proficiency of the participants was assessed with an ASL Picture Naming test and an ASL Recognition test. The results showed that the ASL learners, when compared to the controls, improved on the non-rotated Corsi block-tapping task and the MoveSpan task. There were no group differences in the clock task or mental rotation task. I conclude that even limited sign language experience can improve visual-spatial memory, and I hypothesize that more sign language experience is needed to see improvements in mental imagery abilities. Additionally, the evidence suggested that the participants’ pre-learning Corsi score was a good predictor of their changes in ASL proficiency compared to the Ospan score. This implies that visual-spatial STM might be able to predict how well adults learn sign language as a second language and appears to be more predictive than verbal memory. However, the relationship between visual-spatial memory and changes in ASL proficiency was not clearly established and warrants further investigation.