The aim of this study was to examine the relationship between children’s visual attention to, and acquisition of, Key Word Sign (KWS). Our hypothesis was that children’s visual attention to the clinician’s modelling would be associated with their acquisition and production of KWS. A multiple baseline single case experimental design with additional exploratory analyses was used to examine the effect of visual attention on sign acquisition among three preschool children with ASD. This paper extends from the previous intervention study (see Tan et al. 2014) by examining visual attention as a potential factor underpinning individual differences in sign acquisition documented in the initial study. All three children visually attended to the clinician’s modelling of KWS, and acquired signs to varying degrees following the introduction of KWS intervention. A weak, non-significant correlation was found between the children’s amount of visual attention and their production of signs. The results provide preliminary evidence for a lack of association between children’s amount of ‘looking’ and ‘doing’ during KWS intervention. Replication and examination of other factors impacting on KWS intervention outcomes is required.