Hearing parents of deaf children who are reliant on Sign Language need to learn to sign to ensure communication mode-match with their children. Signing is vital for parent-child interaction, and has implications for the socio-emotional well-being and educational outcomes of the child. However, poor signing skills of parents is repeatedly reported in the literature, with the majority of children in signing educational programmes reported not to be exposed to signing in the home. Teaching parents to sign therefore appears a priority, with sign teaching strategies being debated in the literature. The learning of Sign Language as a second language by hearing parents of deaf children within the bilingual educational approach, which regards Sign Language as the first language of deaf children, raises the challenges of cross-modality language learning for hearing parents. Reports on teaching methods are mainly anecdotal with only a few studies addressing sign learning by hearing individuals. While the use of graphic representations of signs is a common practice in teaching signs, there is no empirical data on their influence on the learning of signs. This study explored the contribution of graphic representations of signs in sign teaching. The main aim of the study was to describe the impact of sign illustrations on the teaching of signs to hearing mothers. Two sub-aims were formulated to compare the conditions of sign learning with and without the use of sign illustrations in graphic displays in terms of (a) sign reception and sign production, and (b) the amount and nature of assistance required in learning signs. An Adapted Alternating Treatments Design (AATD), with four theme-based sign sets, and probes balanced for equivalence, was developed and used. Four biological mothers of three boys and a girl in a Grade Three class at a day school for the deaf in an urban area of KwaZulu-Natal, South Africa took part in the study. The results revealed no significant differences between the two training strategies for sign acquisition, in terms of sign reception and sign production post-training. There were however, significant differences between the two training strategies with regard to assistance required while learning signs. The graphics strategy required significantly less trainer assistance (p<0.05). In addition, there were significant differences in the nature of assistance provided with the use of graphic representations. Significantly fewer repeated demonstrations of signs were required by the participants during self practice (p<0.01). There was a significantly higher number of corrections with the graphics strategy (p<0.01) initially, and this decreased over time, unlike with the signing-only strategy. It would appear that the sign illustrations were redundant during the initial stages of sign learning using a multimodal approach, but that they were relied on to trigger recall of signs during the self practice phase. Thus, the study confirmed the supportive role of sign illustrations in sign learning. The use of theme-based graphic displays of sign illustrations emerged as a viable method in teaching signs. The implications of these results and recommendations for future research are discussed.