The aim of this research was to investigate whether video peer modelling and video self modelling (VSM) were effective treatments in reducing fear of dogs in children with autism spectrum disorder (ASD). This study also examined whether video modelling could be used as an effective tool to increase knowledge of dog behaviour and dog safety in individuals with ASD. Participants were recruited with a poster that was distributed through the national ASD organisations and ASD support groups. Four participants were recruited; they were aged between 9 and 17 years, had a formal diagnosis of ASD, and had a fear of dogs. This study used two single-case design experiments; one pair of participants experienced an A-B sequence (a VSM), and a second pair of participants experienced an A-B-C sequence (video peer model followed by VSM). Participants met with the researcher five times: three baseline sessions, and two post-intervention sessions. All four participants showed some increase in steps achieved in their Behavioural Avoidance Test (BAT) scores, with two participants experiencing large changes. The VSM intervention appeared to have had more effect on increasing BAT scores/reducing fear than the video peer model, though this was not proven conclusively. There was little evidence to support video peer modelling or VSM as an effective tool to increase knowledge of dog behaviour and dog safety. Limitations on this study include the small number of participants, and the single case design in the form of AB/ABC sequences used.