

Abstract. The first part of the paper discusses the process of collecting empirical data in a rural community of the Qwa-Qwa region of South Africa in order to establish the community’s views and opinions about different types of visual illustration approaches. The information was subsequently used to guide the production of visually illustrated soy-related nutrition education messages, which were distributed free of charge in the community as part of an integrated nutrition education programme. In the second part of the paper, the discussion of the data collected during three separate data collection events (n=75, n=67 and n=92) focuses on the relationship between (1) image readability, or the absence of readability barriers; (2) the visual aesthetic preferences of the members of the target group, based on the respondent’s choices and verbal explanations when presented with a range of different visual illustration options for comment, and (3) the overall appropriateness of an illustration.