The purpose of the research conducted for this thesis was to understand the combination of different modes of representation (multimodality) in contemporary art-making in order to better prepare arts education for the challenges of communicating in a multimodal world. Through personal experience as an arts educator in secondary and tertiary education, and an adjudicator of an important national competition for young contemporary artists, the Sasol New Signature Art Competition, I became aware of the discrepancy between curricular emphases at schools and universities, and the characteristics of pieces that are selected as winning works by the artistic society. Whereas traditional art education focuses on technical perfection and techniques bound to traditional art genres, contemporary society values meaning making through various modes of expression (multimodality) and by increasingly using technology as a vehicle of delivery. It was therefore the main aim of this study to explore empirically the processes by which multimodal art-making takes place in practice, in order to elicit principles of how multimodality may equip art educators to make aspects of a hidden curriculum visible, and thus better equip learners/students to cope with the demands and challenges of an increasing multimodal world of communication. Through the analyses of various art works it is shown that although the notion of multimodality originated among linguists and was theorised within the framework of Systemic Functional Linguistics, instances of the combination of multiple sense-based modes (sight, sound, movement, tactability, taste and olfaction) of which the visual mode is typically central, have been evident throughout the history of visual art. Theoretically I draw on socio-semiotics, which emphasises the construction of meaning by exploiting the affordances of multiple modes and media. Although the initial rigorous grammatical approach of Kress and Van Leeuwen (1996) gave momentum to the theorising of multimodal meaning making, it has since been replaced by a perspective that retains only the foundations of Systemic Functional Grammar, and further draws upon the notions of continuous design and re-design driven by social discourses and technology. The empirical research, aimed at establishing how and why nascent artists combine particular modes and media, was conducted within a multiple case study design, focused on the qualitative content analysis of questionnaires e-mailed to 10 winners of a national art competition, their winning art works, and the artist statements accompanying the art works. Four prominent themes emerged from the data, namely combination, layering, viewer response and validation of contemporary issues. These themes do not only support the sense-based taxonomy of modes that precipitated from a literature review of prominent sources on multimodality, but also served to underpin a curriculum framework for art education.