This paper describes a Conceptual Framework underpinning "Systems that Care" in terms of educational systems that take account of motivation, metacognition and affect, in addition to cognition. The main focus is on "motivation," as learning requires the student to put in effort and be engaged, in other words to be motivated to learn. But motivation is not the whole story as it is strongly related to metacognition and affect. Traditional intelligent educational systems, whether learner-centred or teacher-centred in their pedagogy, are characterised as having deployed their intelligence to assist in the development of the learner's knowledge or skill in some domain. They have operated largely at the cognitive level and have assumed that the learner is "already" able to manage her own learning, is "already" in an appropriate affective state and also is "already" motivated to learn. This paper starts by outlining theories of motivation and their interactions with affect and with metacognition, as developed in the psychological and educational literatures. It then describes how such theories have been implemented in intelligent educational systems. The first part of the Conceptual Framework develops the notion of a partial hierarchy of systems in terms of their pedagogic focus. These range from traditional, cognitively intelligent systems, essentially concerned with cognition up to "Systems that Care." Intermediate classes of system include Metacognitively Intelligent systems, Affectively Intelligent systems and Motivationally Intelligent systems. The second part of the Conceptual Framework is concerned with the design of systems. This is characterised in terms of (i) the kinds of diagnostic input data(such as the learner's facial expression offering clues as to her demeanour) and (ii) the repertoire of tactical and strategic pedagogic moves (such as offering encouragement), applicable at different levels of the hierarchy. Attention is paid to "meta"cognition, "meta"-affect and "meta"-motivation covering the capability of both the learner and the educational system to understand, reason about and regulate cognition, affect and motivation. Finally, research questions and areas of further work are identified in theory development, the role of the meta levels, and design considerations. (Contains 2 tables and 5 footnotes.)