In philosophy of perceptions, representationalism (indirect realism) is the traditional view supported by Descartes, Locke, Leibniz, Hume, Husserl, Russell, etc. But this view becomes minority in modern times in favor of direct realism. According to reference here:
Representationalists are often motivated by their belief that experience is transparent, that is to say their belief that when we introspect we find that we are only paying attention to the seemingly mind-independent objects and properties that we are perceiving, rather than attending to other distinctive mental features of experience or any apparent non-representational properties of experience. If such a view were true, then perhaps what it is like to have an experience is exhausted by the experience’s contents.
... there are a battery of examples in the recent literature in which, it is claimed, there are experiences that have differing phenomenal character yet the same representational content and vice versa. Examples of this kind would provide counter-examples to the representationalist’s identity or supervenience claim.
So in summary, representational features of sense-data are those mind-independent objects and properties perceived in our mind which can be further abstracted, symbolized and referenced for further simulation or AI applications. The remaining non-representational features are those resist such formalization. For example the blueness quale we all seem to experience when watching the clear sky or sea, for a representational p-zombie it can only manipulate and apply the word blue or some symbol like BL in relation with other representational symbols exactly like our mind if technology is good enough, but it cannot feel the same blueness we humans all can perceive and seem to be able to share without much difficulty, otherwise the traffic lights would never work in any society...