Often, the contents of consciousness are equated with the contents of short-term memory (or working memory), sometimes to a point where they are treated as identical entities. In the present study we aimed to investigate whether they may be modulated independently and thus dissociated from each other. Previously, we have demonstrated that introspective access and conscious content can affect ERP’s and behaviour in observers during a visual task (Overgaard, Koivisto, Sørensen, Vangkilde, & Revonsuo, 2006). If introspective access to content can modulate performance, one may question if conscious content simply can be reduced to a cognitive process like short-term memory. In two experiments, we combined two different measures of short-term memory capacity to investigate how manipulations of set-size affect performance in observers with the Perceptual Awareness Scale (PAS) to measure conscious experience of the stimulus in every trial (Ramsøy & Overgaard, 2004; Overgaard & Sørensen, 2004). We trained observers to report their experience of a visual target stimulus on the four-point PAS scale; ranging from “no experience” to “clear experience”. To measure short-term memory we used a traditional object based post cue variation of the partial-report paradigm (Sperling, 1960) in experiment 1. Whereas we used the more novel feature based resolution paradigm (Wilken & Ma, 2004) in experiment 2, in which observers had to report the orientation of a cued Landolt C ring. Not surprisingly, we found that increasing task-load through varying set-sizes affected performance in the short-term memory task as well as in the PAS ratings systematically in the both experiments. Interestingly, the two experiments also revealed systematic set-size effect within each of the individual PAS categories. These results indicate that the measure of the experienced content of consciousness does not relate to the measure of short-term memory in a 1:1 fashion.