I disagree with its primary assertion, however, that immediate action is based upon conscious perception. As you know from my paper, I think immediate action is performed by the extrinsic neocortex, and subjective experience / episodic memory is formed afterwards, or at least in parallel to that behavior. The hippocampus is dependent upon the extrinsic neocortex to stitch together the movie of perception, but the extrinsic neocortex can behave all day, without needing the hippocampus' report. Like in Driving Mind, the extrinsic neocortex can perform sophisticated behavior, without the feedback loop of perception/memory.
To Matt Faw: Will print out that paper and read (it looks like an unpublished working paper). I need to give a talk about this issue pretty soon, so that's why I am asking. Apparently, attentional and perceptual processes are part of phenomenal consciousness, so the challenge lies in finding appropriate quantitative measures.
Regarding your mention of the extrinsic cortex behaving all day, I think that may suit the default mode network as well (see Raichle & Synder, 2007; URL: Article A default mode of brain function: A brief history of an evolving idea
Jimmy Y. Zhong Yes, I agree about the DMN. I think it is likely working in parallel to the extrinsic neocortex, although the anti-correlational aspect of the executive parts of both networks also reveals something about how attention moves from outside to inside and vice-versa. The paper I posted here earlier is a philosophy paper, not a study, so I'm not sure if it will help you. I just stumbled across it, and it reminded me of this question. If I see anything else in this territory, I'll send it your way.