[1]
See for example Seth et al. (2005), who presented a proposal close in spirit.
[2]
See Overgaard & Grünbaum (2011); Block (2011); Cohen & Dennett (2011); Kouider et al. (2010).
[3]
See the debate about alternative explanations of the findings of atypical perceptual conditions (for example of the Sperling paradigm) in the references above.
[4]
By this I mean that the evidence does not allow us to rule out the whole class of cognitive versus non-cognitive accounts. I think however, that certain accounts within theses classes are vulnerable to evidence; for example, explicit accessibility accounts (Prinz 2012) seem to have a lot less room to maneuver. But as a debate between cognitive versus non-cognitive accounts, the possibilities for interesting general insights seem limited.
[5]
Of course hybrids are possible, so we might have combinations of functional differences and differences in content. I take Tye (1995) to defend such an account.
[6]
See Schooler (this collection), p. 8.
[7]
Bayne & Montague (2011) provide a nice overview of the complex cognitive phenomenology debate in his introduction to his volume. One might think that other contents causally influence the phenomenology of a state. A second option would be that “what it is likeness” is not a useful conceptual distinction at all (Lycan 1996, p. 77; Papineau 2002, p. 227). A third option would be that there are several meanings of “what its likeness”—indeed, in the literature different distinctions have been suggested. I will go into more detail in a later section, when I introduce elements of an improved taxonomy.
[8]
See further discussion in Grush (this collection).
[9]

Regarding visual perception Siegel (2005) has argued that that learning to recognize an object can change the way that it looks—in the phenomenal sense of “look”, which is taken to imply that the cognitive components of such states are necessary for explaining the change in phenomenal character. In contrast, one could argue that the phenomenology does change, but the change can be explained in sensory terms instead of in terms of cognitive components. Either a subject’s concepts do not directly constitute the subject’s phenomenal states, such that they can have a causal influence on their phenomenology (Carruthers & Veillet 2011), or the contrast between both is the result of differences in the way that one processes the information within the sensory system (Tye & Wright 2011). For my purposes here, what matters most is that the phenomenology differs, and that we need an explanation for it.

[10]
As the complex debate about of the possibility of a phenomenology of thought suggests.
[11]
At the very least we would need to insist that there is a family of co-occurring properties playing an explanatory role within theories (Boyd 1999).
[12]
See also, for example Schooler (this collection), pp. 16-17.
[13]
For the purposes of this commentary I neglect biological state theories.
[14]
On a very broad reading of “functional”.
[15]
Initially introduced by Baars (1988, also 1996). More modern proponents would be, for example, Dehaene et al. (2006).
[16]
However, in the end accessibility accounts will not be Schooler’s best bet—after all, I interpreted him above as agreeing that access and awareness differ.
[17]
However, one might be able to resist the distinction between access-consciousness and phenomenal consciousness and at the same time allow for Schooler’s distinction between experienced consciousness and meta-awareness if one claims that access is not what characterizes the meta-level in Schooler’s meta-awareness.
[18]
Schooler (2013) gives a good overview of the performance costs associated with mind-wandering (including reading comprehension, model building, and impairment of the veto-option to automatized responses) and suggests that mind-wandering may represent a pure failure of cognitive control. For this reason it is so useful to study consciousness. He argues that mind-wandering offers little benefit, though it might have a positive role in topics related to autobiographic episodes and information, for example in autobiographical planning and creative problem-solving.
[19]
As formulations such as “take stock what’s going in their own minds” (Schooler this collection, p. 8) suggest.
[20]
For a more detailed discussion of the same issue see Metzinger (2013, p. 11).
[21]
Otherwise we would have a false belief, not knowledge.
[22]

Like most authors, they focus for the most part on the discussion of conscious perception, and especially Sperling (Block this collection; Fink this collection) and Stroop’s paradigms (see Mroczko-Wąsowicz this collection) and what we can learn from them for consciousness. For a more detailed discussion of the pros and cons or an understanding of consciousness as graded within conscious perception see the debate between Cleeremans (2008), Sergent & Dehaene (2004), Seth et al. (2008), and Overgaard et al. 2006).

[23]
However, this section of the target paper goes beyond the discussion of the well-known traditional arguments from the phikosophical debate, including the explanatory gap argument, the Mary argument, and others. Schooler adds a section on the phenomenon of time experience and reductive accounts that explain time. I found the last example very inspiring, because in contrast to the other arguments it is not just based on thought experiments. However, the implications of this for my purposes here do not matter; they are used as a intuition punp to appeal to the necessity of a meta-level, so I cannot cover this aspect in this commentary.
[24]
For example in the the discussion of emotions p. here.