Ned Block argues that the results of the Sperling experiment (Sperling Reference Sperling1960) lend support to the view that phenomenology outstrips or overflows cognitive accessibility. I shall argue that Block fails to make a compelling case for this claim on the basis of the Sperling data.

I agree with Block that the subjects in the Sperling experiment see all 12 letters in the array even though they can accurately report the identities of only three or four. The first point I want to make is that there is an important sense in which all twelve letters are cognitively accessible notwithstanding the subjects' inability to identify two-thirds of them; for, with respect to each letter in the array, the subjects' experiences enable them at least to wonder “What is that letter?” Their experiences thus put them in a position, with respect to each letter, to bring the letter under the demonstrative concept that. The content of the concept varies with the putative query. Were a subject to wonder “What is that letter?” with respect to, say, the letter S in the top right-hand corner of the array, the cognitive content of the query would be different from the content of the query, expressible in the same sentence, with respect to the letter B in the middle of the bottom row. Of course, subjects do not typically so wonder with respect to each letter in the array. They may not so wonder with respect to any letter. The point is that their experiences enable them to wonder, “What is that?” with respect to each letter, whether or not they actually wonder anything, and the content of this possible cognitive response varies with the letter. So, each letter in the array is cognitively accessible via a cognitive act, the content of which is distinctive to that letter.

This brings me to Block's claim that phenomenology outstrips cognitive accessibility. Let us look more closely at how this is supposed to be the case. One way of understanding Block's claim in a particular instance is that the subjects' phenomenal consciousness of the letter S, say, in the top row, is different from the subjects' phenomenal consciousness of the letter B in the bottom row, even under the condition that the middle tone is heard and subjects are attending to the central row. But this does not show that phenomenology outstrips cognitive accessibility, since the possible act of demonstrative wondering with respect to the former letter has a different content from the possible act of demonstrative wondering with respect to the latter.

What happens when the middle tone sounds? One possible view is that there is no change in phenomenology as the subjects attend to the letters in the central row. The subjects simply focus their attention on the row and report what the letters look like. But patently this does not show that phenomenology outstrips cognitive accessibility. I am happy to allow, of course, that when one focuses on a given letter, one is then put in a position by one's experience to identify the letter as being the letter such-and-such and thus to cognitively classify the letter in a way that one could not before. With the shift in attention, then, one's cognitive possibilities enrichen – one goes from possibly wondering “What is that?” to actually thinking that that is the letter P, for example – even though (I am now supposing) the phenomenology stays the same. But this is not to the point. It shows that cognitive accessibility, content-wise, outstrips phenomenology, not the other way around.

I should add that I am actually disinclined to accept that when one shifts one's attention to a particular row, there is no change in the appearance of the letters. It is true that there is no dramatic or big change, but the situation seems to me similar to the following: Focus on an object – a watch, say – lying on a magazine in the middle of your field of view. There will likely be letters on the magazine cover close to the watch that you won't be able to identify even if you try, given your focus of attention. It is not that the letters are fuzzy or obscure; it is just that they are insufficiently determinate in your experience. Now switch your attention slightly from the watch to the letters. With the new focus, you will be able to identify them, but there will be no marked or sudden change in the phenomenology. The letters will not sharply change in their appearance. Still, the phenomenology does change a little. At a conscious level, there is a detail in the letters that simply was not there before – a detail that now enables you to identify them. Subtle changes of this sort happen all the time as we move our eyes around. But these changes are not dramatic or pre-reflectively obvious.

Block sometimes states his position in a more qualified way. He says that, in his view, the reason why subjects are unable to gain access to all the letters in the Sperling experiment without the tones is that “the ‘capacity’ of phenomenology, or at least the visual phenomenal memory system, is greater than that of the working memory buffer that governs reporting” (sect. 9, para. 11). The working memory system has a capacity of four items (or less) in humans whereas the capacity of visual phenomenal memory “could be said to be at least 8 to 32 objects – at any rate for stimuli of the sort used in the described [i.e., the Sperling and Landman et al.] experiments” (sect. 9, para. 11).

The point I wish to stress again is that cognitive access can be achieved via the use of the demonstrative concept that letter. Seeing the 12 letters, one is put in a position by one's experience to query the identity of each letter and to so do, in each case, via a cognitive act that, in part, subsumes the relevant letter under the concept letter. This involves the use of working memory. Furthermore, it permits one to issue the report, “That is a letter.”

Presumably, what Block has in mind when he says that the capacity of the working memory system is four items (or less) under a certain restricted range of conditions is the capacity to identify (for example) letters as the specific letters they are. But now his claim about phenomenal consciousness outstripping or overflowing cognitive accessibility, although not without interest, does not have quite the bite it first appeared to have. Certainly, nothing that Block says here shows that there fails to be an important connection between the phenomenology of the experience one undergoes in seeing an object and the object's general cognitive accessibility.