A ubiquitous feature of perception is that perceptual discrimination is more fine-grained than perceptual identification. Even those with absolute pitch can identify perhaps 100 pitches (depending on exactly how absolute pitch is defined), but can discriminate many thousands of pitches from one another (Raffman 1995). Those who have absolute pitch are no better than other musically literate people in pitch discrimination (Levitin 2005, 2008). Given the disparity between identification and discrimination one might wonder whether our ability to make fine grained perceptual discriminations misleads us as to the precision of our perceptual representations. It certainly seems to us that each of those thousands of pitches has a distinct phenomenology but maybe that judgment feeds more off of the phenomenology of discrimination of differences than off of the phenomenology of individual pitches.
I have argued that the phenomenology of perception does not allow for a large degree of imprecision.[15] I appealed to the “just noticeable difference” of contrast of 2%. I said:
The representationist may retort that the point is not that the contents are fuzzy or represented indeterminately but that they are abstract relative to other contents, as determinables are to determinates, for example as red is to scarlet. But this line of thought runs into the following difficulty: the variation of 6% due to attention is way above the ‘just noticeable difference’ threshold, which for stimuli at these levels is approximately 2%. (Or so I am told. In any case, just looking at the stimuli in Figure 4 [Figure 9 here] shows that the difference is easily detectable. And you may recall that in the discussion of the tilt aftereffect, there was evidence that at higher levels of contrast, the increase due to attention was as much as 14%.) The point is that there is no single ‘look’ that something has if it is 22% plus or minus 6% in contrast. By analogy, consider the supposition that something looks as follows: rectangular or triangular or circular. That disjunctive predicate does not describe one way that something can look—at least not in normal perceptual circumstances (Block 2010, p. 52).
Jeremy Goodman (2013) has criticized my reasoning. He says:
Ned Block, when considering the hypothesis that perceptual appearances are ‘abstract relative to other contents, as determinables are to determinates, for example as red is to scarlet’, objects that ‘the variation of 6% due to attention is way above the “just noticeable difference” threshold, which for stimuli at these levels is approximately 2%’ (p. 35).
Goodman goes on to speak of “Block’s objection that our discrimination thresholds place an upper bound on the unspecificity of perceptual appearances…” (2013, p. 35). Although it may have sounded that way, I did not intend to claim that discrimination places an upper bound on either representational or phenomenal imprecision. But I think that a certain kind of discrimination is relevant to both imprecisions as I will explain.
An ability to discriminate between two observable magnitudes does not prove that one’s percepts of the magnitudes actually differ (in either representational content or phenomenology). One example that I have used to illustrate this point (Block 2007, p. 540) is the phenomenon of “beats” (alternating soft and loud sounds) caused by interference between guitar strings of very slightly different pitches even when the two pitches are phenomenally the same on their own. (The frequency of beats in response to two pure pitches is the difference in frequencies.) Another is the color border effects that allow one to see that two colors are different even when the colors themselves would not be distinguishable if separated slightly. Even for achromatic objects, slight differences are amplified by a well known border phenomenon that is responsible for the famous “Mach Bands” illusion. Goodman uses the example of two trees that look to have slightly different heights because of how far they stick up above the tree canopy. His point is that vision might represent overlapping but slightly different intervalic values, but one could also use the example to illustrate heights that don’t look at all different when seen separately while nonetheless allowing one to see a difference when seen next to one another.
It is intuitive to think that the way the visual system detects differences between one thing and another is by registering the properties of each thing separately and comparing those registrations. But this is not always the case: Differences are often detected via different processes than the processes that register the entities or properties that differ. Beats are produced by interference between two sound waves, allowing one to detect differences between sounds that would otherwise be inaudible.
As the examples just given illustrate, discrimination may be possible without any difference in the phenomenology of the individual percepts. Two pitches can be indiscriminable even if one knows they differ because of beats. However, there is no reason to think that specialized discrimination mechanisms are at work in the experiments described. Specialized discrimination mechanisms can be expected to depend on the specific features of the perceptual situations and so not robust to changes in the situation of the perception. For example, if you change your angle of view you might see the full vertical length of the trees but not their differential protrusion above the canopy. Border contrast effects are fragile—move the color samples just a bit apart and the effect vanishes. (This is nicely illustrated in the Wikipedia entry for “Mach Bands”.) However, the attentional effects I have been talking about apply to color, speed, size of a moving object, spatial frequency (stripe density), time of occurrence, flicker rate, motion coherence (the extent to which many moving items are going in the same direction), as well as to contrast and gap size. What is the chance that there is some specialized discrimination method at work for all these magnitudes? Most impressively, these effects can be exhibited in visual short term memory—that is, they don’t even require simultaneous perception. This was shown by Martin Rolfs & Marisa Carrasco using a different experimental paradigm than the ones so far discussed (2012; Rolfs et al. 2013). I won’t describe it except to say that the patches are compared in respect of contrast by comparing a patch seen earlier with a currently seen patch, and with similar results to those already described. As I mentioned in section 3, a similar experiment shows that a perceived patch at one location in the visual field can be compared with a remembered patch at a different location with results that show the inhomogeneities in the visual field (Montaser-Kouhsari & Carrasco 2009). The likelihood that there is some method of comparison that does not depend on the individual percepts themselves but survives all these variations does seem slight.
So the kind of discrimination that is not based on specialized mechanisms of detecting differences independently of registering absolute value can be used to make a better case.
But even if we can make very fine grained discriminations and even if the percepts involved in the discriminations are distinct, it does not follow that the percepts are not highly imprecise—as mentioned earlier. Suppose for example, that perceptions of contrast are so imprecise as to cover nearly all the range of contrasts. Consider a representation of contrast of 4%-98%. Still, 4.1%-98.1% would be another equally imprecise content that is nonetheless distinct from the first one. And so more generally discriminability has little in the way of immediate consequences for imprecision.
The notion of a just noticeable difference is not very useful for my purposes. Discrimination can be finer than absolute registration as in the case of beats. And strong ability to discriminate is compatible with a high degree of imprecision. Further, the notion of a just noticeable difference combining as it does, perception with cognition, allows the possibility of a difference in conscious percepts that is not cognitively accessible.