This section describes some assumptions and terminology. A simple percept consists of a representation of an environmental property and a singular element that picks out an individual item (Burge 2010). The representational content is the condition of veridicality and is satisfied only if the referent of the singular element has the property represented by the property-representation. The precision of a representation—in my terminology— is a matter of the range of values attributed. For example, consider two visual representations of the height of a person, one representing the person as between 5’6” and 6’ tall, the other representing the person as between 5’8” and 5’10” tall. The latter has a narrower precision. Precision in the sense used here is not a matter of indeterminacy of borders but rather the size of the range.[2]
The claim that the precision of a representation is wide is a form of the claim that perception is “intervalic”. There are other measures of perception that are easily confused with precision. One of them is acuity—also known as spatial resolution. Acuity is the ability to resolve elements of stimuli. Common measures in the case of vision are the extent to which the subject can distinguish one dot from two dots, detect a gap between two figures, determine whether a rotating figure is rotating clockwise rather than counter-clockwise, ascertain whether two line segments are co-linear, distinguish a dotted from a solid line or detect which side of a Landolt Square a gap is on. (See Figure 4 for an example of a Landolt Square.)
These and other items of terminology are gathered together in a glossary at the end of the article. Of course other quite different definitions of ‘precision’ and ‘acuity’ are just as legitimate as these. Note in particular that I am not using the notion of precision as the inverse of variance or the notion of precision associated with the predictive coding literature.
Representationists (also known as representationalists and intentionalists) think that what it is like to have a perceptual experience—that is, the phenomenology of perceptual experience—is grounded in the representational content of the perception. (Not that representationists have used the notion of grounding, but I believe that it captures what they have meant.) Representationism is sometimes framed as an identity thesis (e.g., Pautz 2010; Tye 2009): what it is for an experience to have a certain phenomenal character = for it to have a certain representational content. But the identity formulation is inadequate because the phenomenology is supposed to be based in the representational content and not the other way around. Identity is symmetrical. The grounding characterization of representationism avoids this problem since grounding is asymmetrical. To say that perceptual representation grounds perceptual phenomenology is to say that it is in virtue of the representational content of a percept that it has the phenomenology it has. And in virtue of is asymmetrical. (See Fine 2012 on the concept of ground and my 2014a for further discussion of grounding in philosophy of mind.)
Representationism is often framed in terms of supervenience: no difference in the phenomenology of perception without a difference in its representational content. But supervenience does not capture a key motivation behind representationism: that the representational content of perception is the source of the phenomenology of perception, that it is in virtue of the representational content of the perception that it has the phenomenology it has. A supervenience formulation would entail that a difference in the precision of phenomenology requires a difference in representational content. However, on a supervenience formulation of representationism it would be a further question whether the phenomenology of perception could increase in precision without a commensurate increase—or even with a decrease—in precision of its representational content. On the grounding characterization, any change or difference in phenomenological precision is dependent on a commensurate change or difference in representational precision.
The grounding formulation of representationism rules out some but not all kinds of multiple realization. Suppose that red782 is an example of the most fine-grained color we can experience. And suppose that the representationist theory of the experience as of red782 is that this experience is grounded in representation of red782. Different experiences as of red782 can be realized by different representational states so long as they all involve the representation of red782.
The grounding characterization captures a representation-first view and excludes phenomenology-first doctrines that are often portrayed as representationist. Phenomenology-first views suppose that phenomenology grounds at least some kinds of representational contents (Hill 2009; Kriegel 2011, 2013; Shoemaker 2007). And it also excludes versions of representationism that treat both the phenomenology and representational content of perception as grounded in something else (Chalmers 2006; Siegel 2013). That is a plus for the grounding characterization—distinguishing between fundamentally different points of view. Although I won’t talk about this much here, I think the considerations I will be raising will cast doubt on views that phenomenology grounds any kind of representational content.[3]
The reader may feel that both peripheral and unattended perception are odd and unimportant phenomena that cannot be the test of any theory of perception. However, peripheral unattended perception is ubiquitous. The fovea is the high density center of the retina. If you hold your hand at arm’s length, your foveal perception encompasses about double the width of your thumb. Much of perception at any fixation occurs outside that area and a similar point applies to attention. However, even if you think that both peripheral and unattended perception are atypical, you should recognize that atypical cases often are a window into the nature of a phenomenon. The experiment in which a beam of light goes through two slits was crucial in demonstrating a wave aspect of light (Feynman 1988).