Mind that Epistemic Gap! Perceptions Reside Within, not Outside of, Reality
Tuesday, September 22, 2015
Hoffman and colleagues propose the Interface Theory of Perception (ITP), which asserts that perceptions evolved to render organisms sensitive to the objective world in terms of payoffs, or fitness functions, not truth. According to ITP, “…perception is about having kids, not seeing truth”, so evolution ensures that perceptual representations simplify an organism’s search for survival/reproduction-relevant dimensions. Accordingly, the issue of whether or not perceptions constitute veridical(i.e., accurate) representations of “the objective world” is simply beside the point.
Hoffman and colleagues clarify the assumptions of ITP by contrasting it with a variety of realisms (e.g., naïve realism, omniscient realism, critical realism, and hybrid realism), all of which are referred to as “realisms” because they assume perceptions correspond to (i.e., represent) the objective world, “as it is,” to some degree or another. For example, if one perceives a snake, according to Hoffman and colleagues, a realist would claim that some aspect of the observer’s perceptual representation (e.g., the aspect that represents shape, or movement, or size) accurately corresponds to something in the objective world—for example, the movement aspect of the representation might correspond to movements that actually exist as movement in the objective world.
According to Hoffman and colleagues, all realisms assume some degree of veridical correspondence between perception and the objective world, whereas ITP does not. Instead, ITP assumes perceptions represent fitness, not truth. As a result, ITP is not a form of realism. Instead, the authors refer to ITP as an interface theory.
At first glance, ITP seems quite different from realisms, for while they require veridical correspondence between perceptions and the objective world, ITP does not. Upon further reflection however, it turns out the two classes of theory are actually quite similar, in that both believe perception corresponds to (i.e., represents) something. In short, both ITP and the realisms it critiques constitute correspondence theories of perception, a term that I coined in earlier work with Brian Day. According to our earlier work, the problem with correspondence approaches to perception is their age-old practice of defining perception in contrast to reality (i.e., the objective world). This maneuver is problematic because it contradicts itself (i.e., it is logically incoherent): Perception cannot be coherently defined in contrast to reality because all perceptions are entailed within reality. There is no set we can refer to as “perception” that is not entailed within the set “reality.” For example, if I see a bracelet on the ground, reach to grab it, and then perceive a snake instead of a bracelet, both the snake and the bracelet are “real” in the sense that both perceptions really occurred—they both constitute real phenomena. As a result, both reside within reality, and defining them in terms of how they differ from reality doesn’t make sense because the perceptions, themselves, are part of reality.
Hoffman and colleagues, however, attempt to define perception in contrast to reality in the equation P: WèX, in which W refers to “measurable spaces” of the world, X refers to perceptual experiences, and P is a perceptual strategy. While describing this equation, the authors state, “…X is not a subset of W…”. At another point in the paper, they state, “The interface theory of perception certainly runs counter to our normal intuitions about the relationship between our perceptions and reality”. Also, “The interface theory of perception…says that reality differs from our perceptions not just in this or that specific case, but in a far more fundamental way…” Thus, while the authors claim in P:WèX, that X is not a subset of W, their later equating of W and reality contradicts P:WèX, for X is most certainly a subset of reality.
To be sure, Hoffman and colleagues could argue that ITP is not a correspondence theory of perception because ITP asserts that perceptual representations correspond to fitness, not “truth.” In consequence, ITP representations do not have to necessarily correspond to anything in the objective world. However, Hoffman and colleagues include the objective world in their definition of fitness: “Fitness is a function of the objective world. However, a fitness function depends not just on the objective world, but also on the organism, its state, and an action”. And later, “Fitness is, in general, a complicated function of the objective world that depends on an organism, its state, and its action”. Clearly, perceptual representations based on fitness functions represent, in one way or another, be it veridical or non-veridical, aspects of the objective world. It follows that ITP, too, is a correspondence theory of perception.
Further evidence of ITP’s status as a correspondence theory of perception can be found in the authors’ assertion that fitness representations can be “right” or “wrong.” “Interface strategies can be tuned entirely to the right information (italics added) whereas realist strategies are necessarily tuned to the wrong (italics added) information”. The authors’ use of the terms right and wrong clearly indicate the belief that fitness representations correspond to (i.e., represent) fitness either correctly (i.e., veridically) or incorrectly (i.e., non-veridically). Clearly, Hoffman and colleagues assume that perceptions can be veridical, not in the psychophysical sense the authors attribute to realists, but in the functional sense that a representation can accurately represent an organism’s fitness. Thus, given ITP’s dual assumptions that (1) perceptions constitute representations, and (2) perceptions can vary in their degree of veridicality, be it psychophysical or functional, ITP clearly falls within the long history of correspondence approaches to perception, stemming at least as far back as Descartes, that define perception in terms of how it is differs from, and corresponds to, something else.
David Vinson and I recently referred to the gap that correspondence theories create between perception and reality as an epistemic gap. In the attempt to bypass such epistemic gaps, my colleagues and I have proposed Wild Systems Theory (WST), which conceptualizes organisms as embodiments of the multi-scale contexts from which they emerged and within which they sustain themselves. From this perspective, the body of a fish (i.e., its fins as well as its neuromuscular architecture) can be coherently conceptualized as an embodiment of the properties of water and the constraints that need to be addressed to propel a mass through water.
An advantage of conceptualizing organisms as embodiments of context is that it logically nests the organism, and all aspects of the organism, within reality. As a result, we need not generate epistemic gaps that lead us to assume the need for bridging principles, or “representations” that allow the organism to cross an epistemic gap. Every aspect of an organism is a “representation,” at some time scale, of the phylogenetic, ontogenetic, and immediate constraints the organism and its species has had to embody in order to sustain itself within reality (i.e., within context). As a result, organisms do not need to be “informed” by their context, and they do not need to process “information” about their context. Organisms, as embodiments of context, are naturally and necessarily “about” the contexts they embody. Therefore, what is traditionally referred to as “perception” in correspondence-driven frameworks, is conceptualized as modulation, not representation, in WST.
And while WST assumes there is something outside of the organism, as do correspondence frameworks, including ITP, WST does not assume the organism resides on the other side of that something. Rather, the organism is nested within that something. And as an embodiment of that something, it is naturally and necessarily about that something and, therefore, inherently meaning-full. In short, knowledge is not something organisms have. Rather, knowledge is something organisms are.