1 Introduction
Feyerabend (1981), Hanson (1958), and Kuhn (1962) argued that what someone thinks, hopes, or desires determines what they perceive. Perception became theory-laden, conceptually modulated, and cognitively penetrated (CP). (The relations among theory-ladenness, conceptual modulation, and cognitive penetrability are not as clear as they seem, but as I have argued [Raftopoulos 2009], under some independent assumptions, the equivalence holds true.) Furthermore, Sellars (1956) attacked one of the main âdogmasâ of classical empiricism, to wit, the view that perception functions independently of concepts and delivers to us the world in its own guise without any conceptual influences. This âgivenâ, empiricists thought, could be used as a neutral basis to provide justification and determine the truth of both perceptual beliefs and scientific theories, exactly because by being free from any conceptual influences it reflects only the environment. The rejection of this assumption undermined the epistemic role of perception in justifying perceptual beliefs since the fact that prior beliefs affect perception makes it possible that prior beliefs , by shaping the percept (i.e., the content or character of perceptual experience)1 that is subsequently used to support rationally a belief, provide indirectly (through the intermediate perception) support either for themselves or for beliefs that are congruent with them. This is a form of confirmation bias that is, epistemically highly problematic, or, at worst, a simple vicious justificatory circle whereby a belief justifies itself through the intermediary of perception; this clearly undermines the epistemic role of perception .
Siegel (2016, 2) explains the threat that the cognitive effects on perception pose for the traditional view concerning the epistemic role of perception , according to which perception yields receptively and passively a given that is neutral with respect to oneâs conceptual frameworks and can be used as a neutral ground to justify beliefs and relate one with the world.
Experiences are not uniformly receptive. They are not always a landing pad for information (or misinformation) that tumbles in along arational channels, naively open to the objects, properties, and events that are there for us to perceive. Experiences can fail to be receptive in these ways, not because they are off the grid of rational assessment, as an undirected âraw feelâ would be, but because the grid has a place for experiences that arise in some of the same ways as irrational beliefs do. And this opens the possibility that the notions of ill-foundedness that we use to convict some beliefs might convict some experiences as well.
Indeed, if prior beliefs affect perceptual processing , one might worry about this affects the justificatory role of perception. It is intuitive to argue that if the belief that X is F causally affects the perceptual processing of a visual scene in which an X is present and as a result of this process, viewers have an experience with content âX is Fâ on which they subsequently base their belief that X is F, one should suspect that the role of the prior belief in affecting the content of perception diminishes the rational support for the perceptually based belief undermining the rational standing of the belief; the belief is epistemically compromised. Siegel (2011, 2013a, 702â703) calls the phenomenon in which CP leads to epistemically compromised beliefs , the downgrade principle. The experience E that results from, among other factors, the causal influence of a prior belief and, owing to this casual influence, has its justificatory role diminished, is epistemically downgraded.
The underdetermination of the justificatory role of perception paved the way to constructivism . Constructivists claim that mind-independent objects are epistemically inaccessible. Epistemological Constructivism undermines realism by arguing that our experience of the world is mediated by our concepts. One cannot examine directly which aspects of objects belong to them independently of our conceptualizations because perception is cognitively penetrable and, thus, they cannot compare their representations of objects and the mind-independent objects these representations represent. This view clashes with epistemological realismâs thesis that perception relates mind-independent objects and us (Kitcher 2001).
I draw a distinction between cognitive effects on perception and CP. The reason is that it is widely accepted (see the next chapter for a discussion) that not all cognitive effects on perception constitute cases of CP. It is also true that the cognitive effects on perception that are not cases of CP may threaten the justificatory role of perception. CP is not the only sort of cognitive influences on perception that could affect the epistemic role of perception ; other cognitive effects that are not usually taken to constitute cases of CP could downgrade perception. When, for example, cognitively driven attention causes the eyes to focus on a certain part of the environment selecting the visual scene to be perceived, or when attention selects in the same way a particular object among the objects in a visual scene boosting its perceptual processing this may affect the epistemic role of perception . It could make one ignore relevant evidence , or give credence to irrelevant evidence , so that an initial hypothesis concerning the identity of the object (that is constructed in late vision as we shall see in Chapter 5) be supported against the testimony of the environment. At the same time, in view of the fact that this effect of attention introduces an external link in the causal chain from the penetrating cognitive state to perception is not deemed to be a case of CP, a claim endorsed by most philosophers with few exceptions.
I will contain my discussion on the epistemic effects of CP. I will return in the next two chapters to discuss the indirect cognitive effects on perception in detail, and I will argue that the epistemic repercussions of indirect cognitive effects on perception differ from the epistemic effects of CP since they affect a different perceptual stage. Cognition directly affects late vision , rendering it CP, and may downgrade the percept , but it does not affect directly early vision . Indirect cognitive effects do impact early vision but do not affect its epistemic role. The indirect cognitive effects , by selecting the environmental scene to be perceptually processed, impact on the epistemic role of perception and may downgrade this role. Siegel (2016, 5) calls this sort of effects âthe selective mode â in which perception can depend on cognition (Siegel talks about desires but her discussion can be extended to all sorts of cognitive effects on perception), and in other works she calls them the effects of global selection (Siegel 2013a, 2016). The downgrade, however, is different in the direct and indirect cases, since the indirect cognitive effects do not affect perceptual processing itself but only the selection of perceptual input, while the direct cognitive effects affect perceptual processing itself. As we shall see, this is a significant difference because the epistemic role of early vision , which is influenced by indirect cognitive effects only, is not affected by these indirect cognitive effects . In contrast, late vision , owing to the fact that is directly affected by cognition, has its epistemic role influenced by cognition.
This chapter concerns the repercussions of the CP of perception for the epistemic role of perception in founding empirical beliefs , which calls for a definition of CP first. CP is thought to encompass cognitive influences on perception, where cognition is widely understood so as to include emotive states, such as desires, hopes, etc., but since not all cognitive effects on perception are considered to be cases of CP, a principled way to distinguish those cognitive effects that do signify the CP of perception from those that do not is required. I will discuss ways to make this distinction in the next chapter, where I attempt to define CP. In this chapter, I use a very generic notion of CP that is also used by Siegel (2013a, 2013b, 2016) and Pylyshyn (1999), according to which CP occurs when some cognitive states affect perceptual processing itself and not some pre-perceptual or post-perceptual stage. In the former case, cognition just selects the where or what one focuses and takes in perceptually, while in the latter case cognition affects the interpretation of the perceptual output. Should this be the case, the contents of the affecting cognitive states influence the contents of the affected perceptual states and some sort of semantic relation between the two is established. Macpherson (2012) and my self (Raftopoulos 2009) have argued that when this happens, concepts enter the perceptual contents.
Discussions concerning the effects of the CP of perception for the epistemic role of perception in grounding perceptual beliefs center on whether the cognitive effects affect the justificatory role of perception and, especially, on whether the cognitive effects hinder the epistemic role of perception either by rendering perception less sensitive to the data and, thus, less reliable, for those with externalistic epistemological inclinations, or, for those with epistemological internalistic tendencies, by introducing an illicit etiology in perception whereby the percept is the result of an illicit, ill-founded perceptual inference, which makes the percept to which the inference issues the result of an irrational process.
There are many views concerning the way perception justifies perceptual beliefs that are roughly divided into two main categories; those that fall within internalism and those that fall within externalism . According to internalism , the justification of perceptual beliefs by perception is independent of truth-related factors. Externalists reject this thesis, tying perceptual justification to externalist , relational factors that are truth-related. The two camps differ on the way they interpret, and account for, the problems that CP engenders for the epistemic role of perception . The disagreement follows mainly from a difference about the content of mental, in general, and perceptual, in particular, states.
For the internalist, perceptual content is inherently intrinsic to the viewer and does not constitutively depend on the viewerâs relation to the environment; the environment is causally implicated in the formation of this content but it is not constitutively involved in this content. For the externalist , the perceptual content is inherently extrinsic, that is, it constitutively depends on the viewerâs relation to the environment at the time of the viewing act. For some externalists , the representational content of perception (throughout this book I assume that perception has representational content) includes both phenomenal content, which is the phenomenal character (or part ...