The Perceptual Belief Problem

The explainability of robotic systems depends on people’s ability to reliably attribute perceptual beliefs to robots, i.e., what robots know (or believe) about objects and events in the world based on their perception. However, the perceptual systems of robots are not necessarily well understood by the majority of people interacting with them. In this article, we explain why this is a significant, difficult, and unique problem in social robotics. The inability to judge what a robot knows (and does not know) about the physical environment it shares with people gives rise to a host of communicative and interactive issues, including difficulties to communicate about objects or adapt to events in the environment. The challenge faced by social robotics researchers or designers who want to facilitate appropriate attributions of perceptual beliefs to robots is to shape human–robot interactions so that people understand what robots know about objects and events in the environment. To meet this challenge, we argue, it is necessary to advance our knowledge of when and why people form incorrect or inadequate mental models of robots’ perceptual and cognitive mechanisms. We outline a general approach to studying this empirically and discuss potential solutions to the problem.

Skeptical appeal: the source-content bias

Radical skepticism is the view that we know nothing, or at least next to nothing. Nearly no one actually believes that skepticism is true. Yet it has remained a serious topic of discussion for millennia and it looms large in popular culture. What explains its persistent and widespread appeal? How does the skeptic get us to doubt what we ordinarily take ourselves to know? I present evidence from two experiments that classic skeptical arguments gain potency from an interaction between two factors. First, people evaluate inferential belief more harshly than perceptual belief. Second, people evaluate inferential belief more harshly when its content is negative (i.e. that something is not the case) than when it’s positive (i.e. that something is the case). It just so happens that potent skeptical arguments tend to focus our attention on negative inferential beliefs, and we are especially prone to doubt that such beliefs count as knowledge. That is, our cognitive evaluations are biased against this specific combination of source and content. The skeptic sows seeds of doubt by exploiting this feature of our psychology.

Perception and Perceptual Belief

This chapter shows how perception is experiential, relational, and representational: in broad terms, a phenomenally representational, discriminative, non-deviant causal relation to an object. In seeing, we have visual experience; in visual experience that is perceptual and not merely sensory, as in hallucinations, some object is visually represented as having certain properties; and genuinely seeing an object entails that it exists. This view of perception is a version of realism. It is realistic about both the objects of perception and the phenomenal properties we instantiate in perceptual experience. The view leaves open, however, the ultimate ontological status of those objects and properties. The properties, however, must be understood to have a character that enables us both to experience what, phenomenologically, it is like to see and what, externally, the things seen are like physically.

The Myth of Stochastic Infallibilism

There is a widespread attitude in epistemology that, if you know on the basis of perception, then you couldn't have been wrong as a matter of chance. Despite the apparent intuitive plausibility of this attitude, which I'll refer to here as “stochastic infallibilism”, it fundamentally misunderstands the way that human perceptual systems actually work. Perhaps the most important lesson of signal detection theory (SDT) is that our percepts are inherently subject to random error, and here I'll highlight some key empirical research that underscores this point. In doing so, it becomes clear that we are in fact quite willing to attribute knowledge to S that p even when S's perceptual belief that p could have been randomly false. In short, perceptual processes can randomly fail, and perceptual knowledge is stochastically fallible. The narrow implication here is that any epistemological account that entails stochastic infallibilism, like safety, is simply untenable. More broadly, this myth of stochastic infallibilism provides a valuable illustration of the importance of integrating empirical findings into epistemological thinking.

Are Perceptual Reasons the Objects of Perception?

This paper begins with a Davidsonian puzzle in the epistemology of perception and introduces two solutions to that puzzle: the Truth-Maker View (TMV) and the Content Model. The paper goes on to elaborate TMV, elements of which can be found in the work of Kalderon (2011) and Brewer (2011). The central tenant of TMV is the claim that one’s reason for one’s perceptual belief should, in all cases, be identified with some item one perceives which makes the proposition believed true. I defend an argument against TMV which appeals to (a) the claim that the reason for which one believes should always to be identified with the explanans of the rationalizing explanation to which one’s belief is subject and (b) the claim that the explanantia of rationalizing explanations must be identified with truths. I finish by replying to two objections to the argument.

The Phenomenal Presence of Perceptual Reasons

This chapter argues for Phenomenal Rationalism about perceptual experiences: the claim that our basic awareness of reasons for perceptual belief is phenomenal and non-conceptual. The main idea is that, from the inside, perceptual experiences seem to be reason-giving insofar as they seem to be relations to, and determined by, external objects and their features. The argumentation centres partly on the claim that assuming the phenomenal presence of the relationality and determination of perceptual experiences provides the best explanation of why so many good philosophers were convinced of the soundness of the argument from hallucination. That we enjoy phenomenal access to reasons prior to any normative beliefs also helps to reconcile the Humean insight that infants (who lack concepts like ‘reason for’) can be motivated to act or form attitudes with the Kantian insight that motivation is a matter of recognizing and responding to reasons.

Phenomenal Presence

Many different features figure consciously in our perceptual experiences, in the sense that they make a subjective difference to those experiences. These features range from colours and shapes to volumes and backsides, from natural or artefactual kinds to reasons for perceptual belief, and from the existence and externality of objects to the relationality and wakefulness of our perceptual awareness of them. The topic of this collection of essays is the different ways in which features like these can be phenomenally present in perceptual experience. In particular, the focus is on features that are less often discussed, and the perceptual presence of which is less obvious because they are out of view or otherwise easily overlooked, features given in a non-sensory manner, and features that are categorical in the sense that they pertain to all perceptual experiences alike (such as their justificatory power, their wakefulness, or the externality of their objects). The book is divided into four parts, each dealing with a different kind of phenomenal presence. The first addresses the nature of the presence of perceptual constancies and variations, while the second investigates the determinacy and ubiquity of the presence of spatial properties in perception. The third part deals with the presence of hidden or occluded aspects of objects, while the last part of the volume discusses the presence of categorical aspects of perceptual experience. Together, the contributions provide a thorough examination of which features are phenomenally present in perception, and what it is for them to figure in experience in this way.

A Better Disjunctivist Response to the ‘New Evil Genius’ Challenge

This paper aims for a more robust epistemological disjunctivism (ed) by offering on its behalf a new and better response to the ‘new evil genius’ problem. The first section articulates the ‘new evil genius challenge’ (neg challenge) to ed, specifying its two components: the ‘first-order’ and ‘diagnostic’ problems for ed. The first-order problem challenges proponents of ed to offer some understanding of the intuition behind the thought that your radically deceived duplicate is no less justified than you are for adopting her perceptual beliefs. In the second section, the author argues that blamelessness explanations are inadequate to the task and offer better explanations in their place—that of ‘trait-level virtue’ and ‘reasonability’. The diagnostic problem challenges proponents of ed to explain why it is that classical internalists disagree with them about how to interpret new evil genius considerations. The proponent of ed owes some error theory. The author engages this problem in the third section, arguing that classical internalists are misled to overlook disjunctivist interpretations of new evil genius thinking owing to a mistaken commitment to a kind of ‘vindicatory’ explanation of proper perceptual belief.