Madeleine Ransom

The predictive coding (PC) theory of attention identifies attention with the optimization of the precision weighting of prediction error. Here we provide some challenges for this identification. On the one hand, the precision weighting of prediction error is too broad a phenomenon to be identified with attention because such weighting plays a central role in multimodal integration. Cases of crossmodal illusions such as the rubber hand illusion and the McGurk effect involve the differential precision weighting of prediction error, yet attention does not shift as one would predict. On the other hand, the precision weighting of prediction error is too narrow a phenomenon to be identified with attention, because it cannot accommodate the full range of attentional phenomena. We review criticisms that PC cannot account for volitional attention and affect-biased attention, and we propose that it may not be able to account for feature-based and intellectual attention.

While the view that we perceive aesthetic properties may seem intuitive, it has received little in the way of explicit defence. It also gives rise to a puzzle. The first strand of this puzzle is that we often cannot perceive aesthetic properties of artworks without training, yet much aesthetic training involves the acquisition of knowledge, such as when an artwork was made, and by whom. How, if at all, can this knowledge affect our perception of an artwork’s aesthetic properties? The second strand of the puzzle arises when we widen the scope of aesthetic experience. The very same aesthetic properties that seem to require training for their perception in artworks do not appear to require training to perceive in objects of everyday aesthetic appreciation and natural phenomena. In this paper I argue that a prominent extant attempt to explain how training is compatible with aesthetic perception—cognitive permeation—is an inadequate solution. I also develop a positive view of aesthetic perception that provides a unified solution to both strands of the puzzle.

A study of culturally-embedded perceptual responses to aesthetic value indicates that learned perceptual capacities can secure compliance with social norms. We should therefore resist the temptation to draw a line between cognitive processes, such as perception, that can adapt to differences in physical environments, and cognitive processes, such as economic decision-making, that are shaped by social norms. Compliance with social norms is a result of perceptual learning when that same compliance modifies perceptible features of the physical environment.

Kendall Walton’s project in ‘Categories of Art’ (1970) is to answer two questions. First, does the history of an artwork’s production determine its aesthetic properties? Second, how – if at all – should knowledge of the history of a work’s production influence our aesthetic judgments of its properties? While his answer to the first has been clearly understood, his answer to the second less so. Contrary to how many have interpreted Walton, such knowledge is not necessary for making aesthetic judgments; perceiving an artwork as belonging to a (correct) category of art requires no art-historical knowledge whatsoever. Moreover, contextualist attempts to incorporate art-historical knowledge via the mechanism of cognitive penetration are incompatible with Walton’s claim that categories of art must be perceptually distinguishable. Here I propose a way of elaborating Walton’s view that avoids this difficulty and reconciles contextualism with aesthetic perception, the view that we perceive aesthetic properties.

Does the scope of beliefs that people can form on the basis of perception remain fixed, or can it be amplified with learning? The answer to this question is important for our understanding of why and when we ought to trust experts, and also for assessing the plausibility of epistemic foundationalism. The empirical study of perceptual expertise suggests that experts can indeed enrich their perceptual experiences through learning. Yet this does not settle the epistemic status of their beliefs. One might hold that the background knowledge of experts is the cause of their enriched perceptual experience – what is known as cognitive permeation – and so their subsequent beliefs are only mediately justified because they are epistemically dependent on this background knowledge. I argue against this view. Perceptual expertise is not the result of cognitive permeation but is rather the result of perceptual learning, and perceptual learning does not involve cognition in a way that entails cognitive permeation. Perceptual expertise thus provides a means of widening the scope of the immediately justified beliefs that experts can form.

In this paper we argue that predictive processing (PP) theory cannot account for the phenomenon of affect-biased attention prioritized attention to stimuli that are affectively salient because of their associations with reward or punishment. Specifically, the PP hypothesis that selective attention can be analyzed in terms of the optimization of precision expectations cannot accommodate affect-biased attention; affectively salient stimuli can capture our attention even when precision expectations are low. We review the prospects of three recent attempts to accommodate affect with tools internal to PP theory: Miller and Clark’s (2018) embodied inference; Seth’s (2013) interoceptive inference; and Joffily and Coricelli’s (2013) rate of change of free energy. In each case we argue that the account does not resolve the challenge from affect-biased attention. For this reason, we conclude that prediction error minimization is not sufficient to explain all mental phenomena, contrary to the claim that the PP framework provides a unified theory of all mental phenomena or the brain ‘s cognitive functioning. Nevertheless, we suggest that empirical investigation of the interaction between affective salience and precision expectations should prove helpful in understanding the limits of PP theory, and may provide new directions for the application of a Bayesian perspective to perception.

It has recently become popular to suggest that cognition can be explained as a process of Bayesian prediction error minimization. Some advocates of this view propose that attention should be understood as the optimization of expected precisions in the prediction-error signal (Clark, 2013, 2016; Feldman & Friston, 2010; Hohwy, 2012, 2013). This proposal successfully accounts for several attention-related phenomena. We claim that it cannot account for all of them, since there are certain forms of voluntary attention that it cannot accommodate. We therefore suggest that, although the theory of Bayesian prediction error minimization introduces some powerful tools for the explanation of mental phenomena, its advocates have been wrong to claim that Bayesian prediction error minimization is ‘all the brain ever does’.

Perceptual learning is an enduring change in the perceptual system – and our resulting perceptions – due to practice or repeated exposure to a perceptual stimulus. It is involved in the acquisition of perceptual expertise: the ability to make rapid and reliable high-level categorizations of objects unavailable to novices. Attentional weighting is one process by which perceptual learning occurs. Advancing our understanding of this process is of particular importance for understanding what is learned in perceptual learning. Attentional weighting seems to favour the hypothesis that what experts learn is simply to better direct their attention to elements of the stimulus relevant to making post-perceptual categorizations. Here I argue against this hypothesis. Attentional weighting is an integral component of the process that allows experts to come to represent high-level properties in perceptual experience.

The fallacies, as traditionally conceived, are wrong ways of reasoning that nevertheless appear attractive to us. Recently, however, Woods (2013) has argued that they don’t merit such a title, and that what we take to be fallacies are instead largely virtuous forms of reasoning. This reformation of the fallacies forms part of Woods’ larger project to naturalize logic. In this paper I will look to his analysis of the argumentum ad hominem as a case study for the prospects of this project. I will argue that the empirical literature on implicit bias presents a difficulty for the reformation of the ad hominem as cognitively virtuous. Cases where implicit bias influences our assessment of the truth or claim or argument are instances of ad hominem reasoning, and these qualify as fallacious on Woods’ own definition.

The acquaintance debate in aesthetics has been traditionally divided between pessimists, who argue that testimony does not provide others with aesthetic knowledge of artworks, and optimists, who hold that acquaintance with an artwork is not a necessary precondition for acquiring aesthetic knowledge. In this paper I propose a reconciliationist solution to the acquaintance debate: while aesthetic knowledge can be had via testimony, aesthetic judgment requires acquaintance with the artwork. I develop this solution by situating it within a virtue aesthetics framework based on Ernest Sosa's virtue epistemology. I go on to apply the solution to the debates on moral testimony and expert testimony. An interesting variant on Gettier cases emerges: cases in which subjects have knowledge, but it has been formed by the wrong competence.

Attempts to engineer a generally intelligent artificial agent have yet to meet with success, largely due to the (intercontext) frame problem. Given that humans are able to solve this problem on a daily basis, one strategy for making progress in AI is to look for disanalogies between humans and computers that might account for the difference. It has become popular to appeal to the emotions as the means by which the frame problem is solved in human agents. The purpose of this paper is to evaluate the tenability of this proposal, with a primary focus on Dylan Evans’ search hypothesis and Antonio Damasio’s somatic marker hypothesis. I will argue that while the emotions plausibly help solve the intracontext frame problem, they do not function to solve or help solve the intercontext frame problem, as they are themselves subject to contextual variability.