Thomas Alrik Sørensen

There has been a long-standing debate in philosophy and psychology about the role of representation in visual perception. Here, we argue on the basis of evidence from philosophy, psychology, and neuroscience that episodic and schematic memory representations are pivotal to the visual perception of objects and scenes. In the visual perception of objects and scenes, sensory information is initially matched with object and scene templates, or schemas, in long-term memory. The most relevant representations are then selected for encoding in working memory. We furthermore argue that activations of episodic memory representations contribute to the fineness of grain of visual representations. The representational view of visual perception that emerges is what we call the "template tuning view." According to this view, prior information – specifically, long-term memories – shape the representational content of visual perception. In the final section of the chapter, we argue that unlike representational conceptions of visual perception, naïve and direct realist theories have difficulties accommodating these findings.

The basis of perception is the processing and categorization of perceptual stimuli from the environment. Much progress has been made in the science of perceptual categorization. Yet there is still no consensus on how the brain generates sensory individuals, from sensory input and perceptual categories in memory. This chapter argues that perceptual categorization is highly variable across perceivers due to their use of different perceptual strategies for solving perceptual problems they encounter, and that the perceptual system structurally adjusts to the strategies that are most successful. Despite this variability, the different sensory individuals that result when different perceivers rely on different strategies nonetheless correspond to actual external world objects/features. This view is called 'perceptual pluralism'.

Visual short-term memory (VSTM) has traditionally been thought to have a very limited capacity of around 3–4 objects. However, recently several researchers have argued that VSTM may be limited in the amount of information retained rather than by a specific number of objects. Here we present a study of the effect of long-term practice on VSTM capacity. We investigated four age groups ranging from pre-school children to adults and measured the change in VSTM capacity for letters and pictures. We found a clear increase in VSTM capacity for letters with age but not for pictures. Our results indicate that VSTM capacity is dependent on the level of expertise for specific types of stimuli.

Whether visual perceptual consciousness is gradable or dichotomous has been the subject of fierce debate in recent years. If perceptual consciousness is gradable, perceivers may have less than full access to—and thus be less than fully phenomenally aware of—perceptual information that is represented in working memory. This raises the question: In virtue of what can a subject be less than fully perceptually conscious? In this chapter, we provide an answer to this question, according to which inexact categorizations of visual input may result in a representation of the visual information in working memory that is less than fully available to the perceiver, and of which the perceiver is therefore less than fully phenomenally aware. The latter proposal is a natural extension of a theory of perception we have proposed in previous works, namely, the template tuning theory (TTT). We argue that although TTT is compatible with both gradable and dichotomous conceptions of perceptual consciousness, the available empirical evidence favours a gradable conception of perceptual consciousness.

Predictive processing models of perception take issue with standard models of perception as hierarchical bottom-up processing modulated by memory and attention. The predictive framework posits that the brain generates predictions about stimuli, which are matched to the incoming signal. Mismatches between predictions and the incoming signal – so-called prediction errors – are then used to generate new and better predictions until the prediction errors have been minimized, at which point a perception arises. Predictive models hold that all bottom-up processes are signals conveying prediction errors to higher areas, which respond by updating their predictions. We take issue with this claim and argue that object recognition requires bottom-up processing that cannot be understood in terms of prediction errors. Along the way, we expand on previous work casting doubt on the framework's ability to account for attention. Specifically, we argue that the type of attention that allows us to rapidly extract the gist of an object or scene presents an additional challenge to the predictive approach. We conclude by considering how the framework may be augmented to avoid these problems.

To study whether the distinction between introspective and non-introspective states of mind is an empirical reality or merely a conceptual distinction, we measured event-related potentials elicited in introspective and non-introspective instruction conditions while the observers were trying to detect the presence of a masked stimulus. The ERPs indicated measurable differences related to introspection in both preconscious and conscious processes. Our data support the hypothesis that introspective states empirically differ from non-introspective states