The Center for Cognitive and Brain Sciences Student Organization will be hosting a workshop—titled Number, Cognition, and Perception—on March 13 and 14 in the Harris and Hill Seminar Room (University Hall 353). Please find the speakers, titles, and abstracts of the presentations below:
March 13 (Friday)
3:30 PM – 5:30 PM
Jacob Beck (Philosophy, York University)
Presentation Title: Softening the Perception–Cognition Border
Abstract: Approaches to the perception–cognition distinction tend towards two extremes. On the one hand, many embrace a hard border, treating perception and cognition as mutually exclusive, non-overlapping categories. On the other hand, eliminativists deny that any principled, theoretically useful distinction exists between perception and cognition. I will offer a third way, describing a principled but soft border between perception and cognition. This non-exclusive approach differentiates perception from cognition while allowing that they overlap. I’ll motivate this soft-border approach with three examples: approximate number, language, and perception/imagery. I’ll argue that each involves capacities that belong to both perception and cognition. This problematizes hard-border approaches, including architectural approaches, format approaches, and my own previous stimulus-dependence approach. I’ll then explain how the stimulus-dependence approach can be modified to soften the border between perception and cognition.
5:30 PM – 6:30 PM
Reception
March 14 (Saturday)
8:30 AM – 9:30 AM
Continental Breakfast
9:30 AM – 10:45 AM
Shao-An Hsu (Philosophy, The Ohio State University)
Presentation Title: An Epistemological Argument for Numerical Perception
Abstract: Can we visually perceive the exact number of speckles on a hen? I think we can. In contrast with the mainstream position that, for cardinalities greater than four, perceptual states represent them only approximately (Dehaene, 2011), and with the view that exact cardinality is available perceptually only up to four via parallel individuation (Carey, 2009), my view allows for exact perceptual representation of larger cardinalities. Although we cannot report the exact number of speckles without counting, that doesn’t imply that we fail to see each speckle and their total number. For how could we know the exact number of speckles empirically unless we see how many there are? Our visual experience is what grounds our knowledge in this case. I argue that our visual experience already contains the information needed to know the exact number of speckles; what is required is a suitable method for accessing that information. The method is counting. Counting is the procedure that extracts cardinality content already implicit in a perceptual state and converts it from an iconic format into a discursive format, thereby making it ready for rational thoughts.
11 AM – 12:15 PM
Richard Samuels (Philosophy, The Ohio State University)
Collaborators: Joonkoo Park (Psychology, The University Massachusetts, Amherst)
Eric Snyder (Philosophy, Ashoka University)
Presentation Title: What is a number, that a man can (not) perceive it?
Abstract: At least since Warren McCulloch’s seminal paper (1961), there’s been a widespread recognition that scientific theorizing about human numerical capacities is sensitive to broadly philosophical issues regarding the nature of number. Yet we maintain that a) the vast and intricate empirical literature on numerical perception routinely invokes inadequate and conflicting assumptions regarding number and that b) these assumptions have exerted a negative influence on several ongoing empirical debates.
The purpose of the paper is threefold. First, we provide several illustrations of how philosophical assumptions about number have exerted an influence on empirical research on numerical perception, especially concerning the approximate number system. Second, drawing on work in linguistics and the philosophy of mathematics, we seek to clarify the relevant conceptual issues, for example, by drawing a distinction between monadic and relational notions of number, and between various notions of abstractness that figure in the literature. Finally, in the light of these distinctions and clarifications we show why some of the assumptions operative in perception research are problematic, and how some longstanding disputes turn out to be largely verbal disagreements. In doing so, we aim to clarify what genuine empirical issues remain to be addressed.
12:30 PM – 2:00 PM
Lunch Break
2:00 PM – 3:15 PM
Sami Yousif (Psychology, The Ohio State University)
Presentation title: Number, adaptation, and perception
Abstract: If you stare at a purple image that suddenly turns grey, you will experience a repulsive aftereffect: You will not experience grey, but green. If you stare at a waterfall with downward motion and then you stare at something stationary, you will experience another repulsive aftereffect: You will now experience upward movement. According to an orthodox view in vision science, these adaptive aftereffects are not limited to lower-level visual features like color and motion but also proliferate to higher-level features like variance, causality, and number. These instances of high-level adaptation are thereby said to answer deep, otherwise intractable questions about the nature and content of visual processing — but is this true? In recent work, my lab has questioned this orthodoxy. Here, I'll begin by considering number adaptation as a paradigmatic case of high-level visual adaptation, reiterating recent arguments against the existence of number adaptation and offering additional reasons for skepticism. I'll then argue that even if number adaptation were to exist, it could not serve the roles expected of it in mainstream philosophy and vision science since adaptation of the proposed sort fails to usefully demarcate perceptual content. Despite these conclusions, I'll briefly mention some better reasons to believe that number is represented by the visual system.
3:30 PM – 5:30 PM
Sam Clarke (Philosophy, The University of Southern California)
Presentation Title: Mad-Dog Nativism, Neural Networks, and The New Riddle of Concept Learning
Abstract: This paper contrasts two arguments for concept nativism. The first is Fodor’s infamous argument for “mad-dog nativism”, the view that all or virtually all concepts are innate. I argue that recent work in computer science undermines Fodor’s argument, but not because it reveals Fodor’s conclusion to be absurd or untenable (as many commentators assume). The real reason why Fodor’s argument is ineffective is that – while its conclusion reflects a serious empirical possibility – it motivates a view which is nativist in name only, allowing concepts to be acquired in ways which many contemporary empiricists would welcome. Against this backdrop, I proceed to develop a lesser known, second argument for concept nativism: The New Riddle of Concept Learning. I show that this argument avoids the shortcomings of Fodor’s mad-dog nativism, dictating a robust brand of concept nativism in myriad contested domains – including that of natural number and theory of mind – without spiraling into absurdity and implying that concepts like CARBURETOR are similarly innate.