For video footage from past you can visit the individual event pages, or go to our YouTube Channel.
To filter by event category, click on the event category link in the table below or use the menu on the right.
List of Past Events
Objects, Object Files, and Object Principles
E. J. Green
Thursday, September 10, 2015, 01:00pm - 02:00pm
Graduate Student, Rutgers University, Department of Philosophy and Center for Cognitive Science
Many authors have posited the existence of an “object file” system, which is recruited by visual processes involved in the selection and tracking of individual objects. Moreover, several theorists have proposed that the object file system internalizes principles that specify what counts as an object, and that it relies on these principles in order to track an object over time. In virtue of internalizing these principles, the object file system is purportedly “keyed” or “tuned” to a particular kind of entity.
In this talk I’ll consider a recent view of this kind, found in Tyler Burge (2010) and Susan Carey (2009), according to which mid-level vision selects and tracks objects in accordance with the principles of three-dimensionality and cohesion. I contrast the approach found in Burge and Carey with a different view, found in the work of a number of vision scientists, on which the visual system selects and tracks objects in accordance with familiar Gestalt criteria of perceptual organization. I’ll show that the criteria of perceptual organization are, in general, far more permissive than the three-dimensionality and cohesion principles—more objects satisfy the former principles than satisfy the latter. Thus, I call Burge and Carey’s position the Restrictive View, and I call the alternative position the Permissive View.
I’ll argue that the available evidence (including the evidence often cited in support of the Restrictive View) is consistent with, and may even support, the Permissive View. Critically, processes of selection and tracking in mid-level vision appear not to impose a three-dimensionality requirement, and there is little evidence that they make a sharp distinction between discrete, cohesive objects and perceptual groups. I then argue that work on the visual selection and tracking of so-called “nonmaterial” entities, such as holes and negative parts (i.e., “bites” or “indentations” out of a figure), may supply further support for the Permissive View over its more restrictive competitor.