Date:
Location:
The Cognitive and Neural Architecture of Visual Manipulation Memory
Hrag Pailian
Postdoctoral Fellow, Alvarez Lab
In an ever-changing world, intelligent behavior is shaped by our abilities to create and manipulate mental models of our visual surroundings. How do we update our internal representations of the dynamic world and reason beyond what we have perceptually experienced? Whereas traditional approaches towards answering this question have focused on characterizing storage processes in visual working memory, my research places emphasis on the domain-general, “working” component of this system: mental manipulation. I have combined behavioral psychophysics with developmental, comparative, and computational modeling techniques, to identify a behavioral signature of manipulation limits that are driven by the misbindings of object features. To further characterize these constraints, I have used electroencephalography and non-invasive brain stimulation to delineate the neural mechanisms and substrates underlying manipulation ability. Not only do these findings suggest independence between visual storage and mental manipulation in their neural resources, but additional behavioral data suggest that these functions operate on separate – yet competing - representations that differ in format (perceptually rich vs. abstract representations). These results have significant implications for how we conceptualize the architecture of the system, the format of its representations, and how VWM interfaces with the rest of cognition (e.g. long-term memory, quantitative reasoning, verbal comprehension, etc.).