BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:NeuroLunch: Cheng Tang (Jazayeri Lab) & Cyn Fang (Kanwisher La
 b)
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260915T091402Z
UID:tag:localist.com\,2008:EventInstance_50762342147776
DTSTART:20251020T160000Z
DTEND:20251020T170000Z
DESCRIPTION:Speaker:  Cheng Tang (Jazayeri Lab)\n\nTitle: Position\, Direct
 ion\, and Time: Three-Way Factorization Underlies Vector Compositionality 
 in the Dorsomedial Frontal Cortex\n\nAbstract:\n\nThe ability to flexibly 
 combine elementary concepts into novel\, complex ones—known as compositi
 onal generalization—is a hallmark of intelligence in both humans and non
 -human primates. Yet\, the neural principles that enable such compositiona
 lity remain poorly understood. To probe this question\, I trained two maca
 que monkeys on a 2D vector addition task that required adding spatial coor
 dinates and dynamic flow-fields defined by direction and duration\, in a v
 ector arithmetic fashion. After learning a subset of combinations\, the an
 imals were tested on novel pairings to assess compositional generalization
 . Electrophysiological recordings from the dorsomedial frontal cortex (DMF
 C) revealed that successful generalization was consistently associated wit
 h factorized neural representations—independent encoding of coordinate a
 nd vector components. Conversely\, error trials showed non-factorized\, en
 tangled representations. Moreover\, within the flow-field representation\,
  direction and time were factorized at the single-trial level. Together\, 
 these findings suggest that three-way factorization—of position\, direct
 ion\, and time—within the dorsomedial frontal cortex (DMFC) provides a n
 eural basis for vector compositionality in the primate brain. More broadly
 \, the capacity for compositional generalization may critically depend on 
 which task variables are factorized within neural representations\, highli
 ghting the importance of representational structure in flexible cognition\
 n\n \n\nSpeaker: Cyn Fang (Kanwisher Lab)\n\nTitle: Is it food?\n\nAbstrac
 t: Three recent publications reported a component of the fMRI response in 
 the human ventral visual pathway that responds selectively to images of fo
 od. However\, all three studies were based on the same Natural Scenes Data
 base (NSD)\, in which high-level categories like food are correlated in th
 e image set with other properties of the image such as the color of object
 s or distance of the scene. To test whether the reported visual food selec
 tivity might reflect these or other correlates of food images rather than 
 (or in addition to) food per se\, we constructed novel stimuli that manipu
 lated these properties on both food and nonfood images. Our results indica
 te that the previously reported food component is not strictly food-select
 ive\, and that it is driven in part by contextual or material features unr
 elated to food.
GEO:42.362302;-71.091766
LOCATION:Building 46\, 3310
SUMMARY:NeuroLunch: Cheng Tang (Jazayeri Lab) & Cyn Fang (Kanwisher Lab)
URL;VALUE=URI:https://calendar.mit.edu/event/copy-of-neurolunch-daoyuan-qia
 n-fiete-lab-josefa-scherrer-fee-lab
END:VEVENT
END:VCALENDAR