With the generous support from the Dartmouth Graduate Student Council (GSC), I was able to attend the 2019 Kavli Summer Institute in Cognitive Neuroscience (a.k.a. Brain Camp) hosted at University of California at Santa Barbara. The summer program lasted for two weeks with each week focusing on different themes. The first week was directed by Dani Bassett (U Penn) and Jörn Diedrichsen (U of Western Ontario) and focused on “Intersections Between Neural Representations and Network Models.” The second week was directed by Leah Somerville (Harvard U) and Luke Chang (Dartmouth College) and focused on the “Computational Social Neuroscience: Advances, Challenges, and New Directions.” Approximately 70 or so fellows attended the conference from around the world.

Amongst the numerous talks, the very first experience that stood out were the human neuroanatomy labs and the neuropathology video sessions led by Robert (Bob) Knight (UC Berkeley) and Robert (Bob) Rafal (U Delaware). I’ve been through multiple Neuroanatomy and Psychology classes but these duo put up a class that was far more informative and interesting than any other. Especially with their long careers in Neurology, the descriptions of the pathology really came to life while they showed videos of patients showing various symptom that they themselves tested to localize the neuroanatomical sites responsible for the problems. In addition to the neuropathology lectures, talks on representations generally hovered around the notion of studying visual representations in the brain by comparing them to layers in the deep convolutional neural networks using encoding models, decoding models, and representational similarity analyses. Talks with network models discussed the application of network and topology theories on analyzing functional and anatomical connectivity of brain regions and their insights into the organization and functioning of the brain.

The second week, notable talks included those from Read Montague (Virginia Tech), Daniela Schiller (Mt Sinai), Steve Chang (Yale U), Molly Crocket (Yale), and Matthew Apps (U Oxford). I particularly enjoyed the moral neuroscience talks that tried to model how we make decisions about whether to help others. For example, Molly Crockett shared with us a paradigm in which participants had to decide between two options which had different ratios between an electric shock and a monetary gain. The basic finding was that most people would require a greater ratio or shock to dollars than for the self ($2/shock for self but$4/shock for other). These behavior were accompanied by neural activity such that the striatum activity increased more for selfish choices while the lateral prefrontal activity encoded the conflict of a moral choice such that it was most activated when inflicting pain yielded minimal profit. Moreover, the connectivity between the lateral prefrontal cortex and the dorsolateral striatum was modulated by whether the individual was making selfish or kind decisions. The correlation between the two regions increased when participants were making harmful or nonsocial choices. I thought that research like these could be developed further in the future by looking at more naturalistic moral decisions which would provide a more ecologically valid and comprehensive model of moral decisions.