AARON BAKER

aaron@aajbaker.com

Home
Interests
Positions
Research
and CV
Computational Models of Cognition
One of my primary interests in the field is modeling cognitive processes computationally. Computational models allow us to think critically about how we believe people incorporate – or fail to incorporate – information to make decisions. In order to build these models, we must formalize what is motivating our subjects, influencing them, and what information is important and available to them. Not only does this give us the fun opportunity to build dynamic programs that help us understand and visualize these cognitive systems, but it also coerces us into formalizing what we think is happening beneath the hood.

I have only scratched the surface of designing these models myself (though I have built many programs in my AI course at Cal), but it is one of my primary interests moving forward. In order to orient myself to the work that's already been done and do the necessary background to build my norms model, I made this guide to computational models of social cognition and decision-making. It mostly follows James Bot, my name for the agent that learns to add layers onto its decision-making process.
James Bot
Social Cognition
Scenario: You're in Ikea perusing the furniture and it's time to leave. Only one problem–– someone removed the arrow decals on the floor and you're completely, hopelessly lost. Thinking hard, you remember that the way out is directly ahead, but it seems like all the other people in the store are heading the other direction. Assuming they're all heading for for freedom, you follow the pack. In the end, the crowd was going for the meatballs at the foodcourt, leaving you to find a new path out.

In this scenario, you decided that the suggestion of the crowd's behavior overwhelmed the certainty of your personal hunch, but ultimately made a false inference about their aggregate goal. Humans lead incredibly social lives, and many of our behaviors are based in what we observe from others. But which behaviors specifically? And how much do we weigh what we learn from others versus what we piece out ourselves? Also, how do we selectively weigh information coming from different kinds of sources? For example, would you have decided to join the Ikea crowd if it was half as large? Or if it consistented entirely of your family members?
I would like to find out, and I will get out of this Ikea.
Ikea
Social implications of Computing
Were we destined to live in an optimal world? Are we optimized to live in the world we're destined for? Computing has become a pervasive part of our society and contributes real structure to it. In many ways it has improved our lives. But it also has unintended consequences, warping our perceptions of the world and our place in it.

While teaching introductory computer science, we emphasized reflections on the social implications of computing technologies, believing that it was important for a new generations of engineers to be conscious of the downstream consequences of their programs. As a cognitive scientist, I find it important and insightful to look into the ways our psychology can be at odds with the computational forces around us.
Atlas