About Josh Moller‑Mara

Hey, I’m Josh Mōller-Mara. I recently graduated with a PhD in Neural Science from NYU. Before that, I graduated with a BA from UC Berkeley with a triple-major in Computer Science, Cognitive Science, and Statistics.

Much of my work involves building internal tools and systems for running experiments, collecting data, visualizing it, and otherwise using it to improve our own decision-making. I enjoy learning, teaching, learning how to learn, teaching people how to learn, and building tools to measure and improve learning. This website is mostly for experimental tools and notes I’ve made for myself.

Josh Moller-Mara smiling with the Golden Gate Bridge in the background
A photo of me from .

Interests and Research

I’m interested in neuroeconomics, decision-making, Bayesian statistics, computational reproducibility, and free software.

For my PhD, I researched the neuroeconomics of risk. Specifically, how do mice and rats make decisions under risk? And can we compare their preferences across time and between species?

Answering these questions requires a large amount of data. So, in the Erlich Lab in Shanghai, I helped to design and build a high-throughput rodent experimental system that automated the process of collecting and analyzing hundreds of thousands of behavioral choices. Here I applied my background as a Linux systems administrator, using an infrastructure-as-code approach to ensure systems were reliable and reproducible. Some of the tools I used were Ansible (and occasionally Docker), Clojure (for backend real-time analyses), ClojureScript + React/reagent (for a front-end dashboard for our system), MariaDB, RabbitMQ, and Nix (for reproducible analyses).

To infer preferences from choices, I designed and wrote Bayesian statistical models in Stan and R. For more on this check out my Society for Neuroeconomics 2021 poster.

Technology

I exclusively use free and open-source software. (Check out the GNU project’s explanation on why free software is important.) I believe that free software is more accessible (especially considering socioeconomic status) and reproducible than proprietary counterparts, and importantly it encourages mutual cooperation and the sharing of knowledge.

To this end, I use Emacs for my text editor (check out mollermara.com for blog posts on Emacs, or /emacs), GNU/Linux (NixOS) for my operating system, and programming languages like R (instead of MATLAB). Check out more of what I use on my /uses page or in my /notes/.

I also enjoy learning about and using “old” technology. At home in Berkeley, I use an old walnut desk from the 1940s and write with a dip pen. I often use my great-grandfather’s Keuffel & Esser 4088-1 pocket slide rule and a pocket-sized brass abacus. I feel sometimes that people confuse technology with whatever’s “new” or “modern”, but for me a tool isn’t valuable for being “modern” but for how it helps you. Understanding old tools has helped me value and use new tools, like using physical files has helped me learn to organize digital files.

Contact

Check out my /contact page for my contact info.

FAQ