Risky decision-making in men who have sex with men
As part of SAND Lab's larger project on risky decision-making, I'm analyzing an intervention game that gives men who have sex with men the opportunity to make decisions about safe and risky sexual choices within an fMRI scanner. The game is designed with a sex-positive, pro-condom message to encourage players to make safer choices in the future.
In my work I'm examining the neurological and behavioral correlates of those choices.
These results are currently being written up for publication. Preliminary results are available here.
Motivated decision-making simulation
Available for download at https://github.com/bjsmith/motivation-simulation
Our actions in everyday situations are determined by a complex combination of the stimuli in our environment, prior model-based and model-free learning, and current proprioceptive need relevant to environmental stimuli. I built a computational simulation of decision-making based on this model. The model can be easily applied to any situation in which people face parallel or serial dilemma about various decisions, where prior learning, current proprioceptive state, and/or environmental stimuli are relevant.
MVPA analysis of fMRI data in R
Available for download at https://github.com/bjsmith/r-mvpa
I've written a wrapper package to facilitate a large number of machine learning methods, through the caret package, being applied to fMRI datasets, to build a convenient, easy-to-use, useful pipeline from analysis to visualization. As Revolution R Open and other tools improve up the day-to-day processing power of R scripts, and neuroscientists seek new ways to quickly identify patterns in their data, R is increasingly a viable option for analysing fMRI data.
Self-interested Moral judgments
With the help of Sara Doyle and Amit Oldak, I'm investigating the role of self interest in moral judgments; particularly how self-interest could shape judgements even about unrelated moral judgements by analogy.
A Nietzschean hypothesis: the moralization of power and equality visible through natural language processing
With collaborators in the USC Dornsife Department of Classics, I'm working on a natural language processing project examining the historical and cultural origins of our society's psychology of morality. In 1489 Latin documents available and written between 99 BCE and 400 CE, documents that discussed power in a positive sense tended to avoid discussion of morality compared to documents that did not discuss power so much.
No comments:
Post a Comment