Selected Publications

Historically, facial expression research has followed from discrete emotion theories, which posit a limited number of distinct affective states that are represented with specific patterns of facial action. Much less work has focused on dimensional features of emotion (e.g., positive and negative affect intensity). We use computer-vision and machine learning (CVML) to identify patterns of facial actions in 4,648 video recordings of 125 human participants. Our results show that CVML can both (1) determine the importance of different facial actions that human coders use to derive positive and negative affective ratings when combined with interpretable machine learning methods, and (2) efficiently automate positive and negative affect intensity coding on large facial expression databases.
In PLOS ONE., 2019.

The Iowa Gambling Task (IGT) is widely used to study decision‐making within healthy and psychiatric populations. Here, we propose the Outcome‐Representation Learning (ORL) model, a novel model that provides the best compromise between competing models. We test the performance of the ORL model on 393 subjects’ data collected across multiple research sites, and we show that the ORL reveals distinct patterns of decision‐making in substance‐using populations.
In Cog. Sci., 2018.

Machine learning is becoming more widely accepted across a broad range of scientific disciplines. The easyml (easy machine learning) package lowers the barrier to entry to machine learning, allowing users to apply best-practice regression and classification schemes to their own data.
In BioRxiv, 2017.

There is a growing interest in psychology to apply advanced computational models to decision making data collected from psychiatric populations to better understand maladaptive choice patterns. However, there are currently no easy-to-use tools for those who may not have the sophisticated mathematical/programming background to use such methods. Here, we present an R package which can fit an array of decision making models to a variety of different tasks with a single line of code.
In CPSY, 2017.

Recent Publications

Using computer-vision and machine learning to automate facial coding of positive and negative affect intensity

Details PDF Code

The Outcome‐Representation Learning Model: A Novel Reinforcement Learning Model of the Iowa Gambling Task

Details PDF Code

Easyml: Easily Build And Evaluate Machine Learning Models

Details PDF Code

The Indirect Effect of Emotion Regulation on Minority Stress and Problematic Substance Use in Lesbian, Gay, and Bisexual Individuals

Details PDF

Revealing Neurocomputational Mechanisms of Reinforcement Learning and Decision-Making With the hBayesDM Package

Details PDF Code

Recent & Upcoming Talks

Recent Posts

More Posts

The Reliability Paradox Defining Reliability In 2017, Hedge, Powell, and Sumner (2017) conducted a study to determine the reliability of a variety of of behavioral tasks. Reliability has many different meanings throughout the psychological literature, but what Hedge et al. were interested in was how well a behavioral measure consistently ranks individuals. In other words, when I have people perform a task and then measure their performance, does the measure that I use to summarize their behavior show high test-retest reliability?


Introduction In this post, we will explore frequentist and Bayesian analogues of regularized/penalized linear regression models (e.g., LASSO [L1 penalty], Ridge regression [L2 penalty]), which are an extention of traditional linear regression models of the form: \[y = \beta_{0}+X\beta + \epsilon\tag{1}\] where \(\epsilon\) is the error, which is normally distributed as: \[\epsilon \sim \mathcal{N}(0, \sigma)\tag{2}\] Unlike these traditional linear regression models, regularized linear regression models produce biased estimates for the \(\beta\) weights.


1. Goals of Paramter Estimation When estimating paramters for a given model, we typically aim to make an inference on an individual’s underlying decision process. We may be inferring a variety of different factors, such as the rate at which someone updates their expectations, the way that someone subjectively values an outcome, or the amount of exploration versus exploitation that someone engages in. Once we estimate an individual’s parameters, we can compare then to other people or even other groups of people.




A cross-platform Python toolbox for analyzing facial expression data.


An R/Python toolbox for easily fitting a variety of machine learning models.


An R toolbox for fitting an array of decision making models with hierarchical Bayesian analysis.


Abnormal Psychology

I am currently teaching PSYCH 3331 at The Ohio State University. My course takes a dimensional and developmental perspective, and is divided mainly into sections for internalizing, externalizing, and psychotic forms of psychopathology. Further, I use principles of active learning to engage students during class. My course design was inspired largely by Ziv Bell, who developed a “flipped class” curriculum for his students to better understand and apply principles of abnormal psychology to everyday life.