Hi 👋, 

I'm Athul Sudheesh. 

I'm an Applied Statistician and Researcher who specializes in building explainable models of human cognitive functioning in realistic, real-world contexts. Rather than relying on controlled laboratory experiments, I specialize in analyzing and modeling human behavior as it naturally occurs in everyday settings. 


What is cognitive functioning?

Cognition functioning is an umbrella term used to refer multiple mental abilities like learning, thinking, reasoning, remembering, problem-solving, decision-making, and attention. 

What are explainable models?

Explainable models aim to provide causal explanations of why things happen the way they do in the phenomena we are interested in.  These are constrative to the black-box modeling approaches widely prevalent in today's AI/ML apps. 

Why building explainable models of cognitive functioning are hard?

Building explainable models of cognitive functioning is challenging because the factors (skills, knowledge, abilities, emotions) underlying psychological phenomena are not directly observable (measurable). We can only make inferences about these factors from the outcomes that are observable as part of the interaction between humans and the tasks that engage their cognitive functions. Hence, these situations first require establishing a strong theory of psychological measurement for the given task before we can build explainable models of cognitive functioning.

As part of the research that I have been doing for the past five years, I have developed a new framework that  provides a unified approach to not only devise and test theories of psychological measurement but also to build and test causal theories of cognitive functioning. This new framework combines ideas from graph theory, probability theory, cognitive task design, and philosophy of measurement.