CAREER: Validating and applying a new class of drift-diffusion models for investigating individual differences in executive control

Missouri Western State University, Saint Joseph MO

Investigators

Abstract

Executive function refers to the ability to control and direct one's own mental processes, which is crucial for everyday behavior. Understanding how and why certain individuals show impairments in executive abilities like focusing attention and inhibiting actions is important for researchers in many areas of behavioral and cognitive sciences. These functions are typically studied by measuring how quickly and accurately an individual can complete different psychological tasks. However, behavior in such tasks can be affected by multiple processes outside of executive function, making the measurement of individual differences in these functions difficult. This problem is addressed in the current research using cognitive models. These models can break down behavior into different psychological components, allowing a deeper understanding of how and why individuals differ in the ability to control their mental processes. This research will develop and test new models of executive function. These models will provide a new class of tools for investigating cognitive function and how it differs across individuals. The researcher will provide publicly-available statistical packages that allow other researchers to use them in their own studies. As executive function tasks are widely used across different domains in psychology and neuroscience, these models will broadly affect our understanding of executive control by providing tested tools for analyzing behavioral data. The primary objective of this research is to determine which computational models, existing or to be developed, provide the best theoretical account of processing in executive function tasks. Competing models will be tested against behavioral and simulated data from a range of executive function tasks meant to measure conflict processing, inhibitory control, and selective attention. Each task will include targeted manipulations of different cognitive components to provide rigorous tests of the models. Successful models will then be evaluated in terms of their ability to decompose behavioral data into its constituent cognitive components, especially in real-world situations where limited behavioral data can be collected. Once the best models have been identified, they will be developed into a statistical package for public distribution. Finally, tutorials and example data sets will be provided to make it easier for researchers to incorporate the model-based analysis into their own research. Overall the research will 1) identify which models are best for analyzing data from executive function tasks, 2) determine the experimental conditions under which the models can be used, and 3) provide packages and instructions for researchers to use the models in their own studies.