Not a current user.

Brown-Iannuzzi Group research activities

The current research seeks to investigate whether a particular method used within psychology, the Reverse Correlation Task (Mangini & Beiderman, 2004), is at risk for an inflated Type I error rate. To investigate this question, we will use simulated data to mimic the Reverse Correlation Task. These simulations, however, utilize a lot of computing power (relate to the desk top computers we are currently working with), and thus the use of the High Performance Computing system at UK would be quite useful.

Simulate Type I error rates under various experimental conditions

We will use the High Performance Computing system to conduct 4 simulations. First, we wish to simulate the Type I error rate under normal experimental conditions. Second, we wish to simulate the Type I error rate when each “participant” (or simulated respondent) is treated independently. Third, we wish to simulate the Type I error rate when averaged across “participants” (or simulated respondents) using several different simulated outcome measures. Finally, we wish to simulate the Type I error rate when we group some “participants” (or simulated respondents) and level other “participants” (or simulated respondents) as independent. I will be working on this project with Dr. Jeremy Cone, a faculty member at Williams College (his website is: https://psychology.williams.edu/profile/jdc2/) and Dr. Ryan Lei, a post-doc at New York University (his website is: https://ryanflei.weebly.com/).

Software:

All simulations will be conducted using R.

Collaborators:

Dr. Jeremy Cone, faculty at Williams College (his website is: https://psychology.williams.edu/profile/jdc2/)
Dr. Ryan Lei, post-doc at New York University (his website is: https://ryanflei.weebly.com/).

Browser not supported