My Research Philosophy. As a graduate student, my advisor impressed upon me the importance of being fair to science. His view and one that I’ve carried with me throughout my career was that the process of doing science is more important than the results one obtains. This view, that “process trumps output” is ever more relevant in today’s research climate. As a believer in the value of transparency and integrity, it is my view that science should be subject to the checks and balances made possible through open and transparent processes. Though one’s theories and ideas are central to science, personal agendas have no place. Mistakes and non-replicable effects are things to acknowledge and learn from, not defend or dismiss. In my own work, I have taken several steps to enhance the scientific integrity of my lab. First, I have transitioned to using the open science framework for documenting and sharing data (https://osf.io/gu6wx/) and require my students to use this platform as well. However, open science only works when fact checking actually takes place. Given the flood of data entering the literature, it is impossible that every finding be verified independently. This leads to the second step: I complete replication studies whenever possible and enter those studies into the scientific record, even when they conflict with my own published finding. Many studies, and in particular replications, are preregistered. Though I'm a proponent of open science and pre-registration, I do believe that there is a need for exploratory research, so long as it is identified as such. Third, I evaluate data not with white gloves, but with a hammer and chisel. Solid results should not depend on a few data points or subset of transformations. They should be robust to violations of parametric assumptions and the presence of unusual data. In my own lab, we use a variety of techniques in the analysis of data. Though Bayesian methods are preferred, we also ensure that our results are consistent across alternative analytic methods, including one developed within my own lab. We also use a duplicative analysis model whenever possible, wherein data are analyzed independently by co-authors to ensure that errors in analysis are caught before publication.

My Research Overview. At the core of my research program is the view that basic research ought to be guided by real-world problems. Although basic research alone can be of substantial theoretical importance, truly transformative research requires that the basic research eventually be scaled-up to address practical problems. Thus, as a cognitive scientist interested in complex cognition, I have tried to identify research questions that are of both practical and theoretical interest, such as:

• How do people generate hypotheses and make judgments?
• What processes are modified through working memory (WM) training and do they generalize?
• Can metacognitive judgments be used to modify people’s learning behavior?
• How can the process of scientific discovery be improved?