|
SQAB Tutorial: What's the Best Model for These Data? Information Theoretical Approaches to Inference as an Alternative to Hypothesis Testing |
Sunday, May 28, 2017 |
9:00 AM–9:50 AM |
Hyatt Regency, Centennial Ballroom D |
Area: EAB; Domain: Basic Research |
PSY/BACB CE Offered. CE Instructor: M. Christopher Newland, Ph.D. |
Chair: Peter R. Killeen (Arizona State University) |
Presenting Authors: : M. CHRISTOPHER NEWLAND (Auburn University), DEREK POPE (Virginia Tech Carilion Research Institute) |
Abstract: Null Hypothesis Statistical Testing (NHST) was developed to provide an objective way to quantify inference. The result is a ritualized technique that is frequently necessary for publication despite criticisms that it is minimally informative, misleading, and produces unreproducible results. NHST tests the probability of the data given a null hypothesis that is rarely of interest and is often implausible. The result is a torturous statement of whether the data are likely to have occurred. An alternative approach, called Information Theoretic (IT) based inference, does not carry many of these problems because it returns a different probability. IT approaches ask the question of interest in model building: Of a set of models, which ones are best? And by how much? By building upon Akaike Information Criteria, IT inference returns the probability of the models considered given the data, numbers that are readily interpretable. Unlike NHST, the approach actually encourages the testing of many models in order to increase the chances of including good ones. Corrections for multiple comparisons are neither necessary nor appropriate. The tutorial will identify criticisms of NHST, offer a (relatively) nontechnical background for IT approaches, and provide examples of IT-based inference using spreadsheets. |
Instruction Level: Intermediate |
Target Audience: Certified behavior analysts, graduate students, licensed psychologists. |
Learning Objectives: At the conclusion of the event, the participant will be able to: (1) describe how null hypothesis statistical testing has resulted in current concerns about replicability in the social and biomedical sciences; (2) describe a new, Information Theoretic approach to statistical inference that is well-suited to model development in behavior analysis; (3) describe how this new approach can be implemented in a spreadsheet. |
|
M. CHRISTOPHER NEWLAND (Auburn University), DEREK POPE (Virginia Tech Carilion Research Institute) |
Chris Newland earned his Ph.D. from Georgia Tech, did postdoctoral work in Environmental Health at the University of Rochester, and is now a Professor of Psychology at Auburn University. His research, which has been funded mostly from the National Institute of Environmental Health Sciences, applies behavioral principles to explore the impact of drugs and environmental contaminants that act on the brain. A life-span development approach is threaded through his research, so he has examined early development, aging, and, more recently, adolescence. A key element of his work is the application of quantitative models taken from behavior analysis to characterize mechanisms by which chemicals disrupt behavior. With his students, he has become interested recently in model-based inference, hence this tutorial. Dr. Newland has served on the editorial boards of JEAB, The Behavior Analyst, Neurobehavioral Toxicology and Teratology and is an Associate Editor of Neurotoxicology. He has served on numerous panels reviewing environmental policy and served as a regular member of the Neurotoxicology and Alcohol (NAL) Study Section for the NIH. He is currently examining the impact of exposure to drugs and contaminants during early development and adolescence and is seeking to link behavioral and epigenetic consequences of early neural damage. |
Derek Pope grew up in Washington DC. He attended James Madison University where he graduated magna cum laude and earned his BA in behavior analysis. He then traveled to Auburn University and joined Chris Newland's behavioral pharmacology and toxicology lab, where he earned his Ph.D. in 2016. While at Auburn, he investigated the interactions between genotype, contextual stimuli, and d-amphetamine on delay discounting in mice, the effects of chronic cocaine exposure during adolescence on spatial discrimination reversal, delay discounting, and demand and response output under FR schedules, the effects of chronic methylmercury exposure on interval timing, the acquisition of response chains, and high-rate responding, and, finally, how the application of theoretical and quantitative models may help to understand the effects of various manipulations within and across these studies. He is now at Virginia Tech Carilion Research Institute's Addiction Research Recovery Center under the tutelage of Dr. Warren Bickel where he continues to research executive functioning, economic demand, and delay discounting, and continues to explore and exploit theoretical and quantitative models. |
Keyword(s): Akaike criterion, model selection, model-based inference, statistical inference |
|
|