Genetics, imaging, and cognition: big data approaches to addiction research

Abstract The etiology and trajectory of addictions is complex, caused and moderated by individual differences in cognition that are themselves a function of genetics and of environment. In this chapter, we discuss how “Big Data” can shed light on the cognitive correlates of addiction. Big Data is primarily data-driven, using algorithms that search for patterns in data, with accurate prediction on previously unseen data as the metric of success. In this chapter, we introduce and provide practical advice on Big Data approaches for addiction. In the first part of this chapter, we describe how online methods of data collection facilitate the collection of large datasets. In the second section, we outline some recent advances in neuroimaging, with a focus on prediction of substance use using machine learning methods. In the final section, we present advances in genetics—meta- and megaanalyses—which may provide breakthroughs in our understanding of the genetics of addiction.