Opportunities and challenges in the collection and analysis of digital phenotyping data

The broad adoption and use of smartphones has led to fundamentally new opportunities for capturing social, behavioral, and cognitive phenotypes in free-living settings, outside of research laboratories and clinics. Predicated on the use of existing personal devices rather than the introduction of additional instrumentation, smartphone-based digital phenotyping presents us with several opportunities and challenges in data collection and data analysis. These two aspects are strongly coupled, because decisions about what data to collect and how to collect it constrain what statistical analyses can be carried out, now and years later, and therefore ultimately determine what scientific, clinical, and public health questions may be asked and answered. Digital phenotyping combines the excitement of fast-paced technologies, smartphones, cloud computing and machine learning, with deep mathematical and statistical questions, and it does this in the service of a better understanding our own behavior in ways that are objective, scalable, and reproducible. We will discuss some fundamental aspects of collection and analysis of digital phenotyping data, which takes us on a brief tour of several important scientific and technological concepts, from the open-source paradigm to computational complexity, with some unexpected insights provided by fields as varied as zoology and quantum mechanics.

Automated summary

The widespread use of smartphones has opened up new opportunities for capturing social, behavioral, and cognitive phenotypes outside of traditional research settings. Digital phenotyping, based on existing personal devices, poses challenges and opportunities in data collection and analysis that impact the types of questions that can be answered in scientific, clinical, and public health contexts. This field combines cutting-edge technologies with mathematical and statistical questions in order to improve our understanding of human behavior in an objective, scalable, and reproducible way.