Avshalom Caspi and Terrie Moffitt are professors of psychology who together study mental health and human development.
The overarching aim of our research is to improve knowledge about how to prevent problems of adult mental and physical health, by uncovering when such problems first emerge in life, and why. Our research uses the longitudinal birth cohort method. This method starts with a list of all babies born in a defined place (typically a city or a country), within a defined period (typically a year). This list is referred to as a “cohort.” The babies are enrolled in the study (sometimes before they are born) and then followed and assessed repeatedly throughout their lives as they grow up and grow old. Researchers and participants collaborate on the project together for many decades, which makes this method quite different from the sorts of research projects that most other psychologists do. The research participants generously give of themselves over and over all their lives, vouchsafing the scientists the details of their most intimate, delicate, and even dangerous behaviors and experiences, as well as their body tissues. The scientists, in turn, accept the ethical obligations to ask important questions, publish findings rapidly, preserve the data carefully for use in future years, and guard the participants’ confidentiality at all costs.
“Scientists” is pluralized in the last sentence because a longitudinal cohort study is too ambitious an enterprise to be undertaken by one or even two psychologists on their own. A longitudinal cohort study is a scientific instrument, not unlike a large Hadron particle collider, or a kilometers-wide radio-telescope antenna. Like one of these instruments, a cohort study is expensive but pays for itself because it is a resource for many scientists to make multiple discoveries. Such scientific instruments are science-magnets; they attract collaboration from many different fields and nations. Our work is thus about as inter-disciplinary as it gets, and we routinely collaborate with economists, geneticists, epidemiologists, sociologists, demographers, neuroscientists, immunologists, and medical researchers specializing in everything from the heart to the teeth. In addition to many of collaborators, the engine of our team is made up of Ph.D. students and postdoctoral fellows who bring new ideas to our work, and with whom we work closely on a wide range of topics.
The longitudinal cohort method is sometimes dismissed as “merely observational,” so it is worth noting some of its advantages. Many of the most powerful causes of health conditions cannot ethically be administered to humans in randomized experiences because they are stressful or toxic (for example, we study child maltreatment and lead exposure). Other powerful causes are impractical to study experimentally because causation unfolds over a very long period (for example, we study individuals who use cannabis regularly over two decades). The longitudinal cohort method is a complement to the experimental method. Experiments with animal models and human analogue measures show that manipulation of variable X can cause outcome Y, but longitudinal cohort studies are essential to test whether variable X does predict outcome Y in the human general population, and to what extent (effects of many proven causes are too rare to account for the public’s health). These advantages of the cohort method are magnified when studying mental health, because animal models are limited when the outcomes are uniquely human.
Welcome to our site, and we hope you will find it useful.