Deep Longevity published a paper in Aging-US outlining machine learning methods for human psychology in collaboration with Dr. Nancy Etcoff of Harvard Medical School. researchers have created two numerical models of human psychology based on data from the American Midlife Study.
The first model is an ensemble of deep neural networks that uses information from psychological surveys to predict a respondent’s age and mental health 10 years later. This model describes the trajectory of human psychology as we age. It also shows that the ability to form meaningful connections and the capacity for psychological autonomy and environmental mastery develop with age. It also shows that the emphasis on personal progress continues to decline, but the feeling of having a purpose in life disappears after 40-50 years. These results add to the growing knowledge about socioemotional selectivity and hedonic adaptability in the context of adult personality development.
The second model is a self-organizing map that was created as the basis for a recommendation engine for mental health applications. This unsupervised learning algorithm divided all respondents into groups based on their likelihood of suffering from depression, in addition to identifying the shortest path for any individual to move toward a mentally stable group. Alex Zhavoronkov, Chief Longevity Officer at Deep Longevity, elaborates: “Existing mental health apps provide generic advice that works for everyone, but doesn’t work for anyone. We’ve built a system that’s scientifically based and delivers exceptional personalized service. ."
To demonstrate the potential of this system, Deep Longevity has released a web service, FuturSelf, a free online application that allows users to take the psychological test described in the original publication. At the end of the assessment, users receive a report with recommendations aimed at improving their long-term mental health and can join a coaching program that provides them with stable AI options. The data obtained on FuturSelf will be used to further develop Deep Longevity’s digital approach to mental health.
Professor Vadim Gladyshev, an expert in biogerontology from Harvard Medical School, commented on the potential of FuturSelf - "This study provides an interesting perspective on mental age, future well-being and depression risk and demonstrates A novel application of machine learning methods to mental health issues also broadens how we think about aging and transitions through life stages and emotional states.
It is reported that the researchers plan to continue to study humans in the context of aging and long-term well-being. psychology. Now, they are conducting a follow-up study looking at the impact of happiness on physiological measures of aging.
