Measuring effortful motivation with heffort :|psychometric testing and validation using machine learning

Abstract

Lack of motivation can a ect everyone in our modern world, with consequences ranging from poor work and school performance to decreased family functioning. Thus, it is important to understand the mechanisms of motivation and devise personalized treatments. Traditionally, motivational and other psychological states in humans are measured with questionnaires. However, none of the current questionnaires can distin guish between motivation and pleasure, two components of the reward system. This is in contrast to how the brain works which employs the dopamine system to mediate e ortful motivation and the opiate system to mediate pleasure. Importantly, de cits in either system require a di erent treatment and thus it is pertinent that we can distin guish between these psychological states. To address this issue, our group has developed a novel questionnaire (HEFFORT) that we administered to a representative sample of Istanbulites varying in age, income, and lifestyle. We tested the hypothesis that HEF FORT can separate motivation from pleasure. First, we explored the relationships between the novel and validated questionnaires using Spearman correlations. Second, we developed an algorithm that predicts individual di erences in e ortful motivation using supervised machine learning classi cation techniques. Third, we determined the accuracy of the predictions by testing the algorithm on a separate testing dataset. Our results show that HEFFORT is reliable and able to distinguish between e ortful mo tivation and pleasure. Future testing and use can establish HEFFORT as a research and clinical diagnostic tool that assesses di erent components of reward.|Keywords : Motivation, Pleasure, Psychometrics, Quality Of Life, Machine Learning, Arti cial Intelligence, Decision Trees.