NTT has developed a mathematical model that can analyze the long-term goal-achieving behavior of people who tend to overestimate short-term goals, and also determine the optimal intervention to help such people achieve their goals. By applying appropriate interventions derived using this model, it is possible to help individuals achieve their goals related to health, education, etc.
In this research, we focused on ”progress accumulation tasks” among goal achievement tasks. A progress accumulation type task is a task that aims to achieve a ”goal progress” by accumulating a numerical value called ”progress” within a given time limit, and is rewarded when achieved.
In addition, by utilizing techniques used in mathematical optimization technology, we have constructed algorithms that quickly find optimal interventions in various problem settings (optimal goal setting problems, optimal incentive design problems, etc.).
In this research, we developed a new mathematical model by analyzing human behavior under present bias. This model allows us to mathematically write down human behavior under current biases in closed form. Therefore, an advantage of this method is that it can predict future human behavior without conducting simulations.
Present bias is the tendency to overestimate recent gains relative to future gains.
Examples of actual use include supporting people with health care apps to help them achieve walking goals for their health, or helping them increase their savings with a household account book app.