Looks like the value function in reinforcement learning. You assign a high value to being close to your loved one. The value function is like you say a synthesi

Looks like the value function in reinforcement learning. You assign a high value to being close to your loved one. The value function is like you say a synthesis of your past experiences, learning what leads to positive or negative rewards. The rewards themselves are biologically coded in our brains by evolutionary pressures, but the value function is like integrating rewards over chains of situations and actions.