Mathematical modeling of emotions (Q1842464)

Learning of a skill by repetition is improved by stress in a first stage and deteriorates in the long run. First-order scalar differential equations model the learning curve and the averaged stress perturbation. Parameters are chosen so that the added solutions to these differential equations match experimental curves. Thus, under stress, the learning curves are characterized by appropriate time-constants. Modelling the combined effect of two stresses is done by adding the two solutions to differential equations modelling these perturbations but there is not a satisfactory match with experimental data. It is unclear how the physical interpretation of the state variable (``information in human memory'') relates to measured experimental data.