Noisy heteroclinic networks and
sequential decision making
I will talk about sequential decision making models based on
along heteroclinic networks of dynamical systems, i.e.,
saddle-type equilibrium points connected by heteroclinic orbits.
goal is to give a precise description of the asymptotic behavior
the limit of vanishing noise. In particular, I will interpret
times for stochastic dynamics as decision making times and give
result on their asymptotic behavior. I will report on extensive
on decision making in no a priori bias setting obtained in a
psychology experiment (joint with Joshua Correll, University
Chicago), and compare the data with the theoretical results. I
also show that the same behavior of exit times appears
innonequilibrium models of statistical mechanics.