this model includes XX free parameters (α1, α2, β1, λ, pers, go_bias)
here we simulate N agents participating in a 2 stage task that includes 2 states at the 2nd stage 
each stae contains 2 choices 
each if the first stage choices lead constantly (100% of the trails) to a specific state at the 2nd stage

for selecting a choice the agent has to execuate a "Go" or a "No go" 
the type of action is set randomly to the choices at each trail