def test_population_fitting(task, agent, n_sessions=8, n_trials=1000, pop_params=None):
"""Simulate a set of sessions using parameters drawn from normal distributions
specified by pop_params. Then fit the agent model to the simulated data and plot
correspondence between true and fitted paramter values.
"""
sessions = simulate_sessions(task, agent, n_sessions, n_trials, pop_params)
ML_fits, MAP_fits, pop_params = mf.fit_population(sessions, agent, max_iter=15)
rp.plot_true_fitted_params(sessions, ML_fits, MAP_fits)
return (sessions, ML_fits, MAP_fits, pop_params)
