I noticed a small bug in cpm.applications.decision_making.PTSM2025. The variable model_choice should represent the model's predicted choice (0 = safe, 1 = risky). However, the code is as follows:
model_choice = np.random.choice([0,1], p=[policies, 1-policies])
Since policies is the computed probability of choosing the risky option (which I verified is indeed computed correctly), model_choice should be computed as follows:
model_choice = np.random.choice([0,1], p=[1-policies, policies])
I noticed a small bug in cpm.applications.decision_making.PTSM2025. The variable
model_choiceshould represent the model's predicted choice (0 = safe, 1 = risky). However, the code is as follows:model_choice = np.random.choice([0,1], p=[policies, 1-policies])Since
policiesis the computed probability of choosing the risky option (which I verified is indeed computed correctly), model_choice should be computed as follows:model_choice = np.random.choice([0,1], p=[1-policies, policies])