Effects of period length and decision noise on the relative performance of the confirmation model. (a) Effect of period length on reward. The line plot represents the difference in average reward between the confirmation model (with the best confirmatory learning rate combination per period) and the unbiased model (with the best per period single learning rate) in function of the log of the period length and for the four different volatility conditions. The logarithmic transformation of the trial number is for illustrative purpose only. , two-tailed independent rank-sum tests. (b) Effect of decision noise on performance. The line plot represents the difference in per trial average performances of the confirmation model (with the best confirmatory learning rates combination) and the unbiased model (with the single best learning rate) as function of the log of softmax temperature, and for the four different volatility conditions. The logarithmic transformation of the softmax temperature is for illustrative purposes only. , two-tailed independent rank-sum tests.
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.