Table 2.

Model recovery rates of two psychophysics models

Selection method	Model fitted	Data from Stevens	Data from Fechner
AIC, BIC	Stevens	100%	63%
	Fechner	0%	37%
MDL	Stevens	99%	2%
	Fechner	1%	98%

The table shows the percentage of samples in which the particular model was selected. One thousand samples were generated from each model using the same four points for x, which ranged from 1 to 4 in increments of 1. The random error was normally distributed with a mean of zero and a standard deviation of 1. The parameters values used to generate the simulated data were a = 2, b = 2 for Stevens' model and a = 2, b = 5 for Fechner's model. In computing the geometric complexity as well as in estimating best-fit values of the parameters, the following parameter ranges were assumed: 0 < a < ∞, 0 < b < 3 for Stevens' model, and 0 < a, b < ∞ for Fechner's model.