Appendix 2—table 1.

The mean BIC and model parameter values for each model for each experiment. The bolded lines show the models with the smallest BIC, which was model number 2, the dual-learning rate model without separate parameters for day 1 conditions.

DOI: http://dx.doi.org/10.7554/eLife.26801.025

Experiment	Model	BIC	on	off	α+ON	α+OFF	α-ON	α-OFF	β
1	1	996.243	0.1006						60.8972
	2	885.9434			0.1024		0.0041		9.678
	3	992.3985	0.0808	0.0988					36.866
	4	891.5522			0.1184	0.1593	0.0184	0.0097	7.3885
2	1	520.6029	0.1131						231.6615
	2	485.0105			0.1922		0.0319		21.278
	3	510.5098	0.1248	0.0616					116.4691
	4	487.6825			0.3144	0.1825	0.0438	0.1273	7.2812
3	1	807.0322	0.132						101.1408
	2	737.4145			0.2069		0.0253		5.9512
	3	802.9193	0.1059	0.0624					163.4829
	4	746.7926			0.2918	0.1846	0.0521	0.0555	4.104