Table 1.

BIC scores and best-fit parameters of the models fitted to the choices of demand-avoiding participants

Models	BIC/parameters	Experiment 1 (n = 24)			Experiment 2 (n = 17)
Models	BIC/parameters	Median	25th percentile	75th percentile	Median	25th percentile	75th percentile
A, PE models
ST	BIC	95.1	84.0	114.8	103.1	96.1	108.8
EC_unchosen(1): free	EC_unchosen(1)	7.51	4.69	10.61	11.05	6.44	13.16
	α	0.65	0.53	0.85	0.41	0.32	0.60
	β	0.29	0.17	0.61	0.28	0.18	0.42
ST	BIC	91.3	81.8	109.7	99.7	95.0	106.3
EC_unchosen(1) = EC_chosen(1)	α	0.66	0.57	0.87	0.53	0.38	0.65
	β	0.25	0.16	0.53	0.25	0.17	0.41
HL	BIC	83.3	64.2	106.5	80.7	67.9	97.0
EC_unchosen(1): free	EC_unchosen(1)	0.75	0.67	0.87	0.75	0.71	0.96
	α	0.84	0.72	0.96	0.85	0.63	0.92
	β	3.13	2.23	4.49	3.32	2.47	4.48
HL	BIC	78.5	59.2	101.4	81.9	63.3	93.4
EC_unchosen(1) = EC_chosen(1)	α	0.80	0.71	0.96	0.87	0.64	0.97
	β	3.13	2.23	4.46	3.15	2.44	4.32
IC	BIC	135.2	128.6	138.0	130.9	126.8	134.9
EC_unchosen(1): free	EC_unchosen(1)	0.70	0.49	1.11	0.90	0.41	1.29
	α	0.23	0.15	0.38	0.20	0.10	0.40
	β	2.18	−4.15	3.14	2.43	−0.44	3.86
IC	BIC	130.7	123.9	134.0	126.2	124.5	130.8
EC_unchosen(1) = EC_chosen(1)	α	0.29	0.18	0.46	0.20	0.12	0.32
	β	2.90	0.72	3.39	3.60	2.24	4.31
ST-IC	BIC	100.3	85.6	119.7	108.0	99.2	114.0
EC_unchosen(1): free	EC_unchosen(1)	7.59	4.70	10.70	11.30	6.44	14.44
	α	0.65	0.53	0.84	0.40	0.32	0.60
	β	0.29	0.17	0.60	0.28	0.18	0.40
	w_incorrect	0.67	0.00	6.96	0.00	0.00	0.00
ST-IC	BIC	96.4	83.4	114.6	104.1	98.3	111.5
EC_unchosen(1) = EC_chosen(1)	α	0.67	0.57	0.86	0.53	0.38	0.65
	β	0.26	0.16	0.53	0.26	0.17	0.36
	w_incorrect	0.00	0.00	3.14	0.00	0.00	1.28
HL-IC	BIC	88.5	69.2	111.6	85.9	73.1	102.1
EC_unchosen(1): free	EC_unchosen(1)	0.75	0.67	0.89	0.76	0.70	0.96
	α	0.83	0.72	0.93	0.85	0.64	0.92
	β	3.38	2.18	4.40	3.16	2.36	4.58
	w_incorrect	0.00	0.00	0.14	0.00	0.00	0.08
HL-IC	BIC	83.7	64.2	106.5	84.3	68.5	98.6
EC_unchosen(1) = EC_chosen(1)	α	0.80	0.71	0.96	0.87	0.64	0.97
	β	3.13	2.23	4.36	3.15	2.44	4.32
	w_incorrect	0.00	0.00	0.01	0.00	0.00	0.00
B, PE-Rest models
Rest	BIC	132.3	129.0	134.0	131.0	125.5	132.1
	α	0.40	0.06	0.97	0.12	0.06	0.45
	β	0.01	−0.05	0.11	0.05	−0.05	0.12
HL-Rest	BIC	88.7	69.2	103.8	92.8	73.6	109.0
	α_cost	0.80	0.68	0.96	0.87	0.63	0.97
	α_reward	0.00	0.00	0.03	0.00	0.00	0.01
	β	−3.21	−4.54	−2.45	−3.14	−4.32	−2.24
	w_reward	−0.03	−0.43	0.60	0.76	−0.35	2.02
C, pWSLS/PS models
pWSLS-HL	BIC	92.9	79.0	110.8	97.1	91.1	109.4
	p	0.19	0.14	0.27	0.21	0.18	0.27
pWSLS-IC	BIC	111.1	88.6	124.7	100.9	92.7	110.8
	p	0.76	0.67	0.85	0.80	0.76	0.83
PS-HL	BIC	69.4	57.8	87.4	70.1	54.6	90.2
	a	0.95	0.91	0.98	0.97	0.91	0.98
	b	0.47	0.30	0.58	0.48	0.46	0.58
PS-IC	BIC	111.1	88.6	124.7	100.9	92.7	110.8
	a	0.76	0.67	0.85	0.80	0.76	0.83
	b	0.50	0.41	0.55	0.66	0.50	0.77

This table shows the median, 25th, and 75th percentiles of BIC scores and best-fit parameter estimates for 10 variants of PE models (5 assumptions on ActualCost [Solve-Time (SV), High-Low (HL), Incorrect-Correct (IC), ST-IC, and HL-IC] × 2 assumptions on EC_unchosen(1) [free parameter or equal to EC_chosen(1)]), two types of additionally considered PE models (Rest and HL-Rest models), and pWSLS and PS models. EC_unchosen, the expected cost for the unchosen option; EC_chosen, the expected cost for the chosen option; α, learning rate; β, inverse temperature; w_incorrect, weighting parameter; α_cost, the learning rate for cost; α_reward, the learning rate for reward; w_reward, weighting parameter; p, probability to select the different and same option after Win and Lose, respectively; a, probability to select the same option after Win; b, probability to select the same option after Lose.