Incidental exposure to hedonic and healthy food features affects food preferences one day later

Cognitive Research: Principles and Implications

Table 5 Mixed-effect regressions of food preference on predictors that included healthy eating habits (Habits) and BMI, along with food type (Food) and exposure (Expo)

DV: food preference	Model 1			Model 2					Model 3
Predictor	Estimate	SE	t	Estimate	SE	t	R²	AIC	∆R²	AIC
Health Exposure vs. No Exposure
Food Type	.06	.038	1.45
Expo sure	− .03	.018	− 1.38
Healthy eating habits	.01	.018	0.74
BMI	.02	.019	1.15
Food × Habits	− .18	.006	− 27.81	− .18	.018	− 9.94	29	51,259	− 10	53,000
Food × BMI	.04	.007	5.51	.04	.016	2.33	28	51,335	− 6	52,272
Exposure × Habits	− .04	.018	− 2.06	− .04	.018	− 2.03	23	52,163	− 1	52,246
Exposure × BMI	− .04	.019	− 2.01	− .04	.019	− 2.01	22	52,234	− 1	52,245
Food × Expo × BMI	− .01	.007	− 2.10	− .01	.015	− 0.98	28	51,341
Food × Habits × BMI	.04	.007	6.19	.04	.015	2.74	29	51,242	− 7	52,280
Expo × Habits × BMI	− .04	.019	− 2.21	− .04	.019	− 2.20	23	52,157	− 1	52,246
Food × Expo × Hab × BMI	− .01	.007	− 2.09	− .01	.014	− 0.96	29	51,249
Hedonic Exposure vs. No Exposure
Food Type	.11	.038	2.96	.11	.040	2.80	28	51,030	− 6	52,032
Exposure	− .02	.017	− 1.35
Healthy eating habits	.04	.017	2.46	.04	.022	1.91	23	51,885
BMI	.04	.018	2.38	.04	.020	2.24	22	52,010	0	52,030
Food × Exposure	.05	.006	7.28	.05	.015	3.14	28	51,025	− 7	52,077
Food × Habits	− .16	.006	− 25.64	− .16	.020	− 8.27	29	50,867	− 10	52,671
Food × BMI	.08	.007	11.07	.08	.017	4.57	29	51,007	− 7	52,146
Habits × BMI	.04	.019	2.27	.04	.020	2.23	23	51,859	− 1	52,029
Food × Expo × Habits	.02	.006	2.82	.02	.014	1.26	29	50,862
Food × Expo × BMI	.03	.007	3.76	.03	.015	1.69	29	51,011
Food × Habits × BMI	.05	.007	6.52	.05	.016	2.92	30	50,852	− 8	52,067

Regressions were performed on standardized measures. For Food Type, tasty foods were coded + 1, and healthy foods were coded − 1. In the Hedonic vs. No Exposure regression, hedonic exposure was coded + 1, and no exposure was coded − 1. In the Health vs. No Exposure regression, health exposure was coded + 1, and no exposure was coded − 1