Caloric labels do not influence taste pleasantness and neural responses to erythritol and sucrose

The beneficial effects of substituting sugar with non-caloric sweeteners (NCSs) remain uncertain due to the mismatch between their rewarding sweet taste and lack of energy content. Functional magnetic resonance imaging (fMRI) studies indicate an influence of cognitive processes (e.g., beliefs, expectations) on reward system responses to NCSs, thereby changing their rewarding properties. We measured the impact of cognitive influences about the caloric content on brain responses and liking ratings to erythritol, a natural NCS with satiating properties, versus sugar (i.e., sucrose).

We performed a within-subject, single-blind, counterbalanced fMRI study in 30 healthy males (mean ± SD: age 23 ± 0.6 years, BMI 22.5 ± 0.3 kg/m²). Concentrations of erythritol were individually titrated to match the perceived sweetness intensity of a 16 % sucrose solution. During the scan, sucrose and equisweet erythritol solutions were delivered as 1 mL sips with either correct or purposefully incorrect "low-calorie" or "high-calorie" labels. After each sip, participants rated sweetness liking. Water with a "water" label was used as the control condition.

A 2 × 2 ANOVA revealed lower liking ratings for erythritol than sucrose (p < 0.0001), but no main effect of the label, nor label-by-sweetener interaction. General Linear Model (GLM) analysis of brain responses at FDR q < 0.05 showed no main effect of sweetener nor label, nor a label-by-sweetener interaction. However, several patterns of brain activity mediated the differences in subjective liking ratings between the sweeteners. Moreover, different neural responses were found for sucrose vs. water in parcel-wise, SVM, and ROI-based analyses, whereas for erythritol vs. water, only the latter two showed differences. Lastly, sucrose induced a stronger craving signature response compared to erythritol, driven by the pattern specific to drug craving.

Liking ratings were lower for erythritol than sucrose, and they were unaffected by the caloric label. There were no differences in neural responses between the sweeteners and labels, except in comparisons with water.