Arousal influences olfactory abilities in adults with different degree of food neophobia

Abstract

Food neophobia, i.e., the aversion to novel foods, and olfaction are both factors strongly affecting food choices. Mounting evidence suggests a higher arousal towards food as a key factor underlying the reluctance to eat what is unfamiliar to us. As the role of olfaction behind this phenomenon is poorly understood, we explored the associations between food neophobia and trait anxiety, olfactory functions (odor threshold, discrimination and identification) and retronasal aroma release from a reference food in a healthy cohort of 83 adult volunteers. We grouped participants in Low-Neophobics or neophilics (n = 35), Medium-Neophobics (n = 32) and High-Neophobics (n = 16) according to the widely recognized Food Neophobia Scale. Participants with higher neophobic tendencies were found to have marginally higher trait anxiety levels than neophilics (p = 0.10). A lower global olfactory functioning and odor discrimination abilities characterized High-Neophobics, while Medium-Neophobics showed a higher odor sensitiveness than Low-Neophobics. Lastly, High-Neophobics showed a lower extent of retronasal aroma release, likely due to a shorter duration of oral processing and higher anxiety-related physiological responses (such as breathing rate). In summary, this study supports the assumption that the conflicting relationship that neophobics have with food may be led by higher levels of arousal toward foods, rather than different chemosensory functions.

Figure 1

Raincloud plot showing the differences on both global (TDI) and relative subtests (OT: Odor Threshold; OI: Odor Identification, OD: Odor Discrimination) olfactory assessments as a function of FN level (LN: Low-Neophobics; MN: Medium-Neophobics; HN: High-Neophobics). The plots provide a representation of data distribution (the ‘cloud’), individual raw observations (the ‘rain’), the median (black filled circle) ± IQR (perpendicular) within each FN level for both global and subtests olfactory performance. Only statistically significant pairwise differences observed after post hoc Dunn’s test with Bonferroni correction are presented (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

Figure 2

First two factors of the individual factor map (a) from the MFA based on matrices of SIFT-MS parameters. Large squares (light green square, light orange square, brown square) are MFA centroids while the circles (light green circle, light orange circle, brown circle) represent participants’ position in the bi-dimensional space colored according to their FN level (LN, MN, HN). Each MFA centroid is associated (b) with partial MFA groups points for the six SIFT-MS parameters (small circles with same colour as the corresponding MFA centroid). (c) Variable factor map from the MFA model. Dots represent correlations between SIFT-MS parameters (for the 7 monitored compounds) and the first two significant dimensions of the MFA model.

Figure 3

Log-transformed median (transparent) and smoothed (bold) release curves for the 7 monitored VOCs by SIFT-MS as a function of FN level.

Figure 4

Raincloud plot (a) showing the differences in breathing rates between FN levels (LN: Low-Neophobics; MN: Medium-Neophobics; HN: High-Neophobics). The plot provides a representation of data distribution (the ‘cloud’), individual raw observations (the ‘rain’), the median (black filled circle) ± IQR (perpendicular) within each FN level. Only statistically significant pairwise differences observed after post hoc Dunn’s test with Bonferroni correction are presented (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001). b–d: examples of breathing behaviors from a replicate of three participants with similar biological characteristics (b: Female; 51 yo; BMI = 19.27; c: Female; 53 yo; BMI = 25.46; d: Female; 53 yo; BMI = 21.82) but different according to their FN level. The top trace shows the curve of acetone (C₃H₆O) within the task with relative local minima (blue filled circle). The bottom trace is the sum of the 7 monitored VOCs, which represents the volatilome of the candy. Lastly, the transparent rectangles report the duration of the oral processing (s) for each replicate considered.