Flavor perception of wine is unchanged during commercial flight: a comparative field study

Introduction

Modern air travel is a multi-billion-dollar industry and more than four billion passengers flew on commercial airlines in 2017, which generated an industry-wide revenue of 755 billion USD¹. Labor, fuel, and equipment are the three biggest running expenses of the industry, but airline catering is also a costly venture, estimated to constitute an average of 2–3% of total operating costs². Despite this, airplane meals generally have a poor reputation among passengers. This may be due to the sheer quality of the food, but could also be a result of altered perception of flavor and odor in an airplane cabin^3–5.

Food perception is a complex multisensory experience depending on both taste, smell, and mouthfeel, which collectively form an impression of flavor⁶. Simulation studies have suggested that our sense of taste and smell is generally reduced in low-pressure environments such as airplane cabins, diminishing the perception of flavor intensity of food products^7–10. This is thought to be caused by elevated sensory thresholds for detection of tastants in low-pressure environments⁵. Moreover, this transitory impairment of the gustatory and olfactory senses may impair the general enjoyment of food as well⁸. In addition to the low air pressure, several other factors present in a standard commercial airline cabin are also assumed to contribute to this, such as dry recirculating air, cold temperature, and low oxygen levels^5,8,9. While the exact physiology behind these phenomena is not definitively determined, some explanatory mechanisms have been suggested: cold and dry air are impeding the interaction between odorants and olfactory receptors, and low oxygen levels are inducing hypoxia which may modulate receptor functioning^5,11. Other sensory modalities (e.g. hearing) can also impact flavor perception – a phenomenon known as crossmodal correspondence – and in this regard, the constant background noise in an airplane cabin may be another important sensory influence^10,12–15. Psychological stress is also assumed to modulate flavor perception to some degree^16,17, and commercial air travel is a source of – perhaps minor – psychological stress^18,19. Clearly, flavor perception during flight is a multifactorial process that has not yet been sufficiently characterized, which is ultimately to the disadvantage of the end consumer.

Wine has a complex composition of flavors and wine tasting requires a special attention to nuance. Therefore, the perception and appreciation of wine may also be affected by the suboptimal conditions in an airplane cabin. In current literature, the specific mechanisms of this process have not been described in its entirety and not all details are fully understood. In Figure 1 we propose a simplified theoretical framework illustrating how these factors may interact and how environmental and individual factors modulate the process of forming a wine consumption experience. Simulation studies have heavily suggested that wine consumption experience is negatively affected during flights^8–10, and reportedly, the bitterness of wine is more prominent when consumed in the air⁵. However, to the best of our knowledge, the impact of airplane cabin conditions on wine consumption experience has not yet been tested in a field study. In this study, we elected to focus specifically on the concept of flavor intensity and the subjective perception of this.

Figure 1. Theoretical framework proposal.

We hypothesized that the subjective perception of flavor intensity of wine is reduced while on board a commercial aircraft flying at standard cruising altitude compared with on the ground. Furthermore, we speculated that this would lead to a decreased overall liking and appreciation of wine, likely resulting in a worse consumer experience. To test these pre-specified hypotheses, we performed a comparative field study of high and low altitude wine tastings, and we aimed to verify the discrepancy in taste sensation, as suggested by simulation studies.

Methods

Results

In total, 22 participants were enrolled in the study. Of these, 45% were females and the majority were below 40 years of age (77%). 86% chose to drink red wine and the remaining participants drank white wine. The characteristics of the study participants are shown in Table 1.

The experienced aromatic intensities at both high- and low-altitude tastings had a median score of 5 on the 9-point hedonic scale (interquartile ranges of 3.5–6.0 and 4.0–6.5 respectively), and Wilcoxon signed-rank test did not demonstrate any statistically significant difference (p = 0.176). The differences between the remaining taste parameters at high and low altitude were also not significant (p-values 0.153 – 0.858). Nor was there any significant difference in the participants' POMS-A scores at high altitude compared with their scores at low altitude (p-values 0.058–0.705). Among these, there was a non-significant tendency towards a higher level of sleepiness at low altitude than at high altitude (p = 0.058). The results are listed in Table 2.

Discussion

Among our 22 enrolled participants, the flavor perception and overall liking were unchanged between high and low altitude. Additionally, there was no statistically significant change in participant mood evaluated through the adapted POMS-A questionnaire.

Our findings suggest that lay people do not experience a subjective difference in flavor intensity in a real-life aircraft cabin, when flying at standard cruising altitude compared with at ground level. Furthermore, our study was unable to detect a statistical difference in a variety of taste parameters, including overall liking of wine, fruit notes, and bitterness. This disagrees with multiple simulation studies conducted in controlled environments, simulating the conditions of an aircraft cabin and using test panels of both lay people and gastronomic professionals^8–10. Our results suggest that even though the gustatory and olfactory senses may be suppressed, it might not affect the subjective tasting experience of lay people (i.e. the final consumer experience) when tested in a real-life setting. Thus, the poor reputation of aircraft cuisine may not be a reflection of a changed perception of flavor and odor in high altitude environments. The mood of our participants was unaffected by the high and low altitude settings. The only parameter of the POMS-A questionnaire that came close to show a statistical difference was the sleepiness of our participants. This can be explained by the late hour of the low altitude tasting and therefore this result is negligible.

A strength of this study is that participants were neither wine professionals nor sommeliers, but lay people with an average knowledge and experience in wine tasting. Thus, the results of this study have high external validity in the general non-wine-proficient population. However, our study population was fairly limited as only 22 participants were eligible for inclusion, and all participants were Danish healthcare professionals making the study population quite homogenous in that regard. Future studies should include a larger and more heterogenous population in terms of nationality, profession, socio-economic status, etc.

This study was conducted in a largely authentic setting, which ensures high ecological validity and makes our conclusions generalizable. In this context, “authentic” refers to a setting that approximates normal circumstances for wine consumption during flight. However, such an uncontrolled study environment does allow for a possible influence of unrecognized and potentially confounding external factors. This is the nature of a field study and the influence of undesirable external factors is inherently unavoidable. It could be argued that the two testing environments in this study were fundamentally different, and in the future, these results could be expanded upon by performing two taste tests in the same airplane cabin, one at cruising altitude and another at ground level. This would reduce the influence of possible environmental confounders. Nonetheless, it could also be argued that the two testing environments in this study were comparable in several aspects. For instance, the presence of constant background noise is quite similar in an airplane cabin and a bus cabin, and the possible influence of auditory-gustatory crossmodal correspondence is therefore not expected to be a significant confounder^12,13. Since we were not allowed to bring measurement equipment onboard the airplane, the exact levels of background noise were not measured and some degree of auditory influence cannot be ruled out. Psychological stress is also thought to have an impact on flavor perception, but as all forms of public transportation can be considered a minor source of stress, the testing conditions were also in this regard quite similar. Psychological stress is therefore not considered a significant confounder in our study.

Some studies have suggested that the five basic flavors (i.e. sweet, sour, salty, bitter, and umami) may be affected differently by the conditions in an airplane cabin¹³. Consequently, future studies should include a variety of wines of different characteristics and quality in order to exclude any possible influence of the specific flavor composition of the wines chosen for this study. It may also be interesting to make a similar field study as the present, but including wine experts instead of lay people without special wine tasting abilities or training. It is also recommended that future studies should employ measures to control for confounders related to the individual, such as current satiety, appetite, and thirst. A limitation of the present study is the lack of experimental control. Under optimal circumstances, we would have preferred to implement a randomized crossover design, include a pre-flight taste test, and expand the timeframe in order to reduce the effect of recall bias. Unfortunately, this was not practically or economically feasible given the extensive security restrictions of commercial air travel, which is why we settled on the current experimental design. Another limitation of this study is the use of Falcon tubes as a vessel for the wine, as opposed to normal wine glasses. The experience of drinking a beverage is greatly influenced by the properties of its container³⁰, and these narrow plastic tubes could be an inhibiting factor in the consumption experience, e.g. inhibiting the odor sensation and diminishing the overall aesthetic experience and its emotional impact on the consumer^31,32.

In general, future research into improving the quality of airline catering could be of great benefit to the industry, as we believe that the current relationship between catering expenses and the quality of the final consumer experience may be disproportionate. The theoretical framework presented in Figure 1 was developed for generating hypotheses and is only intended as a preliminary proposal illustrating the mechanisms at work. It is intended as a basis for further discussion of the topic, and actual testing or validation of the framework is beyond the scope of this article. Further research is necessary on the physiology and psychology of flavor perception onboard airplanes, which will inspire new and innovative methods and techniques for designing meals and beverages optimized for consumption on airplanes. This is part of an emerging field of science known as gastrophysics, which will hopefully pave the way for future improvements in airline gastronomy⁵.

In conclusion, we were unable to verify the findings of previous simulation studies, as we did not find a difference in either subjective perception of flavor intensity or overall liking of wine between high- and low-altitude tests. This suggests that any possible perceptual change that exists while flying may not affect the subjective taste experience of lay people. Ultimately, current knowledge regarding flavor perception onboard airplanes is inadequate.

Data availability