How Synonymic Taste Words Alter Perceived Taste in American Consumers

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for giving me the opportunity to review this study. I found it interesting, and enjoyed reading it. The reporting structure and clarity is generally very good and the authors are to be commended for making their data available through OSF. I am not a corpus linguist, nor food scientist or advertising expert, so my comments should be seen as coming from an interested member of the applied linguistics research community with general, not specific, expertise in this area. Generally speaking, I think the study is of sufficient quality to publish, following minor revisions per my comments below.

In the method, the description of the taste task left me with a couple of questions that I hope the authors can clarify. It is not clear to me when the participants were exposed to the ‘taste synonym’. I am assuming that they will have seen the word first, before tasting, as this would make logical sense if you were expecting the word to influence their perception of the taste. However, the information about the second question they were asked to answer (“Sour…[etc] accurately describes this product”) suggests that they might have seen the word afterwards. Would this not compromise that logic? So, some clarification here would be useful. I would also like to know what was on the labels on the samples. Was there a sample of the product with a label next to it that just had the word (e.g. “Sour”) on it? Or did the label say something like “Sour Apple”. Maybe an image of how the food was presented (i.e. with the label) and some clarification in the text would help resolve this.

The layout of Table 8 is a bit confusing to me, with the sampled product at the end of each day. This, I guess, is personal preference, but I would feel it more logical to have the name of the product in the first collum, then the more associated and less associated columns. And to have Day 1 above, rather than next to Day 2. It looks as though different products were sampled on different days. Why was this? i.e. Why didn’t all participants sample all products?

Column headings on Table 9 are confusing. Column one says ‘average rating’ but all items in that column are products, not ratings. I think the first column should be ‘Product’, second should be ‘Average rating with more associated label’, and third should be ‘Average rating with less associated label’.

The authors have given some plausible responses to why the results of experiment 2 were as they were. But it all feels like a bit of a stretch (the cognitive load argument seems to really be clutching at straws). Given the small sample size, rating a small number of products, might it just as easily be explained by random noise? i.e. labels don’t make any difference to taste perception (even with the marginally significant results for some pairings?). Did the authors conduct a power analyses (either ante hoc or post hoc, or both)? A type one error doesn’t seem to be out of the question.

Author Response

Please see Table R1 in the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Summary

This paper presents an innovative investigation into the semantic differences between synonymic taste words and their effects on taste perception. Through two complementary experiments—category fluency tasks (CFTs) and taste tests—the authors establish that (1) synonymic taste words exhibit subtle but significant semantic differences in terms of their food associations, and (2) these differences influence consumers' taste perceptions, with most foods being rated more favorably when labeled with more strongly associated taste descriptors. The study extends previous research by Johnson and Pfenninger (2021) on "crispy" versus "crunchy" to include five additional pairs of taste descriptors, demonstrating both the validity of category fluency tasks as alternatives to corpus analysis and the generalizability (with important exceptions) of association effects on taste perception.

Strengths

The paper's most significant contribution is its methodological innovation in using CFTs to identify food-taste word associations effectively. This approach provides a faster alternative to corpus-based methods while still capturing core semantic differences between near-synonyms. The thoroughness of the experimental design is commendable, with careful attention to participant selection, stimulus preparation, and statistical analysis. The authors skillfully reinterpret Johnson and Pfenninger's (2021) findings through the lens of their own data, providing a more nuanced understanding of how association affects taste perception.

The broad definitional approach to "taste words" could be improved through references to neurophysiological research and allows for a comprehensive examination of the linguistic landscape of gustation. The paper's theoretical grounding in both corpus linguistics and experimental psychology creates a solid foundation for the cross-disciplinary work.

Weaknesses

While the laboratory setting was necessary for experimental control, as the authors acknowledge, this limits the ecological validity of the findings. The inability to include "crispy" and "crunchy" in Experiment 2 due to concerns about sensory fatigue means that direct comparison with Johnson and Pfenninger's (2021) work is somewhat limited. The authors also note challenges in explaining why certain foods (candy, milk, watermelon) defied the general preference for more associated taste word labeling.

The paper would benefit from deeper exploration of cross-linguistic dimensions, particularly given that the title suggests a "cross-linguistic approach." The authors mention future research directions in other languages, but the current study focuses exclusively on American English.

Connection to Literature and Extension to Metonymy

The paper lacks significant engagement with  research on taste descriptors, which represents a missed opportunity considering cross linguistic contributions to food discourse. The work of Sanz Valdivieso and López-Arroyo (2022) on wine terminology and metonym demonstrates that taste descriptors in English and Spanish often function metonymically in ways distinct from English counterparts. For example, Spanish wine descriptors like "aterciopelado" (velvety) connect texture and taste through metonymic extension in ways that reveal cultural conceptualizations of taste not captured in the authors' Anglo-centric approach.

López Arroyo and Roberts (2017) have conducted extensive research on sensory descriptors in Spanish and English wine tasting notes, demonstrating significant cross-linguistic variation in metonymic patterns. Their corpus-based analysis reveals that Spanish employs more complex chains of metonymic relationships between sensory domains than English, particularly in mapping texture qualities to flavor experiences. 

Similarly, Sánchez-Manzanares (2013) has documented how taste descriptors frequently exhibit more explicit metonymic relationships between sensation and emotion than in English. 

Cross-sensory metaphoric and metonymic patterns could significantly enrich the authors' theoretical framework, particularly in explaining why certain foods perform differently with associated versus non-associated labels.

Methodological Recommendations

The authors should consider extending their methodology to include:

1. Metonymic analysis: Examining how taste descriptors function metonymically (PROPERTY FOR WHOLE, PART FOR WHOLE) would provide insight into why certain food-descriptor pairings generate stronger responses.

2. Cross-linguistic CFTs: Conducting parallel CFTs in other languages would reveal whether association patterns are language-specific or reflect more universal sensory categorizations.

3. Diachronic analysis: Investigating how taste descriptor associations have evolved over time could illuminate cultural and linguistic factors influencing current usage patterns.

4. Consumer segmentation: As the authors suggest, testing across different consumer populations with varying levels of gastronomic expertise would determine whether association effects vary by demographic factors.

Theoretical Implications

The findings significantly advance our understanding of linguistic relativity in the domain of taste perception. The paper demonstrates that synonymic taste words are not merely stylistic variants but semantic tools that shape sensory experience. This has implications beyond marketing to fundamental questions about how language structures perception.

The study provides evidence for embodied cognition theories by showing that linguistic categories influence sensory processing. The processing load explanation for why less associated taste words sometimes improve perception of disliked products is particularly compelling and merits further investigation.

Practical Applications

The research has clear implications for food marketing, menu design, and product naming. The authors rightly caution against direct application without further research in naturalistic settings, but their findings suggest several promising directions:

1. For most products, using descriptors most strongly associated with those products will enhance perceived taste.
2. For potentially disliked products, less associated descriptors may improve reception by increasing cognitive load and reducing attention to negative taste aspects.
3. The methodology offers a rapid technique for testing descriptor effectiveness without extensive corpus analysis.

Conclusion

This paper represents a significant contribution to both linguistic semantics and sensory science. Its methodological innovation in using CFTs to identify subtle semantic differences between synonyms provides an efficient alternative to corpus-based approaches while still capturing meaningful distinctions. The finding that taste perception is generally enhanced by more strongly associated taste descriptors, with important exceptions that merit further study, has both theoretical and practical implications.

To strengthen future research, I recommend incorporating  cross-linguistic perspectives, particularly focusing on metonymic extensions of taste terminology that may reveal cultural variations in taste conceptualization. Additionally, exploring how different consumer segments respond to taste descriptors would provide more nuanced guidance for practical applications.

The paper succeeds in extending the state of the art regarding taste descriptors and synonymy, but would benefit from more explicit attention to metonymic patterns in gustatory language. Nevertheless, it represents a valuable advancement in our understanding of how linguistic choices shape sensory experience.

References

Fernández Jaén, J. (2016). El olfato como fuente de conocimiento: Procesos metafóricos en los verbos de percepción olfativa. E-AESLA, 2, 285-297.

Ibáñez Rodríguez, M. (2010). El dominio vitivinícola: diversidad de lenguas y culturas terminológicas. In M. T. Cabré et al. (Eds.), Terminología y Sociedad del conocimiento (pp. 447-472). Peter Lang.

Jurafsky, D. (2015). The Language of Food: A Linguist Reads the Menu. W. W. Norton & Company.

López Arroyo, B. (2019). Comunicación multilingüe en el ámbito vitivinícola: Análisis contrastivo de descriptores sensoriales. Revista de Lenguas para Fines Específicos, 25(2), 134-152.

López Arroyo, B., & Roberts, R. P. (2017). Differences in wine tasting notes in English and Spanish: A corpus-based approach to sensory language. Languages in Contrast, 17(1), 69-95.

Sánchez-Manzanares, C. (2013). Valor neológico y lexicográfico de los términos sinestésicos. Revista de Investigación Lingüística, 16, 173-208.

Torres Soler, C. F. (2021). La experiencia del sabor: Metáforas conceptuales del gusto en español. Revista española de lingüística aplicada, 34(1), 319-342.

Zawisławska, M., & Falkowska, M. (2018). A dog's life and butterflies in the stomach: Metaphors and metonymies of TASTE in Polish and English. Advances in Cognitive Linguistics Research, 2(3), 1-19.

Author Response

Please see Table R2 in the attachment.

Author Response File: Author Response.pdf