When words disrupt flavour memory: Verbal overshadowing in gustatory recognition

Verbal overshadowing (VO) refers to the phenomenon whereby recognition of a stimulus becomes impaired after it has been verbally described, an effect that has been extensively documented in the visual domain. However, VO in gustatory recognition remains underexplored. This is despite the fact that gustation is lexically more impoverished than vision across many languages, including English and Chinese. The present study reports three experiments examining whether and how verbalisation disrupts gustatory recognition. In Experiment 1, forty native speakers of Mandarin Chinese were randomly assigned either to provide a verbal description of the target juice prior to recognition (n = 20) or to complete a nonverbal filler task (n = 20). Recognition accuracy was significantly lower in the verbalisation group than in the control group, demonstrating a robust VO effect for gustation. Experiment 2 examined whether explicitly alerting participants to the potential for VO would mitigate its effect. Participants in the informed verbalisation condition were warned that their descriptions might not accurately reflect their later gustatory memory. Nonetheless, recognition accuracy remained significantly lower than in the control group, indicating that awareness alone does not attenuate VO. Experiment 3 further tested whether combining VO awareness with explicit instructions to suppress description-based information would reduce the effect. Even under these conditions, participants showed significantly impaired recognition accuracy, suggesting that verbal interference persists despite deliberate attempts to rely solely on sensory memory. Across the three experiments, participants completed two counterbalanced recognition runs, one for a sour and one for a sweet stimulus set, and analyses were conducted after exclusion of participants who failed the gustatory screening criterion. Together, these findings provide convergent evidence that verbalisation reliably impairs gustatory recognition and that neither increased awareness nor deliberate suppression moderates this effect. The results are discussed with reference to four candidate mechanisms, including recoding interference, transfer-inappropriate processing shifts, retrieval-based interference and criterion-shift accounts.