Speaking of Food: Understanding How People Talk About Food Experiences

Paper Title

Speaking of Food: Understanding How People Talk About Food Experiences

Publication Info

• Topic area: Human-Food Interaction (HFI) and food-related communication in HCI design.
• Keywords: Human-Food Interaction, food experience, multisensory perception, linguistic strategies, temporal dynamics, non-expert communication, sensory profiles, expectation management, ambivalence, HCI design.

Background and Problem

• Problem / challenge: Limited understanding of how non-experts articulate multisensory food experiences in everyday contexts; prior research focuses on expert terminology or isolated sensory dimensions.
• Significance: Insights into non-expert food talk can inform HFI design, enabling systems that better align with users’ sensory and emotional experiences, fostering inclusive and empathetic communication.
• Motivation and related work: Existing studies explore objective food descriptions and technological mediation of sensory stimuli but lack empirical accounts of multisensory food experiences as a whole. Linguistic research has focused on expert discourse, leaving a gap in understanding non-expert communication. This paper aims to address this gap by examining how non-experts verbalise food experiences.

Solution

• Proposed approach: A mixed-methods study combining self-reports and explicitation-based interviews to explore how non-experts describe food experiences across pre-eating, during-eating, and post-eating phases.
• Novelty:
  1. Identification of recurring linguistic strategies: temporal, embodied, associative, and evaluative.
  2. Development of a four-dimensional lens (perceptual, cognitive, affective, and behavioural) to capture the temporal unfolding of food experiences.
  3. Derivation of five actionable design implications for HFI systems.
  4. Integration of diachronic (across phases) and synchronic (within moments) perspectives in food experience analysis.
• Procedure and key techniques:
  • Participants tasted five food stimuli representing basic tastes (sweet, sour, salty, bitter, umami).
  • Sequential sessions included tasting, drawing intensity graphs, verbal descriptions via semi-structured interviews, and rating tasks.
  • Qualitative thematic analysis and quantitative correlation analysis were conducted to interpret verbal accounts and self-reported data.

Results

• Concrete findings:
  • Participants structured food experiences through temporal phases, integrating sensory perception, cognition, affective reactions, and behavioural responses.
  • Familiarity and liking ratings showed high variability across food samples, with potato chips rated as most familiar and liked (M = 3.9, Median = 4).
  • Taste-matching ratings confirmed food samples as prototypical representatives of their respective basic tastes, though deviations highlighted the multidimensional nature of flavour perception.
• Advantage over baselines:
  • Extends prior research by focusing on non-expert food talk and integrating multisensory, temporal, and evaluative dimensions into HFI design.
• Experiments / evaluation:
  • Study involved 20 participants (aged 19–51, diverse demographics) tasting five food stimuli in a controlled lab setting.
  • Data included verbal transcripts, self-reported ratings, and graphical representations of taste intensity.
  • Analysis combined thematic coding and statistical correlation to explore patterns in food descriptions and preferences.
• Limitations and future work:
  • Small sample size and limited cultural diversity; future studies should expand participant demographics and stimuli sets.
  • Controlled lab setting may not reflect real-world eating contexts; ecological methods like experience sampling are recommended.
  • Individual differences in taste sensitivity were not systematically analysed; future work could explore this aspect.

Summary

This study investigates how non-experts verbalise food experiences, revealing recurring linguistic strategies (temporal, embodied, associative, evaluative) that connect sensory perception, cognition, emotion, and behaviour across pre-, during-, and post-eating phases. By integrating diachronic and synchronic perspectives, the authors propose a systematic lens for understanding food talk and derive five actionable design implications for HFI systems, including extending food experiences to full temporal journeys and mediating ambivalence. The findings highlight the importance of prioritising non-expert voices in HFI design to create inclusive, responsive, and culturally diverse food-interaction systems. Future research should explore larger, more diverse participant pools and real-world contexts to deepen understanding and refine design approaches.