The Art and Science of Wine Tasting | Qian Janice Wang | TED

The concept of complexity in wine is often used but poorly defined, with various interpretations across individuals. This presentation explores complexity in smell and taste, drawing parallels with cognitive ideas typically reserved for art. In psychology, the Berlyne curve illustrates the relationship between complexity and liking, suggesting an upside-down U-shape where both too little and too much complexity are disliked, with an ideal "sweet spot" in the middle. This idea was demonstrated with audio clips: a simple note, a slightly more complex tune, and a highly complex composition, with the latter often being less preferred. However, the question arises whether this principle holds true for wine. Wine critic Matt Kramer states that complexity is the greatest standard for wine quality, defining it by the ability to repeatedly discover new aromas and tastes in a glass. He suggests that the greatest wines are "seemingly limitless." As a scientist, the speaker aims to define and investigate this vague concept. Two main approaches to complexity are identified: chemical complexity (the molecules in the glass) and psychological complexity (the perception in the drinker's mind). The first study investigated chemical complexity by blending two wines (50/50) to create a chemically more complex product. Participants tasted six red wines (three single varieties and three blends) blind. Surprisingly, beginners, not experts, were slightly better at guessing which wines were blends. More importantly, the blends were not rated as more complex than the single varieties. This suggests that chemical complexity in the glass does not directly translate to perceived complexity. The only factor that consistently predicted higher complexity ratings and willingness to pay was the perception of oak flavors (e.g., vanilla, spice, cinnamon). The second study focused on psychological complexity, exploring how perception of complexity changes over time. Two types of complexity were introduced: static (how many flavors are perceived at any given moment) and dynamic (how flavors evolve in the mouth over time). Music examples were used to illustrate these concepts. For this study, Madeira wines of different ages (3, 10, and 20 years) were used, as older Madeira is generally believed to be more complex. To eliminate visual cues, black glasses were used. Novices and experts tasted the wines and rated their complexity. Experts consistently rated older wines as more complex, while novices showed only a slight, non-significant trend. This indicates that the idea of increasing complexity with age holds true for experts. Participants also performed a "temporal check all that apply" (TCATA) test, continuously ticking perceived descriptors over 30 seconds, including the aftertaste. A video demonstrated the temporal trajectory of flavors for the three Madeira wines. For experts, the flavor evolution curves for the different aged wines were distinct, particularly the 20-year-old. For novices, the curves were much more similar, indicating less differentiation in flavor evolution. This highlights that complexity is not solely in the glass but also in the mind of the drinker, with novices and experts having different mental experiences and concepts of complexity. While the Berlyne curve suggests a limit to desirable complexity in music, evidence for "too much" complexity being negative in wine is yet to be found. The speaker encourages slowing down and observing flavor evolution when eating or drinking.