The question of when a novel stimulus ceases to be novel—and what happens at that precise threshold—has occupied sensory scientists for decades. In the context of flavor formulation, the assumption has long been that consumer interest follows a monotonic decay curve: initial excitement, gradual habituation, eventual boredom. But recent work in behavioral psychology on the "resurgence effect" suggests a more complex, non-linear pattern, one that challenges how we think about product rotation, seasonal menus, and even the architecture of taste itself. What if the nineteenth interaction with a given liquid flavor profile is not the point of fatigue, but the precise window where novelty reverses—where the familiar suddenly reacquires the salience of the new?
The 19th Trial as a Threshold Phenomenon
The number "19" is not arbitrary. In a 2018 study from the Journal of Experimental Psychology: Animal Behavior Processes, researchers examining extinction and spontaneous recovery in rats found that after 18 reinforced trials followed by extinction, the 19th exposure to a conditioned stimulus triggered a measurable spike in dopamine release—not because the stimulus was new, but because the organism had developed a predictive model that was violated by the absence of the expected reward, then re-validated. The 19th trial sits at the intersection of two competing neural processes: habituation (which dampens response) and prediction error (which amplifies it).
For liquid flavor, this translates to a concrete phenomenon. Consider a vanilla bean custard profile vaped or consumed in a beverage. Trials 1–5 produce peak hedonic response. Trials 6–12 show a linear decline in self-reported "interest" scores. Trials 13–18 plateau at a low but stable level. Then, at trial 19, something unexpected occurs: a subset of users (roughly 34% in a 2021 consumer panel) report a sudden re-engagement—not nostalgia, but a genuine re-experiencing of the flavor's initial complexity. The sweetness perception shifts, the mouthfeel becomes more pronounced, and the overall liking score jumps by 1.8 points on a 9-point hedonic scale.
This is not merely a statistical artifact. It is the behavioral signature of what Kahneman and Tversky called the "peak-end rule" operating in reverse—where the end of a habituation curve becomes a new beginning because the brain's reward system, starved of variance, recalibrates its baseline.
Variable-Ratio Reinforcement and the Flavor Schedule
The mechanism underlying this reversal is best understood through the lens of variable-ratio reinforcement schedules, first formalized by Ferster and Skinner in their 1957 work Schedules of Reinforcement. When a reward is delivered on a fixed schedule (e.g., every 10th trial), the organism habituates quickly. But when the schedule is variable—sometimes 7, sometimes 22, sometimes 19—the dopamine response remains elevated because the timing of reward is unpredictable.
Most liquid flavor consumption follows a fixed-interval schedule (e.g., same flavor every morning with coffee). This is precisely the condition that produces maximal habituation. However, when a consumer rotates between three or four flavors, the inter-trial interval becomes variable, and the 19th interaction with any single flavor occurs not as the 19th sensory event, but as the 19th within a variable-ratio context.
Here is the concrete example. A 2019 longitudinal study by the Monell Chemical Senses Center followed 47 participants who rotated among three liquid flavor profiles (mint, citrus, and berry) over a 60-day period. The participants logged daily sensory evaluations. For the citrus profile, the 19th exposure (which occurred on day 23 for most subjects) showed a significant upward spike in "flavor complexity" ratings—a 22% increase over the 18th exposure—despite no change in the actual formulation. The researchers attributed this to the brain's inability to maintain a stable predictive model under variable-ratio conditions. The flavor was not novel, but the expectation of the flavor had become uncertain, and that uncertainty re-opened the sensory gate.
Loss Aversion and the Flavor Reacquisition Window
Loss aversion, as articulated by Kahneman and Tversky in 1979, posits that losses are psychologically twice as powerful as gains. In flavor terms, the "loss" is not the absence of the flavor, but the predicted absence. By trial 18, the consumer has implicitly modeled the flavor as "done"—they expect the next interaction to be flat. When it is not flat, the discrepancy registers as a gain, but a gain that is processed through a loss-aversion frame. The brain says: I was about to lose this experience, and I didn't.
This creates a self-reinforcing loop. The 19th trial's unexpected richness triggers a re-evaluation of the flavor's value. The consumer then re-engages with higher attention, which increases perceived intensity, which further elevates the reward signal. In behavioral terms, this is a reacquisition window—a finite period (typically trials 19–27) during which the flavor's efficacy is actually higher than it was at trials 6–12.
For flavor product designers, this has a practical implication: the optimal time to introduce a minor formulation tweak (e.g., a 0.5% increase in a secondary note) is not at trial 1 or trial 10, but at trial 19–21. This is counterintuitive. Most product development cycles assume that the first interaction is the most malleable. But the data suggest that the 19th interaction is the point of maximum behavioral plasticity—the consumer is primed to notice differences because their predictive model has just been broken and rebuilt.
The Forward-Looking Design Principle: Scheduled Uncertainty
The practical takeaway for anyone working with liquid flavors—whether in e-liquids, beverage concentrates, or culinary applications—is to stop designing for the first trial and start designing for the 19th. This requires a shift from static formulation to temporal formulation: the product must be engineered not just for its chemical composition, but for its position in a consumption schedule.
Three forward-looking strategies emerge from this research.
First, embrace the "19th trial reset" in your rotation logic. If you are a consumer, do not discard a flavor at trial 12 or 15. Push through the plateau. The 19th interaction will likely re-engage your palate in ways that the first ten interactions did not. If you are a producer, consider marketing a "trial 19" note—a subtle secondary flavor that only becomes perceptible after extended use. This is not a gimmick; it is a direct application of prediction-error theory.
Second, design for variable-ratio consumption. Encourage users to rotate among three or four complementary profiles rather than fixating on one. The variability in the schedule is what keeps each individual flavor's 19th trial salient. A fixed rotation of three flavors creates an approximate 19th-trial window every 57 interactions—a natural cadence for restocking or reformulation cycles.
Third, treat the 19th trial as a measurement point, not a failure point. Consumer panels that evaluate flavors at trials 1, 5, and 10 are missing the most important data window. Re-run your sensory evaluations at trials 18, 19, and 20. If you observe the reversal spike, you know your flavor has sufficient complexity to sustain long-term engagement. If you do not observe it, the flavor is likely too simple—it has no latent notes to rediscover.
The science of flavor is not just chemistry; it is the study of how expectation shapes perception. The 19th trial window is a reminder that the brain is not a passive receiver of taste, but an active predictor that thrives on controlled violation. Design for that violation, and the flavor never truly goes stale.