Effects of single plant-based vs. animal-based meals on satiety and mood in real-world smartphone-embedded studies

The study investigates whether a single plant-based (vegetarian or vegan) versus animal-based (meat or fish) cafeteria meal acutely affects satiety, mood, and stress. Motivated by evidence linking higher dietary fiber intake with cardiometabolic and mental health benefits and hypothesized gut–brain signaling effects, the authors proposed that plant-based meals would produce higher satiety, better mood, and less stress than animal-based meals. They further posited that higher fiber would mediate beneficial effects, whereas higher sugar and fat would attenuate them, and that choosing plant-based meals would be more planned than impulsive. The work addresses gaps in understanding short-term physiological and psychological responses to plant-based meals in real-world settings, with implications for dietary interventions promoting planetary and human health. The authors preregistered four hypotheses regarding meal category effects, nutrient mediation (fiber, sugar, fat), decision-making (planned vs. impulsive), and masking effects under non-voluntary choices, and explored moderators including energy intake, fluids, environment, dietary habits, social interaction, personality traits, age, gender, and socioeconomic status.

Prior research suggests higher fiber intake benefits metabolic health and may modulate appetite and mood via gut–brain mechanisms. Meta-analyses indicate fiber-rich foods are associated with reduced non-communicable disease risk and modest weight benefits independent of energy intake. Acute mechanistic evidence links macronutrients and energy density to appetite control, and a randomized crossover study reported greater satiety after plant-based versus macronutrient-matched meat-based meals. Fiber fermentation products (short-chain fatty acids) have been shown to attenuate stress responses in a randomized trial. Psychological factors, such as planned food choices and self-control, influence dietary decisions; vegetarians/vegans have been reported to show higher depressive symptoms in some meta-analyses, though healthier diets and higher fiber intake have also been associated with lower depressive symptoms and anxiety. Interventions in food environments (availability, pricing, labeling) can increase plant-based meal selection without reducing overall sales. However, short-term effects of single meal composition on mood and satiety in real-world settings are under-explored, and reverse causation in food–mood relationships remains a concern.

Design: Three preregistered, real-world studies conducted Jan 2019–Feb 2020 in German (and some Austrian) university cafeterias assessed pre- and post-meal satiety and mood (and stress in sub-studies 2–3). Ethics approval: University of Leipzig Medical Faculty; OSF preregistration (https://doi.org/10.17605/OSF.IO/A7YTS). Samples: - Study 1 (app-based): 16,135 observations included after curation from users of iMensa/Mensaplan apps across most German university cafeterias. Participants reported predominant dietary adherence (omnivorous n=11,600; vegetarian n=3,456; vegan n=911). Reported meals were 47–61% animal-based across studies. - Study 2 (browser-based, free choice): n=173 omnivorous adults (mean weekly meat ~480 g) at cafeterias in Leipzig, Berlin, Halle, Jena. - Study 3 (browser-based, randomized choice): n=71 omnivorous adults, randomized to plant-based (vegetarian/vegan) or animal-based meals after pre-survey. Recruitment and eligibility: - Study 1: All app users invited; broad inclusion of omnivores, vegetarians, and vegans. - Studies 2–3: Cafeteria visitors >18 years with smartphone access; vegetarians/vegans excluded to enable randomization among omnivores; screened to exclude food allergies/intolerances, eating disorders, neurological/psychiatric disorders, or conditions affecting diet/mood. Measures: - Study 1: Pre/post hunger and happiness on 5-point emoji Likert scales; additional items: planned vs. impulsive decision, gender (female/male/diverse), predominant dietary adherence, fluid intake in last 2 h (<0.5 L/<1.0 L/>1.0 L), social interaction (alone/in company), smartphone use during meal (yes/no), finished meal (yes/no), sleep quality (good/normal/bad), and post-meal taste ratings (1–5 stars). Meal category (animal-based vs. plant-based) labeled via cafeteria descriptions and allergens using keyword rules; ambiguous entries excluded. Nutrient data (kcal, carbs, sugar, fat, saturated fat, protein per 100 g/portion) available for a subset (Nmax≈1,262); fiber content not available. - Studies 2–3: Pre/post hunger, happiness (contentment), and stress on 10-point scales; demographics (age, gender, height, weight for BMI), income, estimated portion size, side dishes/desserts, drank fluids (water, tea, coffee, etc.), social context, smartphone use, whether meal was finished. Questionnaires: Food Frequency Questionnaire (DEGS-1) to estimate habitual intake (kcal, macronutrients incl. fiber), NEO-FFI personality, TFEQ eating traits. Optional pre/post meal photos for compliance/time stamps. In Study 3, participants rated contentment with randomized allocation and reasons for (dis)content. Procedures: - Study 1: App prompts before and after a self-chosen cafeteria meal; captured date/time for entries. - Study 2: Free choice of meal; surveys immediately before and after eating; photo time stamps used to validate timing. - Study 3: Randomization to animal-based vs. plant-based category after premeal survey; instructed to eat assigned category; contentment assessed; compliance checked via photos. Data handling and timing: - Data curated for plausibility; duplicates/fake/non-compliant entries excluded; ambiguous meal categorizations removed. - Sensitivity windows defined using time stamps (liberal: >5 min and <3 h between pre/post; conservative: >20 min and <1.5 h). App pre–post intervals varied widely (mean 54±139 min; median 0.5 min); browser studies had tighter intervals (mean 15±7 min; median 14 min). Analytic strategy: - Primary outcomes: changes in hunger and mood (all studies); stress (studies 2–3). Linear mixed models (R lmer) with fixed effects of meal category (plant vs. animal), timepoint (pre vs. post), and their interaction; random intercepts for participant and cafeteria. Multiple-comparison adjusted alpha = 0.05/3 = 0.0167 per study. - Secondary analyses: Added macronutrients (subset), taste ratings, planned vs. impulsive decisions (chi-square), gender stratification, dietary adherence subgroups, and randomization contentment (Study 3). Exploratory analysis of carbohydrate quality via matched meal descriptions (white vs. whole-grain spaghetti bolognese) as a fiber proxy. Power: Predefined; Study 1 anticipated large N via app; Studies 2–3 powered on mood effect sizes from acute physical activity literature (d≈0.30–0.47) allowing for dropouts; Study 3 ultimately underpowered due to curation-related attrition.

- Primary effects (Study 1, n=16,135): • Hunger: No significant meal category×timepoint interaction (p=0.20). Hunger decreased by ~2 points (main effect of time; b=-1.99, t=-179.4, p<0.001). Plant-based choosers reported slightly lower overall hunger (main effect b=-0.10, t=-6.8, p<0.001) irrespective of time. • Mood: Significant interaction with smaller post-meal increases after plant-based vs. animal-based meals (interaction b=-0.06, t=-3.6, model comparison p<0.001). Pre-meal mood slightly higher for plant-based (3.52±1.0) vs. animal-based (3.48±1.0); post-meal mood increased in both (animal-based +0.26; plant-based +0.20). • Stress: Not assessed in Study 1. - Studies 2 and 3 (10-point scales): No significant interaction effects on hunger or mood in preregistered models, though directions broadly resembled Study 1. Stress showed no significant main or interaction effects. - Macronutrients (Study 1 subset, Nmax≈1,262): Plant-based meals had higher carbohydrates and sugar, lower fat and protein (all Wilcoxon p<0.01); energy and saturated fat did not differ significantly. Higher protein content predicted slightly higher satiety post-meal (b=-0.01, t=-3.1, p=0.002), with no interaction by meal category. - Fiber proxy (spaghetti bolognese, white vs. whole-grain; n=442): Whole-grain vs. white-flour versions associated with lower hunger (b=-0.30, t=-1.7, p<2.0×10^-16) and higher mood overall (b=0.15, t=0.8, p<8.0×10^-8); post-meal mood increased after whole-grain (b=0.41, t=1.98, p<0.05); post-meal hunger did not differ (b=-0.08, t=-0.28, p=0.78). Group sizes limited plant vs. animal comparisons for this analysis. - Decision-making (Study 1): Plant-based choices were more often planned (72%) vs. animal-based (62%) (χ²=174, p<2.2×10^-10). Planned decisions interacted with higher post-meal hunger (b=0.04, t=2.8, p=0.006) and higher mood (b=0.13, t=8.9, p<2.2×10^-16); these effects were attenuated for plant-based choices. - Randomization contentment (Study 3, n=71): Pre-randomization contentment similar across meal categories (χ²=9.6, df=9, p=0.27). Post-meal contentment reported more often in animal-based condition (χ²=6.1, df=2, p=0.0487), but contentment did not explain variance in satiety or mood (all model comparisons p>0.52). Mode of choice (voluntary vs. non-voluntary liked/disliked) showed some interactions for plant-based meals on post-meal hunger and mood in combined sub-study analyses (model comparisons p<0.004 and p=0.028, respectively). - Taste (Study 1): Animal-based meals rated slightly tastier (mean 3.86±1.23) than plant-based (3.77±1.27) (Kruskal–Wallis χ²(1)=21.8, p<0.001). Taste strongly associated with higher satiety and mood post-meal (e.g., post-meal×1 star: hunger b=0.6, t=6.4; mood b=-1.3, t=-19.8; post-meal×5 stars: hunger b=-0.5, t=-7.3; mood b=0.5, t=8.5). Taste effects were slightly stronger for animal-based meals (triple interactions all p<0.01). - Gender and context (Study 1): Females showed lower hunger (b=-0.23, t=-15.6) and slightly lower mood (b=-0.01, t=-0.9) vs. males; diverse gender showed higher hunger (b=0.21, t=4.0) and lower mood (b=-0.71, t=-13.8). Eating with company related to higher hunger (b=0.1, t=6.1) and mood (b=0.1, t=6.4). Finishing the meal related to higher hunger (b=0.1, t=5.4) and mood (b=0.4, t=20.2); small interactions with plant-based for post-meal ratings. - Dietary adherence subgroups (Study 1): Predominantly omnivores replicated lower post-meal mood and slightly higher post-meal hunger after plant-based choices; predominantly vegetarians replicated mood effect only; predominantly vegans showed the opposite (higher mood and lower hunger after plant-based). Taste preferences aligned with dietary adherence (omnivores/vegetarians preferred animal-based; vegans preferred plant-based). - Correlations: Greater meal-induced satiety associated with higher mood improvements in Study 1 (Spearman r=-0.19, p<0.001); no consistent hunger–stress correlation; mood–stress changes inversely correlated in sub-studies 2 and 3 (r≈-0.34 to -0.39, both p<0.001).

The preregistered analyses indicate that the category of a single cafeteria meal (plant-based vs. animal-based) has minimal acute impact on post-prandial hunger and only a small effect on mood, with plant-based meals associated with slightly smaller post-meal mood increases compared to animal-based meals. These findings challenge expectations that plant-based meals would acutely enhance satiety and mood via higher fiber content. Lack of fiber data and real-world meal compositions may have obscured nutrient-specific effects; plant-based meals in the sample had higher carbohydrates and sugar but lower protein, and protein modestly predicted satiety. A proxy analysis using whole-grain vs. white-flour spaghetti bolognese suggested potential benefits of higher-quality carbohydrates (likely higher fiber) on mood (and lower overall hunger), supporting the hypothesized role of carbohydrate quality/fiber for affect. Taste emerged as a dominant determinant of both satiety and mood, and plant-based meals were rated slightly less tasty on average, especially among omnivores and vegetarians. This may reflect cultural expectations and culinary familiarity favoring animal-based dishes. Decision-making style differed by meal category, with plant-based choices more often planned; randomization reduced preference effects on satiety/mood, though post-meal contentment was higher after animal-based assignments. Gender and whether meals were finished explained substantial variance in outcomes, whereas social context had smaller effects. Dietary adherence moderated effects: vegans showed higher mood and lower hunger after plant-based meals, consistent with preference alignment and expectations. Overall, results suggest that improving the palatability and protein content of plant-based cafeteria meals may be more impactful for satiety and mood than meal category per se. The small acute effects observed, together with methodological constraints (e.g., missing fiber data, variable timing), highlight the need for controlled and longitudinal designs to capture delayed or cumulative gut–brain effects of fiber and plant-forward eating. The findings inform public health strategies aiming to increase acceptance and satisfaction of plant-based options through culinary optimization and environmental nudges.

Across three real-world studies, single plant-based versus animal-based meals exerted minimal effects on post-meal satiety and only small effects on mood, with overall meal timing (pre vs. post) driving the largest changes. Protein content and perceived taste substantially influenced satiety and mood; plant-based dishes were slightly lower in protein and rated less tasty, suggesting targets for culinary and menu improvements. Results alleviate concerns that plant-based meals are inherently less satiating and emphasize the importance of enhancing palatability and protein quality to boost satisfaction and adoption. Future work should include longitudinal and controlled trials with precise nutrient characterization (including fiber), standardized timing, and broader populations to disentangle acute versus longer-term effects, and evaluate policy-relevant interventions (availability, pricing, descriptive labeling) combined with culinary optimization of plant-based offerings.

- Self-reported, subjective ratings with potential for non-compliance or fraudulent entries; large missing data for some covariates. - Nutritional data available only for a subset; fiber content entirely unavailable, precluding preregistered fiber analyses. - Variable and sometimes very short pre–post intervals in the app dataset; sensitivity restrictions reduced sample size and statistical power. - Emoji-based Likert scales in Study 1 were custom-developed and not fully validated; mood assessed as momentary happiness proxy rather than comprehensive mood scales. - Meal categorization relied on keyword-based cafeteria information; ambiguous entries excluded; no differentiation between vegetarian and vegan in the app study. - Short acute timeframe may miss delayed satiety effects of fiber (3–15 h) and microbiome-mediated mood effects (24–72 h). - Sub-studies 2–3 experienced higher-than-expected attrition after data curation, potentially underpowering some analyses, especially Study 3. - Limited generalizability: young, student population; omnivores only in Studies 2–3; self-identified dietary adherence may be imprecise (e.g., semi-vegetarians/flexitarians within the “predominantly vegetarian” group). - Important psychological and lifestyle factors (health awareness, dieting goals, sport habits, ethical concerns) were not assessed.