Chronotype and eating behavior in adults: associations with circadian preference across body mass index categories

Chronotype, defined as an individual's preference for timing of sleep and activity within the 24-h day, reflects inter-individual variation in endogenous circadian rhythms. While historically studied in the context of sleep behavior, chronotype is increasingly recognized as a determinant of metabolic health and eating behavior, including the behavioral regulation of appetite, with potential implications for body weight regulation and chronic disease risk. Circadian biology research has shown that internal biological rhythms interact with feeding schedules and metabolic processes, suggesting that temporal patterns of food intake may be as relevant to health outcomes as the content and quantity of food consumed (1–3). Several studies indicate that individuals with an evening chronotype are more likely to exhibit eating behaviors associated with adverse metabolic profiles, including delayed meal timing, greater frequency of nocturnal eating, and irregular eating patterns, compared with morning chronotypes (1, 4, 5). These temporal eating patterns are often accompanied by lower diet quality, increased consumption of energy-dense foods, and higher prevalence of unhealthy lifestyle behaviors, which have been linked to higher body mass index (BMI) and increased risk of overweight and obesity (1). Mechanistic studies suggest that misalignment between endogenous circadian rhythms and behavioral cycles—such as eating late in the biological night—can disrupt peripheral clocks in metabolic tissues, leading to impaired glucose metabolism, altered lipid homeostasis, and increased adiposity (3, 6–8). As temporal food intake acts as a potent zeitgeber (i.e., an external environmental cue that synchronizes biological rhythms) for metabolic processes, such misalignment may contribute to long-term energy imbalance and cardiometabolic risk beyond traditional measures of diet composition (9). Beyond meal timing and metabolic outcomes, chronotype appears to influence psychological and behavioral constructs relevant to eating behavior, including appetite regulation and reward sensitivity. Although higher cognitive restraint may reflect better dietary self-regulation, restraint is not a unitary construct and may be either flexible and adaptive or rigid and potentially maladaptive, depending on the psychological context. Evidence from adult and adolescent populations suggests that later chronotypes may show higher tendencies toward dysregulated eating behaviors—such as emotional eating and uncontrolled eating—which can mediate associations between chronotype and unhealthy weight outcomes (10–12). This heterogeneity reflects differences in study design, including the widespread use of cross-sectional approaches and variability in chronotype assessment methods. Despite growing interest in chronotype as a biopsychosocial determinant of eating behavior and metabolic health, research in this area remains heterogeneous and often limited by cross-sectional design and variation in chronotype assessment. Furthermore, potential effect modification by gender and BMI category has not been consistently examined, despite evidence of gender differences in circadian regulation and eating behavior (13). Therefore, a more nuanced understanding of how chronotype relates to specific dimensions of eating behavior and appetite-related self-regulation, across diverse adult populations, is needed. Addressing these gaps has implications for personalized nutrition and chronobiologically informed interventions aimed at improving eating behavior and metabolic health.

The aim of the present study was to investigate the association between chronotype (morning, intermediate, and evening) and key dimensions of eating behavior, with a specific focus on appetite-related self-regulation, examining potential differences across BMI categories and between genders.

A total of 411 fully completed questionnaires were available for screening. Of these, 25 were excluded because participants were under 18 years of age (n = 9) or had a body mass index (BMI) below 18.5 kg/m² (n = 16). The final study sample therefore consisted of 386 participants, including 279 women (72%) and 107 men (28%) (Figure 1).

The mean age of the participants was 31.4 ± 12.3 years, the mean height of the participants included in the study was 1.7 ± 0.1 m, the mean body weight was 68.2 ± 12.8 kg, and the mean BMI was 23.5 ± 3.3 kg/m².

Of the 386 participants, 305 (79%) reported no medical conditions. The remaining 81 participants (21%) reported at least one medical condition. Among these, the most frequently reported conditions were dyslipidemia (21%), hypertension (18%), irritable bowel syndrome (15%), thyroid disorders (9%), celiac disease (4%), type 1 diabetes mellitus (3%), and type 2 diabetes mellitus (2%). The remaining 27% reported other conditions, epilepsy, polycystic ovary syndrome (PCOS), gastroesophageal reflux disease (GERD), Raynaud's syndrome, and other less prevalent disorders.

Analyses were primarily conducted according to chronotype, with additional exploratory stratifications by BMI and gender.

Previous studies conducted on large populations indicate that chronotype plays a relevant role in shaping eating behavior and dietary habits. Individuals with an evening chronotype generally report later bedtimes, greater variability in breakfast timing, higher consumption of caffeinated and alcoholic beverages, and overall poorer diet quality compared with morning chronotypes (1, 2, 17). In contrast, morning chronotypes tend to follow more regular and nutritionally balanced eating patterns, with higher intakes of proteins, carbohydrates, vitamins, and minerals (1, 2, 18, 19).

The findings of the present study partially align with this literature. While no statistically significant differences were observed across chronotypes for uncontrolled eating and emotional eating, individuals with an evening chronotype tended to exhibit higher uncontrolled eating scores, consistent with previous observations of less regulated eating patterns among late chronotypes (1, 2, 18, 19). However, the most distinctive result concerned cognitive restraint, which was significantly higher in morning chronotypes, particularly among normal-weight individuals and women.

This finding is consistent with evidence suggesting that morning chronotypes are more likely to adopt structured and rule-based eating behaviors (4, 18). At the same time, it expands existing knowledge by highlighting that higher cognitive restraint is not necessarily associated with healthier outcomes, as rigid dietary control may coexist with emotional vulnerability and risk of disinhibited eating (20, 21). From a clinical nutrition perspective, higher cognitive restraint should not automatically be interpreted as beneficial. Flexible restraint may support meal planning and self-regulation, whereas rigid restraint may increase stress around eating and may coexist with emotional vulnerability or later disinhibition. Since the TFEQ-R18 does not distinguish between flexible and rigid restraint, the present findings should be interpreted cautiously. This distinction is particularly relevant in clinical practice, where understanding whether restraint reflects adaptive regulation or maladaptive rigidity is essential for nutritional counseling. Notably, direct comparisons between chronotype and cognitive restraint are still limited in the literature, and few studies have examined this relationship while simultaneously accounting for BMI and gender. In this respect, the present study contributes novel evidence to an underexplored area.

The association between cognitive restraint and BMI observed in this study is consistent with prior research indicating that dietary restraint often increases with higher body weight (22). Longitudinal studies further suggest a bidirectional relationship, whereby greater adiposity predicts subsequent increases in cognitive restraint rather than restraint leading to weight loss (23, 24). This mechanism may explain the elevated restraint scores observed among overweight participants, particularly among morning chronotypes.

In contrast, emotional eating was not associated with higher BMI in the present sample, as higher scores were observed among normal-weight individuals rather than overweight participants. This pattern may suggest that cognitive restraint in morning chronotypes is more strongly related to behavioral traits associated with circadian preference than to body weight status per se. This finding differs from some previous reports showing a positive association between emotional eating and BMI (22), but aligns with studies suggesting that emotional eating may not consistently translate into higher body weight, especially in younger or non-clinical populations (21). Differences in age, sample composition, and study design may account for these discrepancies.

Gender-related patterns observed in the present study should be interpreted with caution. Although no statistically significant differences emerged between men and women across BMI categories, women—particularly those with morning and evening chronotypes—tended to report higher mean scores for cognitive restraint and emotional eating. This descriptive pattern is consistent with previous evidence indicating that women generally report higher levels of dietary restraint and emotional eating than men (25, 26). Together, these observations support the potential role of sociocultural and psychological factors in shaping eating behavior among women, including greater dietary monitoring and sensitivity to body image norms (27–31). Correlation analyses further reinforced the complex interplay between eating behavior dimensions. Strong associations between uncontrolled eating and emotional eating were observed across all chronotypes, consistent with previous studies suggesting shared underlying mechanisms (26, 32). Additional correlations between cognitive restraint and emotional eating in morning and intermediate chronotypes highlight the coexistence of dietary control and emotional vulnerability, supporting models that conceptualize eating behavior as a dynamic balance rather than independent traits (20, 21, 33).

Overall, this study suggests that cognitive restraint is the eating behavior dimension most sensitive to chronotype, gender, and BMI, whereas uncontrolled eating and emotional eating appear more stable across circadian preferences. Morning chronotype, in particular, is associated with greater dietary control, which may represent either an adaptive or potentially maladaptive strategy depending on context. These results should be interpreted with caution, particularly in smaller subgroups, as limited sample sizes may reduce statistical power and increase the risk of spurious or unstable findings (34).

Taken together, the present findings underline the importance of considering chronotype in the study of eating behavior and support the integration of circadian preferences into personalized nutritional and preventive strategies.

First, the exploratory cross-sectional design of this study does not allow any inference of causality or temporal direction between chronotype and eating behavior. Consequently, the observed associations should be interpreted as descriptive and non-causal, and not as evidence of cause–effect relationships or clinical implications related to obesity.

Second, the study relied exclusively on self-reported data, including anthropometric measures and questionnaire-based assessments. Because height and weight were self-reported, BMI estimates may have been affected by reporting bias, including possible under-reporting of body weight and over-reporting of height. Although validated instruments were used to evaluate chronotype and eating behavior, self-reporting may be affected by recall bias or social desirability and may not fully reflect actual behaviors or physiological characteristics.

Third, participants were recruited through online platforms and informal social networks and were predominantly university students and their family members. This convenience sampling strategy may have resulted in a sample with relatively homogeneous sociodemographic characteristics, thereby limiting the generalizability of the findings to the broader adult population. In addition, voluntary participation may have introduced self-selection bias. Moreover, body mass index (BMI) was used as the sole anthropometric indicator. While this choice is consistent with the online observational design of the study, BMI does not allow differentiation between fat mass and lean mass. Future studies conducted in in-person settings and including direct assessments of body composition (e.g., bioelectrical impedance analysis) could provide a more nuanced interpretation of eating behavior patterns in relation to chronotype.

Finally, although analyses accounted for BMI and gender, other potentially relevant factors—such as age, socioeconomic status, educational level, physical activity, sleep characteristics, and lifestyle habits—were not included and may have acted as residual confounders. Sleep duration was not assessed in the present survey and therefore could not be included among the study variables, although it may influence both chronotype-related behaviors and appetite regulation. No a priori sample size or power calculation was performed, and some subgroup analyses may have been underpowered, with small subgroup sample sizes potentially resulting in spurious statistically significant findings. Thus, findings from this study should be interpreted with caution.

Chronotype appears to be associated mainly with cognitive restraint, with morning chronotypes showing higher levels of dietary control, particularly among women and normal-weight individuals, while uncontrolled and emotional eating remained stable across groups.

The strong association between uncontrolled and emotional eating highlights the interconnected nature of appetite-related behaviors, and the coexistence of restraint and emotional eating suggests that higher dietary control may not always reflect adaptive regulation.

Overall, these findings support the role of chronotype as a behavioral factor in appetite regulation and suggest its potential relevance for personalized, behavior-oriented nutritional strategies. However, these implications should be interpreted cautiously, as chronotype-based approaches are not currently included in nutritional guidelines.

The author(s) declared that financial support was not received for this work and/or its publication.

The datasets generated and analyzed during the current study are not publicly available due to privacy and data protection regulations, as the data contain information that could potentially allow participant identification. The data are available from the corresponding author upon reasonable request and in accordance with institutional and ethical approval. Requests to access these datasets should be directed to. c.dirosa@unicampus.it

The studies involving humans were approved by the Ethics Committee of the Campus Bio-Medico University of Rome (protocol 61.23 OSS, 17/05/2023). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

CD: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. CS: Formal analysis, Writing – original draft, Writing – review & editing. AG: Formal analysis, Visualization, Writing – review & editing, Writing – original draft. PG: Data curation, Investigation, Methodology, Writing – original draft. SQ: Data curation, Investigation, Methodology, Writing – original draft. SM: Supervision, Writing – review & editing. YK: Conceptualization, Supervision, Visualization, Writing – review & editing.

The authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.