Interoceptive abilities impairment correlates with emotional eating and taste abnormalities in children with overweight and obesity

.


Introduction
Overweight and obesity show increasing prevalence rates amongst children and adolescents over the last years in most European countries (Reinehr, 2020;Yngve et al., 2008).Being overweight or obese in childhood is associated with the development of somatic illnesses like cardiovascular diseases, hypertension or diabetes (Biro & Wien, 2010;Schienkiewitz et al., 2018A., Brettschneider, A. K., Damerow, S., & Rosario, A. S., 2018), which might manifest later in life (l'Allemand-Jander, 2010).Furthermore, overweight and obesity are related to emotional suffering as reflected in higher rates of depression and teasing experiences (Madowitz et al.,2012Knatz, Maginot, Crow, & Boutelle, 2012).Some studies point out that obesity might be related to disturbed general ways to regulate and process emotions, which in turn may indicate emotional eating (Aparicio, Canals, Arija, Henauw, & Michels, 2016;Czaja, Rief, & Hilbert, 2009), a concept originally introduced by Bruch (1964).Emotional eating is considered to be a response to distress, consequent of an inability to distinguish hunger from other aversive internal states, assumed to be rooted in early learning experiences (Harrist, Hubbs-Tait, Topham, Shriver, & Page, 2013;Koch & Pollatos, 2015).Similarly to emotional eating, smell and taste perception and the perception of internal bodily states have been linked to overweight and obesity (Fernández-Aranda et al., 2015;Herbert & Pollatos, 2018;Kure Liu et al., 2019;Robinson, Foote, Smith, Higgs, & Jones, 2021), highlighting the relevance of sensory and interoceptive processing for weight regulation.
Taste perception can be altered by various conditions such as genetic or environmental factors but also hunger and emotions (Noel & Dando, 2015;Risso, Drayna, & Morini, 2020), but the exact mechanisms behind alterations are not conclusively understood (Kure Liu et al., 2019).For instance, while Glöckner et al.,1986, Fikentscher, andUlrich (1986) reported increased taste sensitivity with hunger in obese patients undergoing treatment including fasting, this finding has not been replicated in a sample of fasting students (Pasquet, Frelut, Simmen, Hladik, & Monneuse, 2007).Regarding the influence of mood states and psychological variables on taste perception, Platte, Herbert, Pauli, and Breslin (2013) found that individuals with elevated depression levels rated sweetness as higher than individuals with lower depression levels, and positive and negative mood induction increased perceived intensity for sweet and bitter tastes in the more depressed group.Furthermore, individuals with higher depression levels correctly rated fat increasing with fat concentration, but after mood induction, were unable to distinguish low from high fat stimuli, highlighting the effect of emotional states on taste perception (Platte et al., 2013).Further analyzing this relationship, a study by Noel and Dando (2015) is of great relevance: The authors demonstrated that taste perception could be manipulated by day-by-day emotional states, elicited in hockey games attendees by differing game outcomes (wins, losses and ties).Interestingly, positive emotions were found correlated with enhanced sweet and diminished sour perception intensities while negative emotions were associated with heightened sensibility for sour and decreased sensibility for sweet tastes (Noel & Dando, 2015).
Thus, given that emotions have been found to influence taste perception, exploring the interaction between altered gustatory abilities and heightened emotional eating, particularly in children with overweight and obesity, would be highly relevant for practical implications (Braet & van Strien, 1997;Farrow, Haycraft, & Blissett, 2015).Investigating this relevant link, Braet and van Strien (1997) found in a sample of 9-12 year old children that individuals with overweight and obesity scored significantly higher on emotional eating scores than children with normal weight.In a more recent work on the causes of emotional eating, van Strien (2018) discussed that emotional eating may be an important risk factor for obesity especially in individuals with poor emotion regulation skills, and could serve as a mediator between depression and obesity.Besides depression, evidence also supports a link between emotional eating and anxiety in children and adolescents aged 8-18 years, with higher risk for emotional eating and loss of control over eating in children and adolescents with overweight (Goossens, Braet, van Vlierberghe, & Mels, 2009).A study by Michels et al. (2012) showed that stressful events, problems and negative emotions were associated with emotional eating and a more unhealthy dietary pattern in children.A more recent, cross-sectional study including children from 12 countries supports this notion, also connecting emotional eating with an unhealthy dietary pattern (Jalo et al., 2019).This observed connection between emotional eating and unhealthy eating may thus contribute to the development of overweight and obesity in children.
Regarding the role of taste perception in children and adolescents with overweight and obesity, our understanding thus far is limited and subject to ongoing debate.Donaldson, Bennett, Baic, and Melichar (2009) summarized that few studies so far showed a disturbance in some tastes, with reported reductions in sweet and salt thresholds.A recent review by Harnischfeger and Dando (2021) supports these findings, reporting impaired sense of taste in adults, adolescents and children with obesity.For instance, in children and adolescents aged 6-18 years with obesity, the overall ability to detect tastes, as well as the ability to detect salty, umami and bitter tastes was lower than for non-obese children and adolescents (Overberg, Hummel, Krude, & Wiegand, 2012).Naming taste as the main predictor for eating behavior, Mameli et al. (2019) observed a significantly lower ability to correctly identify taste qualities in children with obesity as compared to normal weight children.However, some findings provide only little support for strong relationships between weight status and taste sensitivity for fat, sweet, salty, sour or bitter tastes, for both adults and children (Cox, Hendrie, & Carty, 2016).Contradicting the assumption of impared taste sensbility in individuals with obesity even further, an increased sensibility in sweet, salty or fat tastes in individuals with obesity has also been observed (Pasquet et al., 2007).As gustation is so closely linked with olfaction (Mazzola et al., 2017), the sense of smell should be and has been considered in investigations on this matter.In support of this, a study conducted by Obrebowski, Obrebowska-Karsznia, and Gawliński (2000) revealed heightened smell perception in children with obesity.Additionally, in the context of adults, Fernández-Aranda et al. (2015) observed a decreased sense of smell but found no evidence of altered taste capacity in individuals with obesity.Simchen, Koebnick, Hoyer, Issanchou, and Zunft (2006) reported age-dependent interactions between taste and smell perception score and the BMI: Younger participants who had a BMI >28 exhibited lower odor detection, odor identification and taste perception scores.Overall, while the exact mechanisms and reasons for altered olfaction and gustation remain unclear, alterations in smell and taste perception can widely and robustly be observed in individuals and children with obesity.
Other than olfactory and gustatory perception, the perception of internal states such as hunger and satiety is essential for eating (Herbert & Pollatos, 2018).Emotional eating, which is linked to overweight and obesity as established above, implies eating while diregarding these cues for hunger and satiety (Jansen et al., 2003;Schachter, 1971).Therefore, alterations in the perception of these internal states may contribute to overweight and obesity.A prospective study by Koch and Pollatos (2014b) was able to demonstrate that externally driven eating behavior leads to a decrease in the accuracy to detect bodily changes in children with overweight and obesity, as measured in a heartbeat perception task.This study illustrates the relevance of interoceptive abilities in the context of weight regulation.Interoception, commonly defined as the perception of internal bodily states, involves the process of sensing and processing internal afferent information from various body tissues by the central nervous system (Craig, 2003;Khalsa et al., 2018).This capability allows the organism to carry out a range of regulatory processes, from homeostatic reflexes to emotion regulation (Chen et al., 2021;Quadt, Critchley, & Garfinkel, 2018).An individual's interoceptive experience can be described across various modalities and dimensions (Herbert, 2021;Herbert & Pollatos, 2018;Nord & Garfinkel, 2022;Pollatos & Herbert, 2018).Interoceptive modalities pertain to the different inner organs from which the afferent body information originates (e.g., cardiovascular, respiratory, and gastrointestinal systems; (Critchley & Garfinkel, 2017;Mai, Braun, Probst, Kammer, & Pollatos, 2019;Quadt et al., 2018).In a recent framework, Suksasilp and Garfinkel (2022) distinguished different dimensions of interoceptive processing, one of which is interoceptive accuracy (assessed through objective performance in behavioral tasks, such as the heartbeat tracking task; schandry task (Schandry, 1981) and interoceptive sensibility when referring to the self-reported sensations often assessed with questionnaires.
Until now, many studies focused on the cardiovascular system (Critchley, Wiens, Rotshtein, Ohman, & Dolan, 2004;Khalsa, Rudrauf, Hassanpour, Davidson, & Tranel, 2020;Larsson, Esposito, Critchley, Dienes, & Garfinkel, 2021;Nord & Garfinkel, 2022;Pollatos, Gramann, & Schandry, 2007;Pollatos, Schandry, Auer, & Kaufmann, 2007) and by doing so found that interoceptive accuracy was lower in both children (Koch & Pollatos, 2014b;Robinson et al., 2021;Willem et al., 2021) and adults (Herbert, Blechert, Hautzinger, Matthias, & Herbert, 2013;Herbert & Pollatos, 2014) with overweight and obesity.Evidence suggests that already children differ in their ability to detect ongoing signals deriving from the heart, and that this ability might function as a basis for their emotional experience in similar ways as it has been reported for adults (Koch & Pollatos, 2014a;Opdensteinen et al., 2021;Pollatos, Mönkemöller, Groppe, & Elsner, 2022).However, the recent review and meta-analytic work by Robinson et al. (2021) emphasizes that evidence for a strong connection of interoception and obesity is inconclusive.Interoceptive development has been highlighted in the broader context of mental health (Khalsa et al., 2018), and specifically in children and adolescents: Hechler (2021) proposes a bidirectional relationship between interoceptive processes and the development of mental health problems and chronic pain.Within a complex framework encompassing physiological, cognitive, and emotional aspects, somatic responses may contribute to the co-occurrence of mental health problems through interoceptive fear conditioning.Emphasizing the significance of interoceptive abilities in various cognitive and emotional processes, Opdensteinen et al. ( 2021) discovered a positive correlation between S.A. Mai-Lippold et al. cardiac interoceptive accuracy and emotion regulation.Similarly, Pollatos et al. (2022) identified advantages in decision-making associated with interoceptive accuracy in a large sample of children.Besides evidence for a connection of interoceptive processes with overweight and obesity, interoception has also been linked to taste, showing higher interoceptive accuracy in sweet-liking women as compared to sweet-disliking women (Iatridi, Quadt, Hayes, Garfinkel, & Yeomans, 2021).
To summarize, previous research suggests associations of overweight and obesity with altered taste and smell perception and heightened emotional eating, which in turn also found to be linked with each other.Further, obesity has been associated with impaired interoceptive abilities, and interoception has been connected to taste.Exactly how these factors intercorrelate, and whether interoceptive processes are relevant for the hypothesized interaction between emotional eating and gustatory as well as olfactory abilities in children with overweight and obesity is to our knowledge still unknown.
The goal of the present study was to investigate the relationship between olfactory and gustatory functions and emotional eating in children with overweight and obesity.We expected children with overweight and obesity to have higher emotional eating scores, and differ from normal weight children in their olfactory and gustatory perception.Additionally, interoceptive abilities, as quantified by the accuracy to detect cardiovascular signals and interoceptive sensibility, were investigated, and we expected to find impairments in children with overweight and obesity.Findings of this study might provide crucial information regarding the development and treatment of overweight and obesity in children.

Participants
We tested 23 (n = 8 boys) children whose BMI was above the 90th percentile, and 23 (n = 10 boys) whose BMI fell between the 24th and 76th percentile.All children were aged between 12 and 16 years.Eight children were classified as overweight (BMI >90th BMI percentile), and fifteen as obese (BMI >97th BMI percentile).Legal guardians had provided written informed consent before testing.Approval for the study was obtained from the ethics committee of Ulm University.Children received a small gift for their participation after the experiment, which they could choose from different options (e.g.drawing equipment, small board games, sports equipment like a badminton set or a small football).
We conducted Post-hoc Power analysis using G Power (Faul, Erdfelder, Buchner, & Lang, 2009), which revealed that we achieved statistical Power between 0.52 and 0.99 for observed significant effects, and statistical Power between 0.05 and 0.55 for non-significant observations.

Procedure
Children with overweight were recruited in local therapy groups (Jumbo Kids, Big Friends) in the southwest of Germany and tested individually by a trained student research assistant during their therapy sessions in a different room.Matched controls were recruited through advertising in schools and youth clubs in Ulm and tested at the laboratory of the Clinical and Health Psychology department.
Upon arrival, participants first filled in two short questionnaires and were familiarized with the procedure before body measures were taken.Then, interoceptive accuracy was assessed using a heartbeat perception test.Afterwards, taste and smell abilities were examined in a counterbalanced fashion.The whole experiment lasted about 40 min.

Body related measures
We determined children's height to the nearest 1.0 cm using a rail guided measuring tape to guarantee exact and reproducible measures.Children's weight and percentage of body fat were measured to the nearest 0.1 kg/% by means of a calibrated digital body fat scale (Tanita BC-532).BMI was calculated as the standard ratio of weight in kg divided by the square of height in meters.Individual BMI-values were also converted to Z-scores (BMI-SDS values, standard deviation score values) based on the national reference data for German children.

Interoceptive accuracy
The heartbeat perception task was performed following the Mental Tracking Method proposed by Schandry (1981) which has been shown to be sufficiently reliable and correlated with the ability to detect changes in other automatically innervated organs (Herbert et al., 2012).Just recently, a good intraclass correlation coefficient of 0.88 has been measured for this task regarding interoceptive accuracy (Santos et al., 2022).This procedure has also been adapted for children (Koch & Pollatos, 2014a).A short training interval of about 10s was followed by three intervals of 25, 40 and 35s.During each interval, children counted their own heartbeats by concentrating on their heart activity, while they were seated and not permitted to attempt any physical manipulation.
Meanwhile, children's actual cardiac activity was recorded using the mobile heart frequency monitor RS800CX (Polar Electro Oy, Kempele, Finland), a mobile device that enables the easy, non-invasive and -reactive recording of inter-beat-intervals and whose validity and reliability compared to alternative Electrocardiogram (ECG) measurement devices could be shown in children and adults (Gamelin, Baquet, Berthoin, & Bosquet, 2008;Kingsley, Lewis, & Marson, 2005;Nunan et al., 2008;Radespiel-Tröger et al., 2003;Rauh, Mahlke, Gottschalk, & Mück-Weymann, 2003).Signals were sampled at 1000 Hz and analyzed by the corresponding Polar ProTrainer 5 software (version 5.40.172).Interoceptive accuracy was determined via the following transformation: Higher scores indicate higher sensitivity to heartbeats, so that the maximum score of 1 indicates absolute accuracy of heartbeat perception and the minimum score of 0 indicates that the child did not perceive any of his or her heartbeats.

Interoceptive sensibility
Interoceptive sensibility was assessed by the 10-items subscale interoceptive awareness from the Eating Disorder Inventory-2 (EDI-2) (Garner, 1984;Paul & Thiel, 2005).All subscales of the EDI-2 show satisfactory internal consistency (α of 0.74-0.95) and this subscale provides a good test-retest reliability of 0.84 (Thiel & Paul, 2006).Questions were rated on a 6-point scale with higher scores indicating problems in the perception of internal signals related to hunger cues.The EDI-2 is suitable for children 10 years and older.

Emotional eating
Emotional eating was assessed by the subscale "emotional eating" of the German questionnaire "Fragebogen zum Essverhalten"/FEV, which is validated for the investigated age group of this study (Pudel, V. & Westenhöfer, J., 1989).The subscale consists of 10 items (e.g.Item 5: "When I feel troubled, worried or tense, I want to eat something.")and is scored on a five-point Likert scale (from 1 to 5), with high scores indicating a higher degree of emotional eating.All subscales of the questionnaire have satisfactory internal consistency (α between 0.74 and 0.87).

Taste assessment
The "Taste Strips" test consists of 16 taste impregnated filter papers (length 8 cm, tip area of 2 square cm) with one of these four taste qualities: sweet, sour, salty, and bitter (Landis et al., 2009).Internal consistency for each taste quality is as follows: sweet, 0.70; salt, 0.55; sour, 0.59; bitter, 0.69 (Fjaeldstad, Niklassen, & Fernandes, 2018).In accordance with other studies (Baskoy et al., 2016;Weiland et al., 2011), we used four varying concentrations for each taste quality (sweet: 0.4, 0.2, 0.1, 0.05 g/ml sucrose; sour: 0.3, 0.165, 0.09, 0.05 g/ml citric acid; salty: 0.25, 0.1, 0.04, 0.016 g/ml sodium chloride; and bitter: 0.006, 0.0024, 0.0009, 0.0004 g/ml quinine hydrochloride).The strips were placed on the anterior third of the extended tongue.Before each administration of a strip, the mouth was rinsed with a sip of room tempered natural mineral water.Tastes were presented applying increasing concentrations to the middle, left and the right side of the tongue.Overall gustatory function was calculated as the sum of correctly identified tastes per tongue location.Data were collected in the afternoon with one participant at a time slot.Children were thus not likely to be hungry due to the assessment taking place after lunchtime.No training was conducted before taste assessment.Landis et al. (2009) defined normogeusia, meaning normal taste perception, in adults as a Taste Strips test score of nine and higher (Landis et al., 2009), and hypogeusia, defined as lowered taste perception, as a score below nine (Landis et al., 2009).We used the same classification for our sample in this study.

Smell assessment
Smell testing was performed using "Sniffin' Kids" odor identification with 14 odors as suggested by Schriever et al. (2014).A test-retest reliability of ρ = 0.44 was found for this test (Schriever et al., 2014).
Using felt tip odor-filled pens, each odor was presented 2 cm under the nose for 3 s.Participants were asked to identify each odor by choosing from four presented descriptors.To do so, participants were allowed to smell each odor as often as necessary.The total of correct answers produced the odor identification score.As defined in the normative data report (Schriever et al., 2014), a score of >8 for children aged 12-14 and a score of >10 for children aged 15-16 indicates normosmia, meaning normal olfactory sensibility.

Statistical analyses
The Statistical Package for Social Sciences (SPSS, version 29) was used for all analyses.Continuous variables (BMI, BMI percentiles, age, emotional eating, interoceptive accuracy) were summarized as mean and standard deviation and compared using two-sample t-tests.Concerning taste, repeated measures analyses of variance (ANOVA) with the factor Taste Quality (sweet, sour, bitter, salty) and the between factor Group (children with overweight and obesity, children with normal weight) were used to examine differences in identification performance (percentage correct).Concerning smell, identification performances (percentage correct) were analyzed with respect to the between factor Group (overweight, normal weight).If assumption of sphericity was not met (Mauchly's sphericity test: p < .05),degrees of freedom for dependent variables were corrected conservatively by Greenhouse-Geisser.In case of significant main or interaction effects, post hoc ttests (with Bonferroni's correction for multiple testing when necessary) were performed.All data met the criteria for parametric analysis.
Pearson correlations and partial correlations coefficients were calculated to investigate the relationship between emotional eating, interoceptive abilities, BMI and taste/olfaction performance.

Sample characteristics, emotional eating and interoceptive accuracy
Sample characteristics as well as questionnaire data and the interoceptive dimensions are depicted in Table 1.T-tests were used to test for statistical significance (see also Table 1).
Children with overweight and obesity had significantly higher BMI and BMI percentiles than children with normal weight.As depicted in Table 1, children with overweight and obesity exhibited significantly higher emotional eating and significantly lower interoceptive accuracy.When controlling for the age difference between the weight groups, the difference in interoceptive accuracy remains statistically significant (F (1,43) = 4.24, p = .045,η 2 = 0.06).No significant group difference was observed concerning interoceptive sensibility.
We observed a descriptive inverse correlation between emotional eating and interoceptive accuracy for all children (see Table 2).When tested for both groups separately, only the children with overweight and obesity exhibited a significant inverse correlation as compared to the control group.As calculated with a calculator provided by Lenhard and Lenhard (2014) via https://www.psychometrica.de/korrelation.html,we identified a significant difference between the two correlations (Fisher's z = 2.273, p = .012).Concerning interoceptive sensibility, a significant positive correlation was observed indicating more emotional eating with higher scores of self-reported problems in interoceptive sensibility (see Table 3).

Taste functioning
Two children in the normal weight group and three children in the overweight and obese group are categorized as hypogeusic based on their test scores.The groups did not differ regarding taste functioning (M Normal = 12.09 (SD 2.57); M Overweight = 11.90 (SD 2.47); t(44) = 0.292, p = .772).The percentage of correct answers for the four taste qualities is visualized in Fig. 1

by group.
We observed a significant difference regarding accuracy in ratings of Taste quality (F(3,132) = 23.12,p < .001,η 2 = 0.345).Sweet and bitter taste were detected with significantly higher accuracy than salty taste.Detection rates for sour taste were significantly lower than all other flavors.Importantly, we observed a significant interaction Taste quality x Group (F(3,132) = 2.96, p = .035,η 2 = 0.063).When controlling for the age difference between weight groups, this interaction remained statistically significant (F(3,129) = 2.73, p = .046,η 2 = 0.216).
Bonferroni-corrected post hoc t-tests revealed that children with obesity and overweight detected sweet tastes with significantly higher accuracy S.A. Mai-Lippold et al. than normal weight children (t(44) = -2.25,p = .03).For all other taste qualities, children with normal weight scored higher than children with obesity and overweight on a descriptive level.This observation was clearest for sour taste (t(44) = 1.83, p = .07).
We observed significant inverse correlations between emotional eating and the percentage of correct identifications of bitter taste (r = − 0.47, p = .012)taste.No significant correlations were observed with respect to interoceptive abilities.

Discussion
The findings of our study unveiled a significant connection between body weight, emotional eating, interoceptive abilities and gustatory abilities in children.Specifically, overweight and obesity were linked to increased emotional eating, poorer interoceptive accuracy, and gustatory abnormalities.Notably, children with overweight and obesity demonstrated significantly higher detection rates for sweet tastes, while no notable differences were found for other taste qualities.Furthermore, higher emotional eating scores were correlated with lower interoceptive accuracy for children with overweight and obesity only, while lower self-reported interoceptive sensibility was significantly correlated with higher emotional eating in all children.Accuracy of the perception of salty and bitter tastes was significantly, negatively correlated with emotional eating.However, no significant effects were observed concerning the tested olfactory abilities.
Our results are in accordance with other studies showing that emotional eating is increased among children with overweight and obesity (Braet & van Strien, 1997;Farrow et al., 2015;Goossens et al., 2009;van Strien, 2018).Further, while we did not find a significant correlation between taste perception and interoceptive abilities in this sample, we did observe impaired interoceptive abilities in children with overweight and obesity, as well as an altered sweetness perception in this group.The present study thus emphasizes the idea that altered taste perception in children with overweight and obesity could be accompanied by a diminished focus on their own bodily signals, such as hunger and satiety.Future investigations should further evaluate the theorized correlative link between alterations in interoception and taste perception change in children with obesity.Considering the lower interoceptive focus observed in this study, it is possible that emotional eating in children with overweight and obesity stems from a withdrawal from one's own bodily signals and a resulting heightened focus on sensory-specific information related to external stimuli, particularly food.Koch and Pollatos (2014b) referenced longitudinal data indicating that eating behavior driven by external cues can hinder the accuracy of perceiving internal signals in young children, supporting this notion.Given that emotional eating is tied to emotional states that heighten attentiveness to negative emotion signals (Levitan & Davis, 2010;Madowitz et al., 2012), a shift in attention towards external stimuli may take place in order to attain positive emotional responses.In line with this idea, Singh (2014) suggests that negative emotional states trigger the consumption of palatable foods to elicit hedonic responses like positive gratification and comfort.Thus, interoception in children with overweight and obesity may be reduced due to emotional processes and a consequently heightened focus on external stimuli such as food.Given that emotion regulation is associated with interoceptive processes (Opdensteinen et al., 2021), a potential vicious cycle may arise.Reduced accuracy in detecting cues related to hunger and satiety in connection with a diminished ability to regulate unpleasant emotions might thus lead to the use of increased food intake as a coping strategy in the long run, aiding the development of overweight and obesity.We assessed interoceptive sensibility using the EDI-2 Interoceptive Awareness subscale (Paul & Thiel, 2005), which focuses assessment on emotional aspects and is specifically designed for eating disorders.While this may be useful for assessment in children with overweight and obesity, a more general measure might be better suited for the entire sample.Future studies could benefit from implementing other measures assessing interoceptive sensibility on a more general level and focusing more on physical sensations, such as the MAIA (Mehling, Acree, Stewart, Silas, & Jones, 2018).
We chose the EDI-2 Interoceptive Sensibility subscale to assess interoceptive sensibility because of its wide use as an interoceptive measure.However, focusing on emotions in 7 out of 10 items, the subscale might be suited to assess alexithymia or emotion regulation difficulty, rather than interoceptive sensibility.If this is the case, given that recent work has shown only weak associations of alexithymia and interoceptive deficits (Trevisan et al., 2019;Zamariola, Vlemincx, Corneille, & Luminet, 2018), the observed link between the EDI-IA subscale scores and emotional eating might indeed represent a link between emotional eating and difficulty regulating emotions, rather than

Table 2
Correlations with emotional eating.

Table 3
Correlations between correct idenfitication of the four different taste qualities and emotional eating, interoceptive accuracy and interoceptive sensibility.interoceptive sensibility impairment.Considering that we did not observe the hypothesized link between interoceptive sensibility and taste and smell perception, future studies might benefit from using different instruments to assess interoceptive sensibility and re-evaluate this potential connection.Given the observed correlation between lower interoceptive accuracy and higher emotional eating in children with overweight and obesity, out study still contributes meaningfully to the notion that impaired interoceptive abilities may be a contributing factor to the development and maintenance of overweight and obesity.
Regarding taste perception, we found that children with a BMI >28 exhibited significantly better detection of sweet tastes, while no significant differences were observed between the two groups for all other tastes.Given that participants completed the interoceptive sensibility questionnaire prior to taste testing, possibly, priming effects could have enhanced their attention on and consequently maybe perception of the presented tastes.However, since this all participants followed the procedure in this order, a potential priming effect could affect overall levels of perception, but not effects of group differences.
Research on taste perception in children with overweight and obesity provides evidence for altered taste perception as compared to normal weight children, but as with our results, findings on the exact mechanisms and direction of the relationship between obesity and the perception of taste qualities are inconsistent (Cox et al., 2016;Fernández-Aranda et al., 2015;Harnischfeger & Dando, 2021).In some studies, such as Bartoshuk, Duffy, Hayes, Moskowitz, and Snyder (2006), taste perception is even observed as unaltered in obesity as compared to normal weight.Based on previous findings highlighting the role of emotion in the perception of taste (Noel & Dando, 2015), we speculate that emotional aspects related to food intake might partially contribute to this effect, while Bartoshuk et al. (2006) suggest that the perceived intensities of taste can vary based on the individuals' experiences with the substances being evaluated.Noel and Dando (2015) discovered an association of enhanced perception of sweetness with positive emotions.Drawing from that, the heightened sensitivity to sweet tastes we observed in our participants may be associated with the pleasurable responses experienced after consuming sweets (Bernard et al., 2019).An anticipation of hedonic responses following food intake might lead to increased food consumption, particularly when emotional regulation is impaired, as can be observed for children with overweight and obesity (Aparicio et al., 2016).Due to the altered taste perception in this group, particularly regarding sweetness or fat (Lanfer et al., 2012;Overberg et al., 2012), an increase in consumption of sweet or fat tasting foods may result, contributing to the maintenance of overweight and obesity.
It is worth noting that gender differences in taste perception have been observed in previous studies, with women often exhibiting superior taste sensitivity (Doty & Cameron, 2009;Vignini et al., 2019).However, due to the small number of male participants in our study, we were unable to control for gender differences.Therefore, future studies should aim to assess gender-specific taste perception abilities in children to gain a more comprehensive understanding of the factors influencing taste perception in this population.An alternative interpretation of our results pertains to the potential link between increased intake of sugar-sweetened beverages and obesity, as indicated in previous studies (Malik, Schulze, & Hu, 2006;Malik & Hu, 2022;Vartanian, Schwartz, & Brownell, 2007).The taste of food provides a physiological feed-forward signal that promotes food intake (Berthoud, Sutton, Townsend, Patterson, & Zheng, 2006;Kure Liu et al., 2019;Smith, 2000).In light of this, it is conceivable that the heightened sweet taste perception observed among children with overweight and obesity may contribute to their preference for sweet foods (Lanfer et al., 2012) and possibly trigger increased consumption of sugar-sweetened beverages.
We did not observe group differences in olfactory functioning.Several reasons might explain this result.First, we only assessed odor identification; a more subtle effect might occur when assessing odor threshold or discrimination and including odors specifically associated with food (e.g.caramel).Second, other studies found altered olfactory function during hunger, fasting and after food intake in healthy participants (Albrecht et al., 2009;Pollatos, Gramann, & Schandry, 2007;Pollatos, Schandry, et al., 2007;Shanahan, Bhutani, & Kahnt, 2021;Stafford & Welbeck, 2011).It can be concluded that odor perception in general varies in regard of the state of satiety.Although studies assessing odor perception among obese have to be interpreted with caution due to heterogeneity, a trend of impaired olfactory abilities among children with overweight and obesity has been observed (Islam et al., 2015;Obrebowski et al., 2000;Simchen et al., 2006).Taking these findings into account, it is likely that methodological issues might have contributed to our results.A broader assessment of olfactory functioning should be implemented in future studies, e.g. using the assessment of odor thresholds as implemented in the "Sniffing Sticks".Furthermore, time to last food intake and consumed amount of food should be examined in association to olfactory functioning, as Kittel and Reitberger (1970) found a link between smell thresholds and time of food intake.They reported highest levels of odor thresholds 90min after food intake with variations in regard of the consumed amount.
We note that methodological limitations during taste assessment should be taken into account when interpreting the revealed association of taste perception and body weight.Since participants were recruited from local therapy groups, it is possible that having already received therapy, taste perception might have been altered.Low, Lacy, McBride, and Keast (2016) reported that decreases in BMI were associated with increased sweet gustatory sensitivity.Whether therapeutic changes in diet habits with probably reduced consumption of sweet foods also impacts taste perception is somewhat unclear and requires further observation (Nishihara et al., 2019).To obtain gustatory function, the taste strips method was used.However, this method has not yet been validated in the age group of our sample.Taking into account that Landis et al. (2009) reported decreasing taste functions with age, normogeusia among children and youths might be classified for overall lower values.Additionally, future studies might also include the assessment of palatability to elucidate if taste perception might be moderated by the subjective palatability of the food.Barthomeuf, S.A. Mai-Lippold et al. Droit-Volet, and Rousset (2009) observed an increased negative affect towards palatable foods (high-fat, high-sugar) among adolescents with obesity undergoing a weight-loss treatment, possibly due to the desire to eat this food when intake was restricted.Furthermore, assessing taste perception in a therapeutic setting did not allow for extensive control of last food intake.Taking into consideration that hunger and satiety are known to impact taste perception (Glöckner et al., 1986;Zverev, 2004) future studies should aim at controlling for last food intake and satiety.We further note that we placed no restrictions regarding food intake or other potential taste confounds like chewing gum or brushing teeth on our participants, which must be considered a limitation to our results.Finally, as we did not assess data on ethnic, socioeconomic, or educational backgrounds of our sample, we are unable to evaluate the influence of these factors on the observed effects.Inclusion of these aspects in future studies will be essential to improve our understanding of the relationship between interoception, taste and smell perception, emotional eating and body weight.
In this study, we were able to demonstrate increased emotional eating and reduced interoceptive accuracy in children with overweight and obesity.We further found alterations in taste perception, which were partly associated with emotional eating as assessed by questionnaire.Due to our limited sample size, future studies with larger samples are required to broaden our findings and further deepen our understanding of impaired interoception and altered taste perception in children with overweight and obesity.Our results reveal valuable insights into the intercorrelation of emotional eating, interoceptive abilities and perception of taste and smell in children with overweight and obesity.While the provided evidence is restricted due to methodological limitations, this study emphasizes the importance of investigating the potential effects of interventions aimed at improving interoceptive abilities and concurrent emotion regulation.It thus contributes meaningfully to the creation of new avenues for innovative interventions targeting lifestyle variables, such as physical activity and food preferences, in children.

Table 1
Characteristics of the sample.
Note: All values are mean ± SD unless stated otherwise; BMI= Body Mass Index; BMIPCT=BMI percentiles; SD= Standard Deviation.