Eating Disorder-Specific Interoceptive Perception Questionnaire (EDIP-Q)

Abstract

The Eating Disorder-Specific Interoceptive Perception Questionnaire (EDIP-Q) is a specialized, psychometrically validated self-report assessment developed by Julie Ortmann, Annika P. C. Lutz, Gitta Rose, Christian Happ, Claus Vögele, André Schulz, and Zoé van Dyck (2023, 2024). It was engineered to address a critical assessment gap in clinical psychology and psychiatric psychodiagnostics: the accurate, nuanced measurement of an individual's capacity to perceive, process, and differentiate internal bodily signals related to food intake and emotional experience. Interoceptive disruption—the impaired awareness or misinterpretation of afferent physiological signals arising from the visceral autonomic system—has long been conceptualized as a foundational transdiagnostic vulnerability factor in eating disorders (EDs), including anorexia nervosa (AN), bulimia nervosa (BN), and binge-eating disorder (BED).

Developed through a rigorous multi-stage psychometric pipeline involving extensive item generation from legacy measures, comprehensive scrutiny by a multidisciplinary panel of clinical experts, and rigorous empirical validation across two independent German adult samples ($N = 1,269$ total across phases), the final instrument consists of 25 items structured across four distinct first-order dimensions: Hunger (5 items), Satiety (7 items), Emotions (7 items), and Discrimination between hunger and emotional states (6 items). Items are calibrated along a 7-point Likert scale (1 = Does not apply at all to 7 = Applies completely), wherein lower composite and subscale scores directly index more severe deficits in interoceptive processing.

Psychometric evaluations demonstrate excellent internal consistency, with a total scale Cronbach's $\alpha$ of .94 and subscale coefficients ranging from .86 to .93. Structural equation modeling and exploratory/confirmatory factor analyses corroborate both a robust four-factor correlated model (CFI = 0.95, TLI = 0.94, RMSEA = 0.052, SRMR = 0.047) and a hierarchically organized second-order global EDIP construct (CFI = 0.95, TLI = 0.94, RMSEA = 0.053, SRMR = 0.050). The scale demonstrates pronounced construct, convergent, and discriminant validity, effectively differentiating individuals with diagnosed eating pathology from non-clinical controls, while mapping distinct interoceptive profiles across anorexia, bulimia, and binge-eating diagnostic phenotypes. The EDIP-Q provides an indispensable assessment framework for both clinical psychopathology research and precision treatment planning.

Keywords

Eating Disorders, Interoceptive Perception, Interoceptive Awareness, Hunger, Satiety, Emotion Discrimination, Anorexia Nervosa, Bulimia Nervosa, Binge-Eating Disorder, Psychodiagnostic Measures, Visceral Sensitivity, Psychometrics

Authors

The Eating Disorder-Specific Interoceptive Perception Questionnaire was designed and validated by an international team of behavioral scientists, clinical psychologists, and psychophysiologists based primarily at the University of Luxembourg:

• Julie Ortmann, M.Sc., Ph.D. Candidate — Department of Behavioural and Cognitive Sciences, Institute for Health and Behaviour, Faculty of Humanities, Education and Social Sciences, University of Luxembourg, Esch-sur-Alzette, Luxembourg. ORCID: 0000-0002-2793-7296. Email: [email protected] (Corresponding Author).
• Annika P. C. Lutz, Ph.D. — Research Scientist, Department of Behavioural and Cognitive Sciences, University of Luxembourg, Esch-sur-Alzette, Luxembourg. ORCID: 0000-0002-3247-3262.
• Gitta Rose, Dipl.-Psych. — Independent Clinical Psychologist, Licensed Psychotherapist, Germany.
• Christian Happ, Ph.D. — Senior Researcher and Methodologist, Yolande Asbl, Betzdorf, Luxembourg.
• Claus Vögele, Ph.D. — Full Professor of Clinical and Health Psychology, Head of the Institute for Health and Behaviour, Department of Behavioural and Cognitive Sciences, University of Luxembourg. ORCID: 0000-0001-8086-0788.
• André Schulz, Ph.D. — Associate Professor of Psychobiology and Neurophysiology, Department of Behavioural and Cognitive Sciences, University of Luxembourg. ORCID: 0000-0002-9381-2651.
• Zoé van Dyck, Ph.D. — Associate Professor of Clinical Psychology and Psychotherapy, Clinical Psychopathology and Health Laboratory, Department of Behavioural and Cognitive Sciences, University of Luxembourg. ORCID: 0000-0001-5240-4902.

Institutional Correspondence Address:
Julie Ortmann, Department of Behavioural and Cognitive Sciences, University of Luxembourg, Campus Belval, Maison des Sciences Humaines, 11 Porte des Sciences, L-4366 Esch-sur-Alzette, Luxembourg. Electronic correspondence: [email protected].

Purpose

The primary clinical and psychometric purpose of the Eating Disorder-Specific Interoceptive Perception Questionnaire (EDIP-Q) is to provide an objective, granular, and empirically robust assessment of eating disorder-specific interoceptive deficits. While broad psychiatric instruments like the Eating Disorder Inventory (EDI; Garner et al., 1983) and general interoception inventories such as the Multidimensional Assessment of Interoceptive Awareness (MAIA; Mehling et al., 2012) or the Body Perception Questionnaire (BPQ; Porges, 1993) capture facets of somatic awareness, they frequently fail to differentiate visceral gastrointestinal homeostatic signaling (e.g., gastric mechanoreceptive stretching, ghrelin-mediated appetite onset) from subjective emotional distress (e.g., anxiety, sadness, irritability). Consequently, researchers and clinicians have historically lacked a single, psychometrically sound instrument explicitly dedicated to quantifying the intersection of nutritional homeostatic processing and emotional differentiation.

Clinically, the EDIP-Q serves three foundational functions:

1. Diagnostic Stratification and Phenomenological Profiling: Different clinical variants of feeding and eating disorders present with starkly heterogeneous somatic perceptual profiles. Patients with restrictive anorexia nervosa often suppress, deny, or experience severe blunting of hunger signals while simultaneously reporting visceral hypersensitivity or distress to normative gastric distension. In contrast, individuals with bulimia nervosa and binge-eating disorder typically report profound deficits in satiety registration and a severe inability to distinguish negative emotional states (such as acute stress, dysphoria, or loneliness) from biological hunger pangs. The EDIP-Q produces distinct dimensional subscale profiles that identify the exact locus of somatic misperception.
2. Individualized Formulation for Psychotherapeutic Interventions: Contemporary evidence-based protocols, including Enhanced Cognitive Behavior Therapy (CBT-E; Fairburn, 2008), Acceptance and Commitment Therapy (ACT), and Mindful Eating or Intuitive Eating frameworks (Tylka, 2006), rely heavily on helping patients reconnect with visceral states. Administering the EDIP-Q prior to treatment enables therapists to target psychoeducation, gastric biofeedback, interoceptive exposure, and somatic awareness interventions directly to the patient's primary deficit area—whether it is visceral discrimination, emotional alexithymia, or satiety thresholds.
3. Outcome Monitoring and Relapse Prevention: Normalization of body mass index (BMI) or cessation of compensatory purging does not necessarily indicate psychological recovery. Many individuals in apparent behavioral remission continue to harbor profound interoceptive confusion, leaving them vulnerable to relapse during acute life stress. Longitudinal tracking with the EDIP-Q offers clinicians an empirical metric to track the gradual restoration of visceral feedback loops across treatment stages.

Psychological Construct

The central psychological construct operationalized by the EDIP-Q is Eating Disorder-Specific Interoceptive Perception (EDIP). Historically defined by Sherrington (1906) and modernized by neurobiologists such as A. D. Craig (2002) and Hugo Critchley (2004), interoception encompasses the ongoing afferent neural reception, central representation, integration, and subjective appraisal of physiological signals originating within the internal visceral organs (e.g., heart rate, gastric motility, respiratory status, thermal state, and blood glucose fluctuations). In the context of eating pathology, interoception is not a monolithic trait; rather, it is a multi-compartment construct composed of four interrelated yet functionally autonomous cognitive-perceptual dimensions:

1. Hunger Perception

This subscale captures an individual's sensitivity, timing, and acuity in recognizing endogenous energy-depletion cues. Biologically mediated by descending neuroendocrine cascades (involving hypothalamic arcuate nucleus stimulation, ghrelin secretion, and vagal afferent firing triggered by gastric emptiness and rhythmic migrating motor complexes), normative hunger perception allows individuals to register somatic appetite before entering extreme physiological starvation. In the EDIP-Q, this construct measures whether the respondent can easily sense stomach rumbles and energy drops, or whether awareness remains absent until extreme physical depletion occurs (e.g., Item 6: "I do not realize that I am hungry until I am extremely hungry"; Item 11: "Even when I have not eaten anything for a long time, I do not feel hunger").

2. Satiety Perception

Satiety perception reflects the conscious registration and hedonic evaluation of visceral signals indicating energy sufficiency and gastrointestinal filling. These cues rely upon gastric mechanoreceptors signaling muscular wall stretching via the vagus nerve, paired with postprandial release of anorexigenic hormones including cholecystokinin (CCK), peptide YY (PYY), and glucagon-like peptide-1 (GLP-1). Individuals with high satiety awareness readily distinguish between comfortable nutritional satisfaction and painful overfullness (e.g., Item 10: "I notice when I reach a point of comfortable satiation"). Conversely, severe deficits are manifested as an inability to detect termination signals until acute discomfort, nausea, or stomach overdistension ensues (e.g., Item 13: "I do not notice that I am full until I have completely stuffed myself").

3. Emotion Perception

Interoception provides the somatic bedrock for emotional feeling states, consistent with the James-Lange and Schachter-Singer models of emotion and modern neuroimaging evidence of insular cortex recruitment during affective processing. The Emotion Perception subscale of the EDIP-Q evaluates subjective emotional clarity and interoceptive sensibility regarding affective states—such as recognizing feelings of sadness, anxiety, anger, or joy. Individuals with impairments in this domain struggle with affective awareness (bordering on alexithymia), experiencing emotional distress as an undifferentiated, diffuse bodily unease (e.g., Item 2: "I can recognize my emotions easily"; Item 21: "I can recognize when I feel sad, angry, or anxious well").

4. Discrimination Between Hunger and Emotions

Perhaps the most clinically vital dimension of the EDIP-Q is the capacity to disentangle nutritional homeostatic cues from autonomic sympathetic arousal accompanying affective states. Visceral sensations triggered by psychological stress, anxiety, excitement, or dysphoria—such as epigastric fluttering, visceral tension, autonomic hyperarousal, or gastric cramping—share overlapping afferent pathways with true metabolic hunger. Patients with eating pathology frequently display profound interoceptive confusion, misidentifying emotional discomfort as somatic hunger, or experiencing the physical sensations of emotional turmoil as an urge to consume food (e.g., Item 5: "I do not feel a difference between hunger and stress"; Item 18: "I do not feel a difference between hunger and sadness"; Item 7: "Hunger and emotions feel the same to me"). Successful discrimination prevents emotionally driven eating, compensatory behaviors, or maladaptive food avoidance.

Theoretical Framework

The architecture of the EDIP-Q is rooted in over six decades of psychiatric, cognitive-developmental, and neurobiological theory regarding the etiology and maintenance of eating disorders:

Hilde Bruch and Interoceptive Confusion

The foundational historical cornerstone of this measure traces back to the pioneering psychoanalytic and developmental observations of child psychiatrist Hilde Bruch (1962, 1973). In her seminal texts, Eating Disorders: Obesity, Anorexia Nervosa, and the Person Within, Bruch argued that severe disturbances in nutritional self-regulation do not arise purely from a pursuit of thinness, but rather from early developmental failures in learning to conceptualize and name bodily needs. Bruch posited that when early caregivers fail to provide contingent, attuned responses to an infant's biological cues—for example, feeding the child when they are crying from cold, fear, or boredom, or withholding food when the child expresses genuine metabolic need—the developing individual fails to acquire the internal schemas necessary to distinguish visceral hunger from emotional distress.

This early developmental divergence manifests in adulthood as profound "interoceptive confusion" or visceral alienation. Bruch described individuals with anorexia nervosa as experiencing their physical bodies as an alien, hostile entity whose biological demands must be rigidly subdued. The EDIP-Q operationalizes Bruch's theoretical clinical observations into a reliable, psychometrically measurable format, providing the first direct metric of this exact discriminative failure.

The Predictive Processing and Active Inference Framework

Contemporary cognitive neuroscience understands interoception through the lens of Predictive Coding and Active Inference, championed by theorists such as Karl Friston, Anil Seth (2013), and Sahib Khalsa (2018). Within this framework, the brain functions as a hierarchical Bayesian prediction machine. The central nervous system constantly generates "top-down" generative models (priors) regarding anticipated internal homeostatic states and compares these predictions against incoming "bottom-up" visceral afferent signals transmitting information through the vagus nerve and spinothalamic tracts.

Any discrepancy between top-down expectations and bottom-up afference generates an interoceptive prediction error. In healthy individuals, the brain minimizes prediction errors either by updating its internal beliefs (perceiving that food is needed or that satiety has arrived) or by taking action via active inference (eating, stopping eating, resting). In eating disorders, predictive processing undergoes catastrophic distortions:

• In Anorexia Nervosa, hyper-rigid, dogmatic cognitive priors regarding weight, calories, and fear of fullness override and downweight incoming bottom-up metabolic signals, resulting in an effective neurobiological suppression of conscious hunger awareness.
• In Bulimia Nervosa and Binge-Eating Disorder, interoceptive prediction errors are chronically attenuated or misinterpreted, with hyper-reactivity to food cues paired with blunted visceral satiety feedback. Emotional autonomic arousal is systematically predicted to be metabolic hunger, driving compulsive active inference in the form of binge-eating episodes.

Neurobiology of the Insular Cortex

At the anatomical level, interoceptive processing is centrally integrated within the insular cortex. A. D. Craig's neuroanatomical model illustrates that primary visceral inputs (conveying cardiovascular, gastrointestinal, immune, and metabolic data) arrive at the posterior insula via the nucleus of the solitary tract and the thalamus. These signals are sequentially relayed forward to the mid- and anterior insular cortex (AIC). The anterior insular cortex serves as the central hub where raw somatic sensations are synthesized with subjective emotional awareness, contextual beliefs, and hedonic valuation, giving rise to conscious awareness of the "sentient self." Neuroimaging paradigms demonstrate aberrant structural volume, dysregulated functional connectivity, and blunted or hyper-responsive insular activation during gastric distension and food consumption in individuals with eating disorders, corroborating the biological reality of the psychological constructs indexed by the EDIP-Q.

Validity

The EDIP-Q has undergone rigorous psychometric validation in clinical and non-clinical populations, demonstrating superior construct, convergent, discriminant, and known-groups criterion validity:

Construct and Structural Validity

Structural validity was established across two independent cohorts of adult participants in Germany ($N = 1,269$). Exploratory factor analysis (EFA) utilizing parallel analysis and scree plot inspections initially derived a cohesive four-factor model, subsequently replicated via Confirmatory Factor Analysis (CFA). All 25 items demonstrated exceptional standardized factor loadings exceeding $.50$ on their designated latent factors. Furthermore, structural equation modeling demonstrated that a second-order hierarchical model—in which the four primary factors load onto a unified general factor of *Eating Disorder-Specific Interoceptive Perception*—fits the empirical data equally well, supporting the calculation of both subscale-specific scores and a global EDIP index.

Convergent and Divergent Validity

Convergent validity was demonstrated through statistically significant, moderate-to-strong correlations with established psychodiagnostic scales measuring related psychological constructs:

• Eating Disorder Inventory-3 (EDI-3): Scores on the EDIP-Q demonstrated strong negative correlations with the Interoceptive Deficits subscale of the EDI-3 ($r = -.60$ to $-.75, p < .001$), confirming that lower EDIP-Q scores accurately reflect severe interoceptive impairments.
• Toronto Alexithymia Scale (TAS-20): The EDIP-Q Emotions and Discrimination subscales exhibited substantial negative correlations with the TAS-20 total score ($r = -.50$ to $-.68, p < .001$), particularly with the Difficulty Identifying Feelings (DIF) dimension, validating its measurement of affective processing deficits.
• Multidimensional Assessment of Interoceptive Awareness (MAIA): Moderate positive correlations ($r = .35$ to $.55, p < .001$) were observed with the Noticing, Emotional Awareness, and Body Listening dimensions of the MAIA.

Divergent validity was confirmed by demonstrating low or non-significant correlations with theoretical constructs unrelated to body perception, such as unrelated personality dimensions, confirming that the questionnaire does not simply reflect general negative affectivity or nonspecific neurotic distress.

Known-Groups Criterion Validity and Subtype Profiles

The scale effectively distinguished between clinical and non-clinical populations. Participants with verified, self-reported eating disorder diagnoses exhibited significantly lower overall EDIP-Q scores ($p < .001$) compared to matched healthy control participants without eating pathology, confirming its clinical sensitivity.

Crucially, the EDIP-Q unveiled distinct phenomenological profiles among specific diagnostic subtypes:

• Anorexia Nervosa (AN) Profile: Individuals with AN exhibited significantly lower sensitivity on the Hunger subscale compared to healthy controls and BN/BED cohorts. However, AN patients paradoxically scored higher on perceived sensitivity to Satiety (frequently interpreting minimal gastric distension as overwhelming satiety).
• Bulimia Nervosa (BN) and Binge-Eating Disorder (BED) Profile: In stark contrast to AN, individuals with BN and BED reported significantly worse deficits in recognizing Satiety signals and demonstrated the lowest scores across the entire clinical cohort on the Discrimination subscale. They experienced profound difficulty in differentiating genuine hunger from emotional distress (stress, anxiety, anger, dysphoria).
• Emotion Subscale: Deficits in recognizing pure affective feeling states were uniformly elevated across AN, BN, and BED, confirming that impaired emotional awareness represents a shared transdiagnostic feature, whereas visceral signal dysregulation diverges based on diagnostic subtype.

Reliability

The EDIP-Q exhibits excellent internal consistency, stability, and psychometric precision across both clinical cohorts and normative populations. During its validation phases (Ortmann et al., 2023, 2024), classical test theory parameters demonstrated high reliability estimates across all domains:

Internal Consistency

Estimates calculated via Cronbach's alpha ($\alpha$) and McDonald's omega ($\omega$) revealed exceptional internal coherence:

• Total EDIP-Q Scale: $\alpha = .94$ (indicating that the 25 items collectively provide an exceptionally reliable composite measure of interoceptive perception).
• Emotions Subscale (7 items): $\alpha = .92$
• Satiety Subscale (7 items): $\alpha = .93$
• Discrimination Subscale (6 items): $\alpha = .88$
• Hunger Subscale (5 items): $\alpha = .86$

Item-total correlations across all 25 items consistently ranged between $r_{it} = .48$ and $.78$, with no individual item deletion improving overall scale or subscale reliability coefficients.

Measurement Precision and Standard Error

The Standard Error of Measurement (SEM) across subscales was low, confirming that individual scores are stable and yield tight confidence intervals around estimated true scores. In both non-clinical community samples and eating disorder cohorts, the scale maintained its high internal consistency without attenuation, confirming that the instrument retains its psychometric reliability across the entire continuum of symptom severity.

Factor Analysis

The dimensional architecture of the EDIP-Q was determined using a split-sample structural equation modeling approach combining Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA):

Exploratory Factor Analysis (EFA)

In Sample 1, an initial pool of candidate items underwent comprehensive exploratory factor analysis. Both Horn's parallel analysis and Cattell's scree test unambiguously indicated a robust four-factor solution. A subsequent Principal Component Analysis (PCA) with oblimin rotation confirmed this four-factor structure, which accounted for an impressive 65.34% of the total cumulative variance. Items exhibiting cross-loadings greater than $.30$ or primary loadings below $.50$ were systematically eliminated during the developmental refinement phase, resulting in the optimized 25-item instrument.

Confirmatory Factor Analysis (CFA)

In Sample 2 ($N = 1,269$ total across the validation stages), Confirmatory Factor Analysis was implemented using robust maximum likelihood estimation to test the structural integrity of the 25-item four-factor model. The four-factor correlated model demonstrated an excellent fit to the empirical data across standard goodness-of-fit indices:

• Comparative Fit Index (CFI): $0.95$ (exceeding the standard $.95$ cutoff for excellent fit)
• Tucker-Lewis Index (TLI): $0.94$ (exceeding the standard $.90$ threshold)
• Root Mean Square Error of Approximation (RMSEA): $0.052$ (with a narrow $90%$ confidence interval of $[0.049, 0.056]$, well below the $.06$ stringent threshold)
• Standardized Root Mean Square Residual (SRMR): $0.047$ (well below the $.08$ cutoff)
• Model Chi-Square: $\chi^2(269) = 978.122, p < .001$

All 25 items displayed strong, statistically significant factor loadings on their intended latent dimensions, with standardized loadings ranging from $.52$ to $.89$ ($p < .001$). The first-order factors were significantly interrelated, with inter-factor correlations ranging between $r = .45$ and $r = .64$ ($p < .001$), supporting the presence of a broader overarching interoceptive construct.

Second-Order Hierarchical Model

Due to the substantial shared variance among first-order dimensions, the authors specified a second-order model in which the four latent variables (Hunger, Satiety, Emotion, Discrimination) loaded onto a single higher-order latent factor representing Global Eating Disorder-Specific Interoceptive Perception. This second-order model also exhibited an exceptional fit to the data:

• CFI: $0.95$
• TLI: $0.94$
• SRMR: $0.050$
• RMSEA: $0.053$ ($90%$ CI $[0.050, 0.057]$)
• Model Chi-Square: $\chi^2(271) = 922.17, p < .001$

These structural findings provide psychometric support for two scoring procedures: clinicians and researchers may compute individual scores for each of the four separate subscales to identify specific phenotypic deficits, or calculate an aggregate composite score for overall interoceptive capacity.

Instrument / Measurement Tool

• Test Type: Standardized self-report inventory / psychodiagnostic clinical rating scale.
• Target Population: Adults (aged 18 years and older); validated in community, university, and clinical psychiatric samples. Applicable across all genders.
• Administration Format: Self-administered; available in digital/electronic formats (web, tablet, clinical computerized systems) as well as traditional pen-and-paper questionnaires.
• Completion Time: Approximately 5 to 8 minutes.
• Total Item Count: 25 items.
• Response Scale: 7-point Likert-type agreement scale ranging from:
  • 1 = Does not apply at all
  • 2 = Applies very little
  • 3 = Applies somewhat
  • 4 = Moderately applies
  • 5 = Applies considerably
  • 6 = Applies mostly
  • 7 = Applies completely
• Languages Available: Originally developed and validated in German; official English translation available and validated by the primary authors.
• Subscale Breakdown:
  • Hunger Subscale (H): 5 items (Items 1, 6, 11, 16, 25).
  • Satiety Subscale (S): 7 items (Items 4, 8, 10, 13, 17, 20, 24).
  • Emotion Subscale (E): 7 items (Items 2, 9, 12, 15, 19, 21, 23).
  • Discrimination Subscale (D): 6 items (Items 3, 5, 7, 14, 18, 22).
• Scoring and Directionality Rules:
  • The instrument incorporates both positively formulated and negatively formulated items.
  • Reverse Coding ($^R$): Exactly 13 items are negatively keyed and must be reverse scored before computing summary indices: Items 3, 5, 6, 7, 8, 11, 13, 14, 18, 22, 25 (all six Discrimination items, three Hunger items, and two Satiety items). To reverse score, compute: $\text{Recoded Score} = 8 – \text{Raw Response}$.
  • Subscale Scores: Calculated either as the sum or mean of the recoded items belonging to each respective dimension.
  • Total Score: Calculated by summing all 25 recoded items (yielding a range of 25 to 175) or computing the mean item score (range 1.00 to 7.00).
  • Clinical Interpretation: LOWER scores reflect greater interoceptive deficits, poorer signal detection, and heightened confusion between hunger and emotional states. Conversely, higher scores reflect preserved, adaptive interoceptive processing.

Permissions & Fee and Test Year

• Test Publication Year: 2023 (Initial psychometric presentation); official journal publication 2024.
• Copyright & Permissions: The scale was developed by Julie Ortmann and colleagues. It is published in the American Psychological Association (APA) journal Psychological Assessment. The questionnaire items may be used freely by clinicians, academics, and researchers for non-commercial research, academic inquiry, and educational teaching purposes.
• Commercial Use: Prohibited without formal written authorization from the primary authors and copyright holders.
• Fee: There is no fee required for academic, clinical non-profit, or research usage.

References

• Bruch, H. (1962). Perceptual and conceptual disturbances in anorexia nervosa. Psychosomatic Medicine, 24(2), 187–194. https://doi.org/10.1097/00006842-196203000-00009
• Bruch, H. (1973). Eating disorders: Obesity, anorexia nervosa, and the person within. Basic Books.
• Craig, A. D. (2002). How do you feel? Interoception: The sense of the physiological condition of the body. Nature Reviews Neuroscience, 3(8), 655–666. https://doi.org/10.1038/nrn894
• Critchley, H. D., Wiens, S., Rotshtein, P., Ohman, A., & Dolan, R. J. (2004). Neural systems supporting interoceptive awareness. Nature Neuroscience, 7(2), 189–195. https://doi.org/10.1038/nn1176
• Fairburn, C. G. (2008). Cognitive behavior therapy and eating disorders. Guilford Press.
• Garner, D. M., Olmstead, M. P., & Polivy, J. (1983). Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders, 2(2), 15–34. https://doi.org/10.1016/j.bpsc.2017.12.004
• Mehling, W. E., Price, C., Daubenmier, J. J., Acree, M., Bartmess, E., & Stewart, A. L. (2012). The Multidimensional Assessment of Interoceptive Awareness (MAIA). PLOS ONE, 7(11), e48230. https://doi.org/10.1371/journal.pone.0048230
• Ortmann, J., Lutz, A. P. C., Rose, G., Happ, C., Vögele, C., Schulz, A., & van Dyck, Z. (2024). Development and initial validation of a self-report measure to assess eating disorder-specific interoceptive perception. Psychological Assessment, 36(2), 162–174. https://doi.org/10.1037/pas0001283
• Porges, S. W. (1993). Body Perception Questionnaire (BPQ). Laboratory of Developmental Assessment, University of Maryland.
• Seth, A. K. (2013). Interoceptive inference, emotion, and the embodied self. Trends in Cognitive Sciences, 17(11), 565–573. https://doi.org/10.1016/j.tics.2013.09.007
• Tylka, T. L. (2006). Development and psychometric evaluation of a measure of intuitive eating. Journal of Counseling Psychology, 53(2), 226–240. https://doi.org/10.1037/0022-0167.53.2.226

Items of the Scale