Publications 

Stafford, L., Munns, L., Crossland, A. E., Kirk, E., & Preston, C. E. (2024). Bonding with bump: Interoceptive sensibility moderates the relationship between pregnancy body satisfaction and antenatal attachment. Midwifery, 131, 103940.  https://www.sciencedirect.com/science/article/pii/S026661382400024X

Research summary: 

Background and problem: The body undergoes transformative changes during pregnancy. This includes how the body looks, such as the development of a baby bump. These changes can make mothers feel connected to their baby, for example by touching or stroking the bump. As a result, some research has linked an expectant mother’s satisfaction with their pregnant body to the bond they have with their unborn baby. Said bond is important for postnatal mental health. However, other research has failed to find an association between satisfaction with bodily appearance and bond with the unborn baby. This might be because they did not consider the role of the internal body in this relationship. The internal body may be important in this relationship as like the external appearance, it also undergoes massive changes and can be used to connect mother and baby. For example, heart rate increases with pregnancy and kicks and movements from the fetus might make the mother feel closer to their baby. 

Aim: Therefore, in order to gain a more holistic understanding of the role of the pregnancy body experience to antenatal bonding, this research considered the relative contribution of the external and internal body. 

Methods: Survey data from 159 pregnant participants. 

Findings: Increased body dissatisfaction was associated with decreased antenatal attachment. However, this relationship was changed by how much they worried about signals from the internal body. For high levels of worry about internal signals, greater dissatisfaction with pregnant bodily appearance was not associated with a weaker bond. For low levels of worry about internal signals, greater dissatisfaction with pregnancy bodily appearance was associated with a weaker bond. 

Discussion: For expectant mothers who are unhappy with their bodily appearance, high levels of worry about internal signals may be protective against a weak bond with their unborn baby. This might be because worrying about internal bodily signals removes their attention from dislike of their body. 

Conclusion: Dislike of the maternal external body can be detrimental to antenatal bonds, but the internal body may be a decisive factor in determining how detrimental this is. 

Stafford, L., Preston, C., & Pike, A. C. (2024). Participant Use of Artificial Intelligence in Online Focus Groups: An Experiential Account. International Journal of Qualitative Methods, 23. https://doi.org/10.1177/16094069241286417

Research summary: 

Background and problem: Adapting methods like focus groups and interviews to take place online via video conferencing software is beneficial for several reasons. Reducing the need for participants to travel increases accessibility and geographical reach. However, one problem with this is that participants may decide to add to the discussion via the chat box rather than their microphone. This is something that we experienced when running online focus groups, despite asking that they have them on in the email invite. There are valid reasons to use the chat box over microphone, like a fear of being overheard by family/housemates. However, another possibility is that they were using the chat box to provide ChatGPT generated answers. 

Aim:  To provide systematic evidence that participants may be pasting ChatGPT answers into the chat box. 

Methods: 12 online focus groups with 78 people. 42 participants provided chat box answers. 

Findings: 9 participants were suspected of providing ChatGPT generated responses.

Discussion: Whilst ChatGPT may be a useful tool to help participants to verbalise their answers, a big problem with this is that ‘new’ data is not created. This is because ChatGPT responds to prompts using data that already exists, surely failing to provide any new insights into a research topic. There are ways researchers could prevent and identify ChatGPT generated responses, but the preventative measures limit accessibility (e.g. for people with speech difficulties) and indicators can run the risk that researchers ‘cherry pick’ their data. Instead, researchers may have to consider consulting with a member of a focus group who answered using the microphone to ensure that, even with the inclusion of the suspected ChatGPT generated responses in the data set, the analysis captures how a phenomenon is subjectively experienced. Alternatively, researchers could compare the findings from the data set that includes ChatGPT generated responses to the findings from a data set that excludes them and then check for substantial differences. Substantial differences can then provide reason for excluding the suspected ChatGPT responses. 

Conclusion: Participants may be using the chat box in focus groups so they can answer questions with ChatGPT generated responses. Researchers may have to conduct an extra step of qualitative analysis to protect the integrity of their data.  

Stafford, L., Pike, A. C., & Preston, C. E. (2025). " I feel full with shame": A qualitative perspective on gastric interoceptive sensibility. Appetite, 108299. https://www.sciencedirect.com/science/article/abs/pii/S0195666325004520, freely available at: https://eprints.whiterose.ac.uk/id/eprint/231904/ 

Background: “Am I hungry? Did I overeat at lunch?” Gastric interoception - the sensing, interpretation, and regulation of signals from the gastrointestinal system - is central to daily behavior and homeostasis. Dysfunctional gastric interoception has been proposed as a maintenance factor in both eating disorders and gastrointestinal symptoms. However, no qualitative research has explored how individuals across these groups, and the general population, subjectively experience gastrointestinal signals, known as gastric interoceptive sensibility. This study aimed to investigate how gastric sensations are sensed, interpreted, and regulated among individuals with eating disorders, gastric disorders, and those without such diagnoses, focusing on identifying shared experiences.

Methods: Fifteen semi-structured focus groups (n = 96) were conducted. Transcripts underwent hybrid deductive and inductive thematic analysis.

Findings:  Four key themes were identified. In “Sensations in the Interoceptive Body”, participants described hunger and fullness as physically aversive or reported an absence of cues related to satiation. “Perceiving the Interoceptive Body” captured the noticing, interpreting, attending to, and reacting to sensations of hunger, satiation, and fullness. In “Affective Experiences of the Interoceptive Body”, participants discussed how these sensations influenced emotional states positively, negatively, or not at all. “Responding to the Interoceptive Body” described participants strategies relating to relief-seeking, compensation, acceptance, distraction, and body checking in response to gastric sensations. 

Discussion: These findings shed light on the nuanced components of gastric interoceptive sensibility and suggest that individuals vary in how they experience and manage gastric signals. This work may inform interoceptive exposure therapies targeting maladaptive interpretations and regulation strategies in eating and gastrointestinal symptoms.

Stafford, L., Crossland, A., Pike, A. C., & Preston, C. (2026, May 2). Afraid of Hunger, Ashamed of Fullness: The Development and Validation of the Negative Appraisal of Gastric Sensations (NAGS) Scale and its association with Disordered Eating. Retrieved from osf.io/preprints/psyarxiv/jufgk_v1, accepted for publicaton in Psychological Assessment 

Background: Interoceptive dysfunction is central to eating disorders, and emerging evidence indicates that the maladaptive appraisal of gastrointestinal signals predicts pathology better than body-general interoception. However, existing tools do not adequately measure this construct; they conflate sensation intensity with appraisal and assess a limited range of emotional responses, likely due to little input from eating disorder populations during tool development. To address this gap, we developed and validated the Negative Appraisal of Gastric Sensations (NAGS) scale to specifically capture the affective interpretation of hunger and fullness.

Methods: Online survey (n=674). 

Findings: Exploratory factor analysis (sample 1 n=337, 80.12% Female, 75.97% White British) and confirmatory factor analysis (sample 2 n=337, 85.16% Female, 79.23% White British) supported a two-factor scale: hunger-fear and fullness-shame. The scale demonstrated robust psychometric properties, including strong internal and test-retest reliability, as well as criterion, concurrent, convergent, and discriminant validity. The subscales successfully distinguished eating disorder from gastrointestinal disorder and non-disordered controls, and this was not attributable to differential item functioning. Exploratory analyses using latent profile analysis and hierarchical regressions supported the measure’s utility as a critical interoceptive factor in eating disorders over and above existing interoceptive measures: increases in fullness-shame predicted higher overall symptom severity, body dissatisfaction, and purge-only category membership, while increases in hunger-fear specifically predicted restriction-only category membership.

Discussion: The scale could be considered a useful tool for researchers studying the somatic experience of eating disorders, but also primary care physicians for distinguishing primary gastrointestinal disorders from eating disorders presenting as gastrointestinal complaints.

Stafford, L., Pike, A. C., & Preston, C. (2026, September 11). The Impact of Gastric State on Perceptual and Attitudinal Body Image Dimensions and Their Moderation by Eating Pathology. Retrieved from osf.io/preprints/psyarxiv/bfnve_v1

Background: Body image encompasses cognitive-affective attitudes and explicit and implicit body size perceptions. Although interoception is implicated in eating pathology, evidence directly linking gastrointestinal states to body image remains limited by caloric confounds, a dearth of research using perceptual measures, and narrow assessments of interoceptive facets. 

Methods: To address these gaps, this study (n=68) manipulated gastric state using the Water Load Test-II to evaluate how gastrointestinal sensitivity, over/underconfidence in sensitivity, and appraisal are related to baseline body image and disordered eating, if gastric state can change body image, and if this is moderated by disordered eating. 

Results: At baseline, elevated hunger-fear and fullness-shame predicted heightened eating shame, dietary restriction/preoccupation, and reduced baseline body satisfaction, with fullness-shame uniquely predicting greater weight/shape concerns and restraint. Heightened gastrointestinal sensitivity predicted less certainty in judging an aperture as passable relative to body size, whereas underconfidence in gastrointestinal sensitivity predicted smaller waist size estimations and greater certainty in judging an aperture as passable relative to body size. Controlling for baseline body image, fullness significantly decreased body satisfaction, increased the width required for an aperture to be judged as passable relative to body size, and increased waist size estimations according to disordered eating severity. Eating disorder severity predicted lower certainty for an aperture as passable relative to body size. 

Conclusions: During nutritional eating disorder recovery, where postprandial fullness is unavoidable, unaddressed negative appraisal, hypervigilance, and over/underconfidence in gastric sensitivity may perpetuate body image distortions and uncertainty of body size. Eating to satiation, rather than complete fullness, may offer a behavioral goal for ongoing treatment once weight is stabilised.