Abstract
Bodily autonomy is essential to Autistic well-being. Interoception supports bodily autonomy through guiding behavior in support of homeostasis. Promoting adaptive interoceptive skills is therefore crucial to improving Autistic well-being. To date, research on interoception in Autistic people has been poorly informed by the experiences and goals of Autistic people, has had a narrow focus on comparison with non-autistic norms in search of deficits, and has had limited impact on Autistic quality of life. In this perspective piece, a team of Autistic and non-autistic researchers review findings to date and make recommendations for future research directions. We argue that participatory research is needed to comprehensively map the varied interoceptive landscape of Autistic people, ensure that new interoceptive measures are autism-relevant, and guide interventions designed to improve the interoceptive and broader well-being of Autistic people. We believe that if interoception research is conducted in partnership with the autism community, the understanding of interoceptive processing and the full range of interpretive differences that result will be impactful and informative.
Community Brief
Why is this topic important?
Interoception is our ability to notice, monitor, and understand signals from our bodies. The accounts of Autistic people emphasize the centrality of challenges with interoception to their lived experience and quality of life, including difficulties with knowing when they are thirsty, need to go to the bathroom, or are feeling unwell. These challenges likely stand in the way of many capabilities that are important for a fulfilling and independent life.
What is the purpose of this article?
The goal of this article is to review what is already known about interoception in Autistic people, discuss how the findings might relate to the everyday challenges Autistic people face, and make recommendations about how research could be better designed in the future to speak to both Autistic people’s everyday experiences of interoception and their research priorities.
What personal or professional perspectives do the authors bring to this topic?
The authors are a team of Autistic and non-autistic researchers who have the experience of conducting research on interoception, being Autistic, and supporting Autistic people to nurture their interoceptive capabilities.
What is already known about this topic?
Previous research has shown differences between Autistic and non-autistic children, adolescents, and adults on various measures of interoception, which corroborates the idea that interoception is affected in many Autistic people. The findings have not always replicated across studies, however, suggesting that there are important components of Autistic interoception that researchers have been missing. Moreover, these findings have not been very helpful in explaining the everyday challenges of Autistic people, nor how we can best support them.
What do the authors recommend?
The authors recommend moving away from studies that simply compare performance between Autistic and non-autistic people on various interoception measures, and instead adopt a participatory research approach to integrate Autistic people’s views and experiences from the get-go.
How will these recommendations help autistic adults now or in the future?
The authors hope that their recommendations for future research will ultimately result in research that better captures Autistic people’s phenomenological interoceptive experiences and leads to better supports for fostering Autistic flourishing and well-being.
Introduction
Interoception is a set of core physiological and cognitive processes that determine the condition of the body. 1 Interoception guides the sensing, interpretation, and integration of the body’s internal landscape. Relevant signals include afferent traffic from the cardiovascular, respiratory, gastrointestinal, genitourinary, nociceptive, thermoregulatory, immune, and endocrine systems, 2 although there remains debate about the exact boundaries between interoception, exteroception, and proprioception. 3 Interest in interoception has increased dramatically in recent years,2,4 and is now recognized as integral to many capabilities that promote bodily autonomy and integrity, including self-awareness,5–8 emotional experience and regulation,9–15 and activities of daily living. 16 Bodily autonomy reflects the ability to make decisions about one’s own body, life, and future, without coercion or violence. Bodily integrity is a related construct that emphasizes the importance of personal autonomy, self-ownership, and self-determination of human beings over their own bodies. Together, bodily autonomy and integrity refer to the human right that everyone should enjoy in making self-determined decisions over their own body.17,18 The bodily autonomy and integrity of Autistic people are sadly violated too frequently, in multifarious ways, through no fault of their own, as evidenced by high rates of coercive interventions,19–22 unnecessary use of physical and pharmacological restraint, 23 and sexual assault.24–26 Here we wish to focus on the subtle ways in which differences in interoceptive systems, which guide the interpretation of bodily sensations, promoting adaptive responses to changes in the internal and external environment, can also present barriers to bodily integrity and autonomy.
In recent years, interoception has become a burgeoning area of study for autism research, but the resultant findings have painted a confusing picture (see Ref. 27 for a recent meta-analysis). Much of the research to date has focused on comparing the average performance of Autistic and non-autistic participants on measures of interoception. Such research often frames a difference between these groups as de facto representing a deficit in Autistic people, even when supportive evidence, beyond the group difference, remains scarce. Due to the nature of the tasks and questionnaires used, almost all research on interoception in Autistic people to date has been conducted with speaking Autistic people without intellectual disability, limiting its generalizability. In this article, a team of Autistic and non-autistic researchers review what is currently known about interoception in Autistic people and make recommendations for future research. We argue that applying participatory research methods, which involve Autistic people and their allies as partners in the research process, 28 will improve interoception research. Manifestations of participatory practices are varied, ranging from consultatory and partnership roles to Autistic leadership, 29 but more power in the decision-making processes around research should help ensure that research is ethically guided by community values, improve the validity and accessibility of methods, and situate findings within the real-world, enhancing the translation of findings into practice.30–33
Efforts to better operationalize interoceptive measures and create a taxonomy of interoception that separates subjective from objective indices2,34,35 offer a structure to organize the mixed findings observed with Autistic people. On tasks where participants are asked to count their heartbeats, Autistic participants often,34,36–38 but not always,39–41 detect fewer of their heartbeats than non-autistic participants. On tasks where participants are asked to report whether external tones are synchronous or asynchronous with their heartbeats, Autistic participants generally respond similarly to non-autistic participants.34,38 On questionnaire measures that inquire how often the respondent pays attention to various interoceptive sensations, or how confident they are in their ability to detect interoceptive sensations, Autistic participants sometimes score higher than non-autistic participants,34,38 but it depends on the framing of the questions. For example, Autistic people tend to score higher on measures that tap a hypervigilant interoceptive attentional style that is associated with anxiety, such as the Body Perception Questionnaire (BPQ 42 ), and lower on measures that emphasize a more mindful and beneficial attentional approach to interoceptive signals, such as the Multidimensional Assessment of Interoceptive Awareness (MAIA 43 ).
Previous research has struggled to account for discrepancies between these different measures of interoception in Autistic people. Some authors have suggested a mismatch between expectations and reality as a possible mechanism. 34 Bayesian accounts of autism propose that Autistic perception reflects a relative underweighting of expectations based on prior experiences, and a relative overweighting of incoming sensory information, such that the gain or “volume” on sensory channels is turned up, and difficult to contextualize based on past explanations for that sensory information.44–49 Bayesian accounts of interoceptive processing posit that altered interoceptive prediction mechanisms could cause sensory integration challenges, including for Autistic people.50–52 These accounts have informed the interpretation of discrepancies observed between dimensions of interoception in Autistic people, with lower ability to detect heartbeats, alongside high subjective awareness of interoceptive sensations described as an interoceptive trait prediction error. 34 While these accounts may have high explanatory value and are starting to inform the development of novel interventions, 53 as yet, they have had little applied value in informing clinical care and the daily lives of Autistic people.
An alternative explanation is that existing interoceptive measures lack validity for Autistic people. The BPQ and MAIA were both developed for use in non-autistic samples. There are few measures specifically developed for quantifying interoception in Autistic people, reflecting the exceedingly scant research that has directly focused on Autistic people’s phenomenological interoceptive experiences. One exception is the Interoception Sensory Questionnaire, 54 which was designed to capture the variability in Autistic interoception, and showed good reliability and convergent and divergent validity in the original validation sample. The variability of experience captured by this measure is important, as emerging qualitative work suggests that Autistic people (in this case, adolescents) experience both hyperawareness and hypoawareness of interoceptive signals. 55 For example, an Autistic person may be unaware of thirst signals but highly attuned to the growth of cells in a keloid scar. Autism also frequently co-occurs with other neurodivergent traits, including attention-deficit coordination disorder (ADHD), sensory modulation disorder, tic disorder, and developmental coordination disorder, which are also associated with variability in interoception.56–60 To date, the impact of these co-occurring traits has been poorly addressed by research.
Interoception is key to independent participation in everyday activities because when interoceptive signals are meaningful and can be trusted, the person feels a sense of safety in their body and a coherent sense of self.12,61 Despite this, the questionable validity of interoceptive measures makes it unclear how such findings could be used to support the everyday lives of Autistic people. The narratives of Autistic people reinforce the idea that they experience challenges with interoception in day-to-day life, 62 and suggest that the ways in which interoception is studied in Autistic people could be better tailored to fostering improvements in these day-to-day challenges. Understanding interoception in autism has the potential to improve Autistic well-being, but the perspectives of Autistic people should be centered to inform research design, interpretation, and dissemination.
Norms for measures of interoception have not been delineated, and even among non-autistic samples, there is large variability, with many non-autistic adults reporting little or no awareness of their heart beating in laboratory settings. 63 What then defines optimal interoception, and correspondingly, suboptimal interoception? Are there benchmarks that can be applied across a population, or might one person’s ideal interoception be problematic for another person? This lack of consensus makes interpreting the impact of interoceptive differences between Autistic and non-autistic people difficult and suggests we may have been too rash to pathologize differences as deficits. In addition, a better understanding and appreciation of the practical needs of Autistic people are warranted. In terms of supports, sensing heart rate can be easily and accurately supported using wearable devices, but knowing when you are likely to need to use the toilet or are thirsty, while likely highly useful, is not currently well supported using available technologies.
Neglected Features of the Autistic Interoceptive Experience
Research on Autistic interoception should be informed by the experiences, needs, and goals of Autistic people. Researchers’ focus on studying heartbeat detection has likely obscured the importance of other organ systems (e.g., respiration and gastrointestinal tract64,65), physiological pathways (e.g., immune and endocrine systems66,67), and processes (e.g., pain and menstruation 68 ). Gastrointestinal issues are particularly common in Autistic people, 69 and include chronic constipation, diarrhea, and abdominal pain.70,71 Connective tissue disorders are also more common than in non-autistic people.72–74 Such challenges may reflect more widespread or impactful differences in interoception than in the cardiovascular system. When faced with chronic conditions that cause pain and discomfort, an adaptive response may be to down-weight interoceptive signals from the gastrointestinal system. This may have widespread impacts on everyday functioning. Toileting, maintaining adequate levels of hydration, and getting a balanced diet are all life skills with which Autistic youth experience greater challenges than non-autistic children.75–78
Some authors have argued that alterations in interoception do not represent fundamental features of autism, and findings instead reflect high rates of co-occurring alexithymia,41,79 which is experienced by approximately 50% of the Autistic population.80,81 This subclinical personality construct describes a difficulty with labeling and expressing emotions, and a tendency for reduced imaginative thinking. 82 There is both evidence for and against this conclusion. When alexithymia, autistic traits, and interoception were examined together in general population adults, heartbeat detection scores correlated negatively with alexithymia, but not with Autistic traits. 41 Alexithymia has also been found to partially mediate the relationship between vigilance of interoceptive signals and anxiety, 83 at least in general population adults, suggesting important contributions to well-being. The association between atypical interoception and alexithymia does not always replicate, however.39,84 Many of the characteristics of alexithymia also appear to overlap with core Autistic features, at least as defined by DSM-5. While the presence of alexithymia is not required for a diagnosis of autism, and many Autistic people do not have alexithymia, difficulties with sharing emotions and imaginative play are listed in the current diagnostic criteria. 85 It is also possible that alexithymia, as a difficulty verbalizing internal states, explains much of the variance on standard interoceptive tasks, as the tasks themselves require participants to verbalize their internal (interoceptive) states. Whether this means that being Autistic is not associated with differences in interoception, however, remains to be conclusively demonstrated. Examining the distinctness of these constructs, and how they relate to interoception, is a key question for future research.
Age-related changes in interoception are also ripe for investigation. In the neurotypical population, interoception gradually declines across the adult lifespan, particularly when measured using heartbeat perception tasks,86,87 and may be partially mediated by increasing body weight with age. 88 In Autistic children, the effects of age on interoception may be additionally moderated by cognitive abilities. 89 Interoception in Autistic adults may be more similar to neurotypical people than in Autistic children, 37 suggesting that Autistic people may develop coping strategies as they age. Perhaps a lifetime of experiences may allow you to become an expert in your own body, but the impact on Autistic adults’ well-being is currently unknown.
Interoceptive Implications for Autistic Well-Being
Interoception is undoubtedly crucial to bodily autonomy, emotion regulation, and physical and mental health. A recent review applied Nussbaum’s capabilities approach17,18 to Autistic well-being, and highlighted the importance of bodily integrity to Autistic flourishing, including using one’s senses and feeling a range of emotions. 90 Interoception touches many of these domains, illustrating the far-reaching effects of interoception on Autistic people’s lives.
Challenges with emotion dysregulation, which, when unmitigated, can lead to meltdowns, shutdowns, and self-injurious behavior, are more common in Autistic people than non-autistic people.91–93 Emotion regulation is a multicomponent process that can involve up- or downregulating the thoughts, behaviors, or autonomic nervous system activity associated with an emotion in service of current goals, using both deliberate control and more implicit efforts. 94 Interoception is likely implicated in these emotion regulation processes at several levels. First, attention must be deployed to detect salient alterations in allostatic norms. Determinations about what is salient may show inter- and intra-individual variability. Second, interoceptive information must be integrated with environmental information and interpreted. Third, a coordinated and appropriate reaction across autonomic, endocrine, and behavioral response systems must be generated. Fourth, the ongoing impact of these changes must be monitored. Autistic people are purported to have differences across many of these regulatory functions.40,95–98 Pain and the associated consequences of poorly maintained physical needs also curtail regulation abilities. A person trying to escape physical discomfort might move around and struggle to sit still for extended periods or may self-injure in a bid for relief or to signal that their body is hurting. The contributions of physical movement to regulation were recently illustrated by a study that demonstrated an association between exteroceptive uncertainty and emotion dysregulation (measured as dissociative symptoms) in those with Autistic and ADHD traits. 99 Bodily processes should therefore be increasingly recognized as important contributors to one’s ability to self-regulate.
Interoception has a large impact on bodily health, and in particular the conversion of signs of ill health into adequate medical care. This is particularly pertinent given the elevated prevalence rates of many medical conditions in the Autistic population.100,101 Autistic people experience reduced life expectancy102–104 and increased mortality across most diagnostic categories. 105 Autistic people have therefore raised physical health as a priority area for research. 106 Autistic people across the lifespan experience higher rates of chronic health conditions, 107 the management of which often requires the monitoring of interoceptive sensations. Alongside under-acknowledgment of clinical signs by professionals, who may find it challenging to understand and communicate with Autistic people,108,109 Autistic people may struggle to identify sensations as potential health concerns and communicate this to others. They may face challenges with being taken seriously because they may report pain or other sensations differently.110,111 They may struggle to localize the source of their pain or underreport pain due to challenges with translating internal sensations into numerical report scales. Autistic adults report frequent doubts about knowing whether symptoms warrant a doctor’s visit and difficulties communicating their symptoms to their doctor. 112 For example, Autistic women report menopause symptoms to their doctors half as frequently as women in the general population (4% vs. 8% 113 ) despite Autistic women listing menopause as a significant challenge,114,115 highlighting a lack of health care support with this transition.116,117 Tools and specialized training that help health care workers communicate with their Autistic patients are therefore needed to address this barrier to accessing timely and effective care.
Interoception is also a key contributor to sexuality, which contributes to bodily autonomy and self-identity. The World Health Organization 118 recognizes sexuality as an integral aspect of healthy human development, encompassing sex, intimacy, and sexual identity. Sexuality is associated with relationship quality and psychological well-being in neurotypical and Autistic adults.119,120 Although Autistic people experience the same sexual needs and desires as non-autistic people,121–123 research has suggested that Autistic adolescents and adults have lower sexual well-being 124 (see also Ref. 125), greater loneliness,126–129 and fewer intimate relationships.130–134 While these concerns are often attributed to social challenges, interoceptive contributions should also be considered. Lower interoceptive listening, noticing, and trust scores on the MAIA are associated with lower concordance between self-reported sexual arousal and genital response in women with sexual interest/arousal disorder. 135 Interoceptive trust has also been linked to orgasm satisfaction in non-Autistic women. 136 Associations between sexuality and interoception are yet to be systematically examined in Autistic people, but in qualitative work, some Autistic adolescents and adults describe difficulties identifying sensations associated with sexual desire, genital hyposensitivity, low or absent sexual desire, and difficulty reaching orgasm, 121 which may relate to differences in the interoceptive processing of these signals.
Monotropism: A Promising Yet Underexplored Explanation for Autistic Patterns of Interoception
We suggest that the concept of Autistic monotropism may help us understand patterns of interoception in Autistic people. Monotropism was developed as a generalized theory of autism by Autistic advocates and researchers and describes a narrowing of attentional focus such that extraneous distractors do not penetrate awareness.137,138 Monotropism may both result from (e.g., as a coping mechanism for painful or confusing bodily sensations) and maintain (e.g., focusing singularly on one environmental feature at the cost of bodily sensations) interoceptive differences. Emerging research suggests that monotropism is often a feature of the Autistic experience. 139 Anecdotal evidence suggests that the theory of monotropism is highly popular in the Autistic community, 140 and is often invoked in explanations of, and practical strategies for, many aspects of Autistic people’s real-world experiences. 141
While the concept of monotropism has not previously been applied to interoception, it offers a promising account of the everyday struggles with interoception that Autistic people describe. Autistic people seem to struggle the most with detecting interoceptive signals when they are occupied with other engaging tasks, but may have no problems in detecting interoceptive information when it is attended to, such as in a relatively controlled laboratory setting, if other distractors are removed. 138 Anatomical features of interoceptive pathways may make affective states relatively difficult to locate in time and space and more likely than other sensory experiences to recede into the background of conscious awareness. 142 Forgetting personal needs is an integral feature of attention narrowing, 143 which is emphasized by the accounts of Autistic people, who describe ignoring interoceptive signals when engaged in a task, “Once I get started, stopping is not an option…you just keep going. Not eating, not drinking, trying to avoid going to the toilet…”, or having no awareness of interoceptive signals altogether, “absolutely no recognition of my body’s signals—I won’t feel hungry, I won’t feel thirsty, I won’t need to go to the toilet. Or I probably do but I don’t notice it”, [I am] “not sure I’m even aware I have a body at such times” and go “for days without sleep”, as well as the negative consequences of not attending to interoceptive needs, “focus eventually breaks and I then realise that I’ve been sitting in the same position for six hours and I’m starving, and I desperately need to pee and every muscle in my body hurts.” 144 These quotes, from a study of Autistic flow states, suggest that Autistic people may not necessarily experience altered conveyance of afferent interoceptive signals to the central nervous system, but may instead have a reduced propensity to access, notice, or engage with interoceptive information.
Much of the interoception research to date has considered performance as a relatively stable, trait-like measure. State-based fluctuations in interoception, however, may be crucial to understanding the Autistic experience. Monotropism would predict that, depending on the task at hand, Autistic people may switch from heightened to reduced interoceptive awareness. Similar intraindividual fluctuations have been observed for exteroceptive sensing in Autistic people, whereby someone might experience hypersensitivity to a sensation at one moment, and hyposensitivity at another. 145 Greater state-based fluctuations could account for the mixed findings in the literature to date, as using so-called resting or baseline conditions does not necessarily ensure that all participants are in a similar exteroceptive and interoceptive attentional state. Future research could compare the effect of different experimental conditions (e.g., monotropic flow vs. mindful interoceptive focus) on the ability to notice interoceptive perturbations.
Opportunities for Support
Many existing therapeutic supports for Autistic people are primarily designed to reduce observable behaviors attributed to autism. This approach may have a negligible—or even an adverse—impact on the well-being of Autistic people. When designing supports targeting interoception, goals should be defined by the Autistic community. One particularly salient target is anxiety, which is often reported by Autistic people to have a greater impact on their quality of life than autistic features. 146 Context and individual factors should be considered as there is unlikely to be a one-size-fits-all approach to support that works for all Autistic people.
Noticing one’s internal landscape is the first step in a series of actions we must all master. Detecting deviations from homeostasis and taking action to address these deviations, before they become urgent, is a fundamental aspect of being able to self-manage, supporting bodily autonomy. Many strategies suggested to Autistic people who struggle with interoception do not address interoception directly, relying on external supports. An alternative strategy would be to leverage more pertinent signals about the internal bodily state to gradually scaffold the developing ability to detect less apparent signals. See Box 1 for a specific example for how this might be achieved in the context of thirst.
Currently, Autistic people who struggle to monitor their thirst are instructed to drink at regular intervals, according to a clock or timer, to prevent dehydration. While ensuring that they take in sufficient fluids, this strategy does little to foster improvements in interoception. An alternative strategy that supports emerging interoceptive skills would foster associations between interoceptive signals and their meaning, over time. In the case of thirst, we might instruct the person to monitor their urine color or smell and pay attention to how they feel in their body (e.g., dry mouth, lips, or throat; headache; dizziness; sluggishness). Over time, the goal is that one starts to associate darker urine color or stronger urine smell with a greater preponderance of sensations or more intense sensations. Ultimately, subtle sensations may encourage drinking without the need to monitor urine directly.
Recommendations
The above overview of the literature underscores the need to integrate Autistic people’s lived experiences into research on Autistic interoception. As it stands, there is currently insufficient information to motivate the goal of fostering neurotypical “norms” of interoception in Autistic people, and research efforts should move away from merely comparing Autistic people and non-autistic people’s performance on various measures at a single time point. Instead, investigating how and why interoception fluctuates in Autistic people will likely be crucial for real-world applications of interoceptive supports. We suggest that participatory research methods should be used to inform the next wave of autism interoception research. This might involve the implementation of community-led priority setting exercises for interoception studies, co-produced measures that emphasize the use of interoceptive signals in everyday life, and consultation on how to disseminate findings such that Autistic people can best benefit from the knowledge that is generated. We make some recommendations below but emphasize that continued partnership with Autistic people and their allies is essential for steering the development of research questions, appropriate methods, and the interpretation of findings.
In terms of therapeutic interventions, interpretation of interoceptive sensations, beyond mere detection, is likely to be critical to supporting bodily autonomy. Observations of elevated attention to interoceptive sensations, including cardiac signals, in other highly anxious populations,147–149 cautions against indiscriminately seeking to boost cardiac awareness without concomitant training in interpretation. Increased ability to detect interoceptive sensations, alongside continuing challenges in interpreting and effectively responding to interoceptive sensations, may have an adverse impact on well-being. Therapies should therefore include training in pragmatic skills that guide interpreting, contextualizing, and responding to interoceptive sensations.
We outline six particularly salient questions for future research below:
How can measures of interoception better reflect the everyday struggles with interoception that Autistic people experience, capturing intra-individual fluctuations and the role of context? How can interoception research be more inclusive of Autistic people with minimal or no spoken language and/or co-occurring intellectual disability? How do the frequent gastrointestinal challenges that Autistic people experience impact the development of interoception and self-care capabilities? What is the relationship between alexithymia and interoception in Autistic people and how does alexithymia contribute to Autistic people’s everyday challenges in identifying and communicating their internal states? Do difficulties with interoception represent causal and/or maintaining factors in the poor bodily health of the Autistic population? What is the developmental trajectory of interoception in Autistic people, and how does this relate to Autistic well-being while navigating the changing demands of adulthood and later life?
Conclusions
Interoception is a fundamental life skill that offers a path to personal autonomy. Research on interoception in Autistic people would benefit from a reduced focus on comparison with poorly defined non-autistic benchmarks and greater focus on how interoception impacts bodily autonomy, quality of life, and the ability to participate in the community. Participatory practices can help guide these research efforts, ensuring that the needs and goals of Autistic people remain at the forefront.
Footnotes
Acknowledgments
The authors are grateful to the Autistic people and their families they have worked with, who have informed these perspectives. They are also grateful to Professor Sarah Garfinkel for comments provided on an earlier version of this article.
Author Disclosure Statement
The authors have no competing or conflicting interests to disclose.
Funding Information
The authors have no funding sources to acknowledge.
Authorship Confirmation Statement
E.R.P., W.B.L., E.G., and E.P. collaboratively designed this perspective piece together. E.R.P. drafted the article. W.B.L., E.G., and E.P. provided revisions. All authors approved the final version of the article. This article has been given solely to this journal and is not published, in press, or submitted elsewhere.
