Diminished Interoceptive Awareness in Attention-Deficit/Hyperactivity Disorder: A Systematic Review

Abstract

Background:

Interoception is the perception of one’s internal physiological state. Altered interoception may play a role in the pathogenesis of attention-deficit/hyperactivity disorder (ADHD). We reviewed the literature on interoception and ADHD symptoms (inattention, hyperactivity/impulsivity, executive function, emotional dysregulation) in individuals with and without the disorder.

Methods:

Studies included assessments of interoception and ADHD symptoms in 1) individuals with/without diagnoses of ADHD or 2) the general population. Interoception assessments included objective measures (heartbeat tracking test) and self-report questionnaires. The literature search included the PubMed/MEDLINE, PsychINFO, and Scopus databases. A protocol was prospectively registered in the PROSPERO database (CRD #42022351726); the PRISMA guidelines directed reporting. The Newcastle-Ottawa scale was adapted to evaluate study quality.

Results:

Of 636 articles identified, 17 articles (based on 18 studies) met inclusion criteria. Five studies compared interoception in participants with/without ADHD, 3 finding that interoception was reduced in participants with ADHD. One study compared interoception between presentations of ADHD, finding no differences. Twelve studies assessed the association of ADHD symptoms and interoception in the general population. Overall, results suggested interoception is reduced in individuals who reported higher symptoms of inattention, hyperactivity, impulsivity, and emotional dysregulation. Studies were of moderate quality; issues included small sample sizes and inadequate reporting.

Conclusions:

Individuals with ADHD may have decreased interoception compared to individuals without. Interoception is negatively associated with ADHD symptoms in the general population. Interventions focused on improving interoceptive abilities may provide an avenue for ADHD treatment. Future work should consider the role of culture in interoception.

INTRODUCTION

Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders of childhood.¹ Hallmark symptoms include dysregulated behavior and attention, i.e., hyperactivity, impulsivity, impaired focus, and decreased memory.¹ Emotional dysregulation (irritability, anger, and aggression) also occurs in up to 70% of cases.^2,3 ADHD has been described as a disorder of self-regulation, as dysregulation in general appears to drive symptoms.⁴ In more than half of children, symptoms persist to adulthood,⁵ increasing the risk of other self-regulatory issues such as gambling, speeding tickets, and substance use disorders.^6–8

Effective self-regulation requires self-monitoring, the awareness and observation of internal physiological and emotional states. The ability to monitor one’s own internal state is informed by interoception, conceptualized as the awareness and interpretation of sensory signals originating inside the body (such as hunger, thirst, heart rate, and breathing frequency),⁹ as well as skin-mediated signals, including the perception of temperature, touch, and pain.^9–11 Further, interoception is also thought to involve feedback and feedforward loops from the brain to the internal organs which contribute to the creation of feelings and emotional states.^12–14 Interoception is considered a critically important process to a variety of fundamental human abilities, such as decision-making and emotional regulation, leading to the hypothesis that interoception is a central transdiagnostic process that causes and maintains mental disorders and physical diseases.¹³

At the conscious level, several dimensions of interoception have been proposed. One of the most frequently discussed frameworks is Garfinkel’s three-dimension model of interoception.¹⁵ This model is composed of: (a) interoceptive accuracy, (b) interoceptive sensibility, and (c) interoceptive awareness.¹⁰ Interoceptive accuracy describes the ability to correctly detect internal sensory signals like heart rate and is measured with objective assessments such as an electrocardiogram (EKG; Figure 1). Interoceptive sensibility is the subjective belief in one’s own interoceptive ability to detect changes in internal states and is assessed with self-report questionnaires such as the Multidimensional Assessment of Interoceptive Awareness (MAIA) questionnaire.¹⁶ Lastly, interoceptive awareness refers to the association between objective interoceptive accuracy and subjective report, i.e., the degree to which confidence in the heartbeat task predicts performance on the heartbeat task. A high level of interoceptive awareness reflects the ability (i.e., meta-awareness) of an individual to know when he/she is making good or bad interoceptive decisions, on the level of interoceptive behavioral accuracy. Interoceptive accuracy has been the most studied dimension of interoception, but these classic interoception measures have important psychometric limitations.¹⁷

Figure 1:

Garfinkel’s three-dimension model of interoception

The conceptualization of interoception has continued to evolve. Khalsa and colleagues have proposed an 8-component model that distinguishes between interoceptive sensibility (defined as the ability to focus on interoceptive stimuli) and interoceptive self-report scales, defined as the ability to reflect and describe one’s own experience of bodily states and make then judgments about them.¹² Murphy and colleagues, in their model of interoception, distinguish between interoceptive accuracy and interoceptive attention, which refers to the degree to which interoceptive signals are the object of attention.¹⁸ Authors suggest that self-report questionnaire measures could be used to quantify both interoceptive attention (e.g., via the Body Perception Questionnaire)¹⁹ and interoceptive accuracy (e.g., via the Interoceptive Accuracy Scale).¹⁸

More recently, some authors have proposed that interoception should also include regulating signals from the brain.¹² The Embodied Predictive Interoceptive Coding (EPIC) model posits that the brain forms a prediction-sensation loop that minimizes exteroception/ interoceptive unexpectedness, contributing to homeostasis.^20–23 Authors postulate that the brain is constantly generating and updating a mental model of the internal and external environment that is used to predict afferent bodily signals that are then compared with the actual input signals from those senses. Finally, Crucianelli et al. has suggested that interoception at perceptual levels should be investigated as a set of independent processes rather than a single construct.⁹ Specifically, they suggest that the capacity to perceive different kinds of interoceptive signals (e.g., cardiac, thermosensory, nociceptive) may vary within an individual and thus distinct interoceptive sub-modalities may need to be probed separately.

Recently, interest in the role of interoception in neurological and psychological conditions has grown. Interoception has been investigated extensively in autism spectrum disorder (ASD), a neurodevelopmental condition that has some shared symptoms with ADHD and co-occurs in 12% of individuals with ADHD.²⁴ Systematic reviews have reported reduced interoceptive accuracy in ASD.^25,26 An interoception-focused intervention, designed to increase interoceptive accuracy through real-time feedback during heartbeat detection tasks, was effective at reducing anxiety in autistic adults.²⁷ The proposed mechanism of this change was an increase in the ability to anticipate, interpret, and modulate internal states, leading to increased self-regulation.²⁷ Identification of internal signals is required for self-management and the regulation of affect and behavior. Given that ADHD is characterized by dysregulation across multiple domains including self-regulation, reduced interoceptive accuracy may contribute to these self-regulation deficits that drive symptoms. If so, interoception may be a potential treatment target.

To our knowledge, no systematic review exists on interoception in ADHD. Therefore, using Garfinkel’s model of interoception, we performed a systematic review to address the following research questions:

1. Are interoceptive abilities (accuracy, sensibility, and awareness) decreased in individuals with ADHD compared to individuals without?

2. Are measures of interoceptive accuracy, sensibility, and awareness associated with symptoms of ADHD in individuals who do not have a diagnosis of ADHD?

Our objective was to systematically evaluate the research investigating interoception in ADHD.

METHODS

This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.²⁸ The protocol for the review was indexed in the PROSPERO database before the review began (CRD #42022351726), and deviations from the protocol were recorded (Supplemental Table S1).

Search strategy

The literature search included the PubMed/MEDLINE, PsychINFO, CINAHL, and Scopus databases. The PubMed search strategy included terms and MeSH headings related to attention, hyperactivity, impulsivity, emotion regulation, ADHD, and interoception, and was modified for use in the other databases (Appendix 1). The reference lists of included studies were searched for other relevant studies. ClinicalTrials.gov was searched for relevant research in progress, the ProQuest database was searched for theses/dissertations, and the Cochrane Database of Systematic Reviews and the PROSPERO database were searched for relevant systematic reviews and meta-analyses. The first ten pages of Google Scholar were reviewed to capture any other relevant literature. Searches were conducted through March 1, 2024, and no limits on publication dates were applied to the literature search. Only studies available in the English language were included due to lack of resources for translation.

Study selection

Two types of studies were considered. The key inclusion criteria for this review were as follows:

Inclusion criteria 1

Study population:

First, studies in individuals who reported a previous diagnosis of ADHD, with or without a comparison group of individuals without ADHD, were included. Studies with participants reporting co-occurring neurodevelopmental disorders, such as ASD, in addition to ADHD were included if ADHD was one of the primary diagnoses for study inclusion. Because we anticipated a small number of studies, the inclusion criteria were intentionally broad.

Study type:

Studies must have compared 1) mean (or other summary measure) interoception scores for the ADHD versus non-ADHD groups or 2) mean interoception scores between different presentations of ADHD.

Inclusion criteria 2

Study population:

Second, studies in community-dwelling adults or pediatric participants without a diagnosis of ADHD who were assessed for symptoms of ADHD using a self- or collateral-report (parent, teacher, or clinician) symptom questionnaire were included. Symptoms of ADHD could include core symptoms (inattention, hyperactivity, impulsivity), or commonly co-occurring symptoms such as executive dysfunction or emotional dysregulation.

Study type:

Studies could enroll participants with diagnoses of ADHD but could not compare mean interoception scores between ADHD and non-ADHD groups.

We considered cohort, case-control, cross-sectional, and pilot studies, plus randomized controlled trials as candidates for inclusion. Case papers were considered if they otherwise met the inclusion criteria, and relevant systematic review and meta-analyses were examined for potentially eligible primary studies. Qualitative studies, conference abstracts, editorials, and opinion papers were excluded. Theses /dissertations which were not published in peer-reviewed journals were considered if they otherwise met inclusion criteria. No age limits were applied to participants, but studies on pediatric/adult populations were examined separately.

Study inclusion and data collection

After the literature search, all identified manuscripts were uploaded into the Rayyan software and duplicates removed.²⁹ Two to three authors independently screened studies in a blinded fashion (AB, LL, NR). Manuscripts that seemed to meet inclusion criteria for the review were retrieved in full, and their complete text assessed by 2+ independent reviewers (AB, LL, NR). Any disagreements were resolved through discussion or by consulting a third author (JJ, DC). Reasons for excluding manuscripts were recorded for reporting in a PRISMA flowchart.

Data were extracted from the included studies by 2+ authors independently using a data extraction tool developed in Microsoft Excel (AB, LL, NR; details in Appendix 2). Sex and/or gender of participants were noted using the terminology in the original paper (sex, gender) per the Sex and Gender Equity in Research (SAGER) guidelines.³⁰ Any disagreements between authors were resolved through discussion or by consulting a third author (JJ). If studies had missing data or unclear results, up to three attempts were made to contact study authors to request clarification.

Types of measures

Requisite outcome measures included ADHD symptom questionnaires and measures of interoception, either self-report or objective.

ADHD:

Eligible outcome measures were parent- or self-report questionnaires on ADHD symptoms, including, but not limited to, the ADHD-Rating Scale IV or the Conner’s Parent or Teacher Rating Scale in any of its forms (short form, revised form, etc.) or computerized tasks of attention, impulsivity, executive function, etc.

Interoception:

Three dimensions of interoception were considered for this review, based on Garfinkel’s model of interoception:¹⁵ interoceptive accuracy, interoceptive sensibility, and interoceptive awareness. An outcome measure that assessed any of these three dimensions was eligible for inclusion.

Assessment of study quality

A study quality score for the studies in individuals with ADHD was calculated from the Newcastle-Ottawa Scale for Non-Randomized Studies, modified for cross-sectional studies.^31,32 A version of the scale used in studies with autistic individuals^* was used as a guide and modified for use in ADHD (Table S2).³⁴ Study quality was assessed by 2+ authors independently (AB, LL, NR) and disagreements were resolved by discussion or by consulting a third author (JJ).

RESULTS

Study selection

Records from 658 studies were identified through database searches and by hand-searching the reference lists of the primary studies. After deduplication and initial title and abstract screening, 590 were excluded for being the wrong population (i.e., participants without ADHD or with some other primary diagnosis of interest, like substance use disorder), the wrong outcome (without assessments of interoception and symptoms of ADHD) or for not being empirical studies in humans. Forty-six full-text articles were assessed for eligibility and 17 published articles based on 18 studies met the inclusion criteria (one published report included data from 2 studies which both met inclusion criteria; Figure 2).³⁵ Another published report included data from two studies, but only one met inclusion criteria.³⁶ There were no disagreements between authors on study inclusion/exclusion. In addition, six studies assessed interoception and emotional regulation strategy use or emotional competence, but not emotion regulation, and therefore did not meet full inclusion criteria but are briefly described below.

Figure 2:

PRISMA flowchart for included studies

Study designs

Studies in participants diagnosed with ADHD

Four cross-sectional studies examined interoception in children with ADHD; one was a dissertation and three were studies published in peer-reviewed journals (Table 1).^37–39 The dissertation enrolled children with diagnoses of ADHD.³⁹ The peer-reviewed studies enrolled children with a diagnosis of ADHD, ASD, or both (hereafter ADHD+ASD) as well as comparison groups comprised of children without ADHD or ASD, ^37,38,40 although one study’s comparison group was composed of children with either high or low levels of autistic traits.³⁸ One of these studies excluded children with any prior training in practices that may impact interoceptive performance such as yoga, mindfulness, or tai chi. Two studies enrolled adults with formal diagnoses of ADHD and comparison groups of adults without ADHD.^41,42

Table 1:

Characteristics of included studies

Author	Country	Population	N	Age	Description of sample
Ide-Okochi 2022	Japan	Diagnosed with ADHD	15	10–18	Children with ADHD, ASD or both; children without ADHD or ASD
Ide-Okochi 2024	Japan	Diagnosed with ADHD	24	NR	Children with ADHD, ASD or both who did or did not have a history of maltreatment; children without ADHD or ASD
Kutscheidt 2019	Germany	Diagnosed with ADHD	30	NR	Adults with and without ADHD
Odeh 2019	US	Diagnosed with ADHD	62	4–17	Children with ADHD
Wiersema 2018	Belgium	Diagnosed with ADHD	37	19–37	Adults with and without ADHD
Yang 2022	China	Diagnosed with ADHD	50	6–13	Children with ASD or ASD+ADHD; children without ASD but with high or low ASD traits
Baiano 2021	Italy	General population	30	20–31	Adults without psychiatric diagnoses or medication use
Bishop 2023	Australia	General population	30	8–12	Children without psychological, neurological, or behavioral diagnoses
Cheung 2023	Australia	General population	25	9–12	Children without psychological, neurological, or behavioral diagnoses
Cicarelli 2024	Canada	General population	76	18–63	Community-dwelling adults, of which n=37 reported they had ADHD
Herman 2018	UK	General population	574	18–45	Community-dwelling adults
Herman 2019 ¥	UK	General population	60	18–37	Students and staff from a university
Kaisari 2018 §	UK	General population	237	18–60	Community-dwelling adults, of which n=79 reported they had ADHD
Kaisari 2018 ¥	UK	General population	142	18–32	University students without psychological diagnoses or medication use
Martin 2023	UK	General population	345	18–60	Community-dwelling adults, some of whom had clinically significant ADHD symptoms
Rae 2020	UK	General population	41	18–28	Adults without psychiatric diagnoses or medication use
Schandry 1981	US	General population	39	18–40	NR
Schuette 2021	US	General population	95	18–29	Undergraduate psychology students

ADHD, attention-deficit/hyperactivity disorder. ASD, autism spectrum disorder. NR, not reported. UK, United Kingdom. US, United States.

^§: study 1 and

^¥: study 2 (published report included data from two separate studies).

Studies in participants in the general population

Twelve studies examined the association between interoception and inattention, hyperactivity, impulsivity, executive function, or emotional regulation. All were cross-sectional studies with self-report questionnaires and/or lab tasks. Two were in children^43,44 while ten were in adults.^{35,36,45–51}

The pediatric studies evaluated the association between interoception and emotional regulation⁴³ or executive function.⁴⁴ Seven adult studies evaluated the association between interoception and inattention and hyperactivity/impulsivity,^48,51 impulsivity alone,^36,46,47 or emotional regulation^49,50 in adults in the general population without known diagnoses of ADHD. Three adult studies were in large community samples of adults reporting a range of ADHD symptoms; they examined the association between ADHD symptoms and measures of interoception.^35,45 Of the ten studies in adults, nine were published in peer-reviewed journals and one was a thesis for a master’s degree.⁵¹

Outcome measures

Objective measures of interoception

The most frequently used objective measure (i.e., laboratory test) of interoception was the heartbeat tracking test. The heartbeat tracking, or heartbeat perception, test was first developed by Schandry et al. in 1981.⁴⁹ In this test, participants wear a heartrate monitor (e.g., a pulse oximeter or electrocardiogram (EKG)) to monitor their heartbeat over a period of time. Participants are asked to count their heartbeats for that same period, without seeing the heart rate monitor or placing their fingers on their own pulse. The number of beats counted by the participant is compared to the recorded beats, and a score is calculated by subtracting the recorded beats from the participants’ counted beats, divided by the recorded beats. This process is repeated and then averaged across many trials (n=number of trials). The formula is as follows:

$$\mathrm{I}\mathrm{n}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{c}\mathrm{e}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{c}\mathrm{o}\mathrm{r}\mathrm{e}=\frac{1}{n}{\sum }_{i=1}^n\left(1-\frac{\mid \mathit{recorded}\mathit{beats}-\mathit{counted}\mathit{beats}\mid }{\mathit{recorded}\mathit{beats}}\right)$$

The result is a score ranging from 0–1 where lower numbers represent lower interoceptive abilities, indicating a larger difference between the recorded and counted heartbeats. A higher number represents higher interoceptive abilities, meaning the recorded and counted heartbeats were closer in number. The studies that used the heartbeat tracking test used either the formula above or variations of it (Table S3).

Similar to the heartbeat tracking test, the heartbeat discrimination or detection task is based on work by Whitehead et al. in 1977.⁵² During this test, a participant is connected to an EKG machine, which monitors their heartbeat. The participant hears a series of tones or sees a rhythmic flashing light and is asked to report if the sounds or lights are synchronous or asynchronous with their heartbeat. Multiple trials are run, with the participant indicating after each trial whether there was synchrony. There are many variants of these methods. For example, participants may be asked to push a button every time their heart beats, and the responses are compared to an objective measure like an EKG. Scoring methods are similar to those from the heartbeat tracking test described above.

One study in children included an eye-tracking interoceptive accuracy test (EIAT), which was previously developed by the study authors^38,53 in order to assess interoception in children with psychiatric disorders. During the eye-tracking test, participants wore an electrode on one finger to measure their heartbeat. They were shown a screen with two jumping rabbits: one coincided with their heartbeat and one did not. Participants were asked to visually focus on the rabbit that was jumping with the same timing as their heartbeat. An eye-tracking device on the computer calculated how long they spent looking at the correct rabbit. The eye-tracking test assesses a participant’s awareness of the rhythm of their heartbeat, rather than their ability to count the number of beats. It is loosely based on the heartbeat discrimination test.⁵²

Self-report measures of interoception

Self-report measures of interoception included questionnaires such as the Multidimensional Assessment of Interoceptive Awareness (MAIA) questionnaire, the Body Perception Questionnaire (BPQ), the Self-Awareness Questionnaire (SAQ), Interoceptive Accuracy Scale (IAS), Sensational Brain’s Sensory Symptoms Checklist, the Reliance on Internal Hunger/Satiety Cues Subscale from the Intuitive Eating Scale (IES), or their derivatives (i.e., translations or adaptations for different populations or age groups).

The most used questionnaire was the MAIA, a 37-item validated self-report questionnaire with eight subscales, all of which assess different dimensions of interoception.¹⁶ Examples of items include “I notice changes in my breathing, such as whether it slows down or speeds up.” Each item has six possible answers, from 0 (Never) to 5 (Always). Subscales are scored by finding the mean of all items on that subscale. It has been adapted for children 7–17 years old (MAIA-y)⁵⁴ and has been translated into at least 28 languages. Details on the other questionnaires can be found in Table S4.

Results of individuals studies

Aim 1: Interoception in individuals with diagnoses of ADHD

1.1. Studies in children/adolescents with ADHD

Four studies examined interoception in a group of children with ADHD (Table 2). Three studies were peer-reviewed^37,38,40 while one was a non-peer-reviewed dissertation.³⁹ Details about demographic characteristics of the samples can be found in Tables S5–7.

Table 2:

Results of studies in participants diagnosed with ADHD

Author, year	Outcome measures		Results
	Objective tests (# trials)	Questionnaires
Studies in children/adolescents
Ide-Okochi 2022	HTT (3)	MAIA	HTT score was not different between ADHD/ASD and comparison group. 6 of 8 MAIA subscales were significantly lower in the ADHD/ASD group.
Ide-Okochi 2024	HTT (NR)	MAIA	The ADHD/ASD-maltreatment group had significantly lower HTT than the group without ADHD/ASD. The ADHD/ASD group had significantly lower scores on 3 of 8 subscales of the MAIA compared to the group without ADHD/ASD.
Odeh 2019	--	SBSSC	Interoception didn’t differ between presentations of ADHD (Inattentive, Hyperactive/Impulsive, Combined).
Yang 2022	EIAT (36)	ADHD-RS-IV, AQ-C, CAST	Children with ASD with and without ADHD had significantly lower eye-tracking scores than comparison children. Interoception negatively correlated with ASD and ADHD symptoms.
Studies in adults
Kutscheidt 2019	HTT (11)	BIS ASRS ADHS-SB WURS-K	HTT scores were significantly lower in ADHD group. There was no correlation between questionnaires and HTT.
Wiersema 2018	HTT (2)	BPQ	HTT and BPQ Awareness subscale were not different between ADHD and comparison groups.

ADHD, attention-deficit hyperactivity disorder. ADHD-RS, ADHD Rating Scale version 4. ADHS-SB, ADHD Self-Rating Scale. ASD, autism spectrum disorder. AQ-C, Autism-Spectrum Quotient-Child. ASRS, Adult ADHD Self-Report Scale. BIS, Barratt impulsiveness scale. BPQ, Body Perception Questionnaire. CAST, Child Autism Spectrum Test. EIAT, eye-tracking interoceptive accuracy test. HTT, heartbeat tracking task. MAIA, Multidimensional Assessment of Interoceptive Awareness. NR, not reported. SBSSC, Sensational Brain’s Sensory Symptoms Checklist. WURS-K, Wender Utah Rating Scale.

The three peer-reviewed studies compared interoception scores between a group of children with ADHD, ASD, or ADHD+ASD and a comparison group. Ide-Okochi et al. 2021 enrolled children with ADHD, ASD, or ADHD+ASD (n=8 total), and children without ADHD or ASD (n=7) who each completed the MAIA and the heartbeat tracking test.³⁷ The specific number of individual children who had ADHD, ASD, or both was not reported. On the heartbeat tracking test, the ADHD/ASD group’s score was lower than the comparison group (mean difference= 0.22; p=0.07) yet not statistically significant, perhaps due to the small sample size. On the MAIA, six of eight subscales were significantly lower in the ADHD/ASD group than the comparison group (the Noticing, Attention Regulation, Emotional Awareness, Self-Regulation, Body Listening, and Trusting subscales).

Another study by the same group (Ide-Okochi et al. 2024) enrolled children with ADHD, ASD, or ADHD+ASD who did (n=9) or did not (n=8) have a history of maltreatment/abuse by their primary caregivers, as well as a comparison groups of children without ADHD, ASD, or a known history of abuse (n=7).⁴⁰ Of the 17 children with ADHD or ASD, four had ADHD only, one had ASD only, and 12 had ADHD+ASD. The study examined differences in the MAIA and in the heartbeat tracking task between the three groups. On the heartbeat tracking task, the score was lower in the ADHD/ASD-maltreatment group than the comparison group without ADHD/ASD (mean difference=0.25; p=0.043), while the score of the ADHD/ASD group without maltreatment was not different than the other two. On the MAIA, 3/8 subscales were lower in the ADHD/ASD group without maltreatment than the comparison group without ADHD/ASD (Body Listening, Self-Regulation, and Trusting subscales); the scores of the ADHD/ASD-maltreatment group were not different.

Yang et al. enrolled children with ADHD+ASD (n=20) or ASD alone (n=30).³⁸ The comparison group was selected from children without ADHD/ASD (n=509) whose parents completed a series of questionnaires about their child. Only children with parent-reported low or high autistic traits were retained, from which children were age- and gender-matched to the ADHD/ASD group, and then randomly selected. Included in the study were 63 total children without ADHD or ASD. Of these, 31 children had parent-rated low autistic traits, and 32 had high autistic traits. All children completed the eye-tracking task (EIAT), and parents completed the ADHD Rating Scale-IV (ADHD-RS-IV), the Child Autism Spectrum Test (CAST), and the Autism Spectrum Quotient-Child (AQ-C).

Eye-tracking scores were significantly different between the four groups: ADHD+ASD, ASD, comparison group with low autistic traits, and comparison group with high autistic traits. Post hoc contrasts revealed all groups were significantly different from each other, except the ADHD+ASD and ASD groups, which had comparable scores (mean difference=3.37; p=0.28). The ASD+ADHD and ASD groups had lower interoception scores than both comparison groups. In the entire sample, interoception score was negatively correlated with symptoms of ASD via the AQ-C (r=−0.33, p<0.01). Interoception score was also negatively correlated with ADHD symptoms via the inattention and hyperactivity/impulsivity subscales of the ADHD-RS (inattention: r= − 0.40, p<0.01; hyperactivity/impulsivity: r= − 0.44, p<0.01).

The dissertation (Odeh 2019) enrolled children with a diagnosis of ADHD. Parents filled out the Sensational Brain’s Sensory Symptoms Checklist (SBSSC), which included an Interoception subscale.³⁹ They examined whether interoception was different between the different presentations of ADHD (Inattentive, Hyperactive/Impulsive, or Combined), and found it was not.

1.2. Studies in adults with ADHD

Two studies examined interoception in adults with/without ADHD. Kutschiedt et al. enrolled participants who completed the heartbeat tracking test and questionnaires (Table 2).⁴¹ On the heartbeat tracking test, the score in the ADHD group (mean=0.55, SD=0.15) was lower than the comparison group without ADHD (mean=0.71, SD=0.17; p=0.025). Participants on medication for ADHD were not asked to abstain from it before the study assessments, but the difference between medicated and unmedicated participants was examined. Neither heartrate nor scores on the heartbeat tracking test were significantly different between the medicated (n=8) and nonmedicated (n=6) participants with ADHD.

Wiersema et al. enrolled adults with ADHD and an age- and sex-matched comparison group without a previous diagnosis of ADHD or ASD.⁵⁵ Participants in the ADHD group who were taking stimulant medication for ADHD were asked to stop 48 hours before the laboratory visit (n=12 regular users; n=7 occasional users; n=4 past users; n=1 never used). All participants completed six trials of the heartbeat tracking test and the BPQ.⁴² Participants in the comparison group had to score below a clinical cut-off for ADHD symptoms on the Self-Report Questionnaire for Attention Problems and Hyperactivity questionnaire, and below a clinical cut-off on the Wechsler Utah Rating Scale for ADHD (WURS), which inquired about ADHD symptoms in childhood.

Heartbeat tracking test scores were not significantly different between the ADHD group and the comparison group (p=0.62). There were no differences between ADHD participants who were and were not currently using stimulant medication, although all participants stopped medication 48 hours prior to the laboratory visit. Scores on the Awareness subscale of the BPQ were not significantly different between groups.

Aim 2: Interoception and symptoms of ADHD in the general population

2.1. Studies in children in the general population

Two studies assessed interoception and either emotion regulation or executive function in children in the general population (Table 3). One study was in children 8–12 years old.⁴⁴ The MAIA-y was used to assess interoception, and the Behavior Rating Inventory of Executive Function, 2^nd edition (BRIEF-2) assessed executive function. The Emotional Awareness subscale of the MAIA-y was significantly positively associated with the total score on the BRIEF-2, as well as two of the three domains. The second study enrolled children 9–12 years old.⁴³ The MAIA-y was used to assess interoception, and the Cognitive Emotion Regulation Questionnaire for children (CERQ-k) was used to assess emotion regulation. The Not-Distracting subscale on the MAIA-y was significantly negatively associated with the Planning subscale of the CERQ-k, and the Self-Regulation and Not-Distracting subscales of the MAIA-y were significantly associated with the Positive Reappraisal subscale of the CERQ-k.

Table 3:

Results of studies of individuals without ADHD or mixed samples

Study	Construct	Measure	Results
Baiano 2021	Interoception	HTT	SAQ score was significantly associated with impulsivity but not inhibitory control. HTT not significantly associated with either.
	Interoception	SAQ
	Impulsivity	BIS-11
	Inhibitory control	Go/NoGo
Bishop 2023	Executive function	BRIEF-2	Emotional Awareness subscale (MAIA-y) associated with 2 of 3 domains on the BRIEF-2 (Emotional Regulation, Cognitive Regulation) as well as total score.
	Interoception	MAIA-Y
Cheung 2023	Emotion regulation	CERQ-k	Not Distracting subscale (MAIA-y) associated with the Planning subscale (CERQ-k); the Self-Regulation and Not-Distracting (MAIA-y) associated with Positive Reappraisal subscale (CERQ-k).
	Interoception	MAIA-Y
Cicarelli 2024	ADHD symptoms	ASRS-v1	Total MAIA score and 4 of 8 MAIA subscales were associated with inattention and hyperactivity/impulsivity.
	Interoception	MAIA-2
Herman 2018	Impulsivity	UPPS-P	BPQ score wasn’t significantly associated with measures of impulsivity.
	Impulsivity	MFFT
	Impulsivity	MCQ
	Inhibitory control	BIS/BAS
	Interoception	BPQ
Herman 2019 (s2)	Impulsivity	BIS	Seven measures of interoception were assessed: BPQ score, HTT, HDT, confidence on HTT, confidence on HDT, and the correlation between score and confidence on the HTT and the HDT. The associations between all seven and the three subscales of the BIS (attention, motor impulsivity, non-planning impulsivity) were investigated. Only the HDT score and non-planning impulsivity were significantly associated.
	Interoception	BPQ
	Interoception	HTT
	Interoception	HDT
Kaisari 2018 (s1)	ADHD symptoms	CAARS-SRSV	The Inattentive, but not Hyperactive/Impulsive, domain of the CAARS was significantly associated with IES-Reliance.
	Interoception	IES
Kaisari 2018 (s2)	ADHD symptoms	CAARS-SRSV	The Inattentive, but not Hyperactive/Impulsive, domain of the CAARS was significantly associated with IES-Reliance. HTT score was not associated with either domain, or with Reliance.
	Impulsivity	BIS
	Inhibitory Control	Go/NoGo
	Interoception	IES
	Interoception	HTT
Martin 2023	ADHD symptoms	CAARS-SV	The Inattentive domain of the CAARS was significantly associated with the IAS score, while the Hyperactive/Impulsive domain was significantly associated with the BPQ-VSF score.
	Interoception	IES
	Interoception	IAS
	Interoception	BPQ-VSF
Rae 2020	ADHD symptoms	CAARS-SS	Interoceptive awareness (correlation between HTT and confidence in HTT) was negatively associated with impulsivity on the Go/NoGo task. Participants with lower confidence in their HTT score responded more impulsively on the Go/NoGo task. Interoceptive insight (the discrepancy between confidence and accuracy on the HTT) was negatively correlated with reaction time on the Go/NoGo.
	Impulsivity	BIS
	Impulsivity	UPPS-P
	Inhibitory Control	Go/NoGo
	Interoception	HTT
	Interoception	BPQ
Schandry 1981	Emotion regulation	FPI*	Good heartbeat perceivers (high interoception) had a higher emotional dysregulation score compared to bad heartbeat perceivers (low interoception).
	Interoception	HTT
Schuette 2021	Emotion regulation	PEC	The Noticing subscale of the MAIA was significantly associated with emotion regulation, but the Body Listening subscale, HTT score, and confidence in HTT score were not.
	Interoception	MAIA
	Interoception	HTT

ASRS-v1, Adult ADHD Self-Report Scale Symptom Checklist version 1. BIS, Barratt Impulsivity Scale. BIS/BAS, Behavioral Inhibition System/Behavioral Activation System Questionnaire. BPQ, Body Perception Questionnaire. BPQ-VSF, Body Perception Questionnaire-Very Short Form. BPQ-VSF, Body Perception Questionnaire-Very Short Form. BRIEF-2, Behavior Rating Inventory of Executive Function, 2nd edition. CAARS-SRSV, Conner’s Adult ADHD Rating Scale, Self-Report Screening Version. CAARS-SS, Conner’s Adult ADHD Rating Scale- Short Scale. CAARS:SV, Conner’s Adult ADHD Rating Scale, Short Version. CERQ-k, Cognitive Emotion Regulation Questionnaire for children. FPI, Freiburger Persönlichkeitsinventar, Emotional Lability scale. Go/NoGo, Go/NoGo task. HDT, Heartbeat Discrimination Task. HTT, Heartbeat Tracking Task. IAS, Interoceptive Accuracy Scale. IES, Intuitive Eating Scale. MAIA-y, Multidimensional Assessment of Interoceptive Awareness - youth. MCQ, Monetary Choice Questionnaire. MFFT, Matching Familiar Figures Task. PEC, Profile of Emotional Competence. PEC, Profile of Emotional Competence. UPPS-P Impulse Behavior Scale.

2.1. Studies in adults in the general population

Ten studies (published in nine articles) assessed interoception and symptoms of inattention, hyperactivity, impulsivity, or emotion regulation in adults in the general population.

2.1.1. Inattention and hyperactivity/impulsivity

Four studies (published in three articles), evaluated interoception, inattention and hyperactivity/impulsivity in large community samples of adults. The first study enrolled a large community sample of adults, some of whom reported a diagnosis of ADHD (33%).³⁵ ADHD symptoms were assessed with Conner’s Adult ADHD Rating Scale, Self-Report Screening Version (CAARS), and interoception with the Reliance on Internal Hunger/Satiety Cues subscale from the IES. The Reliance subscale has 6 items on internal sensations (Table S4). The Reliance subscale was significantly associated with the Inattentive domain of the CAARS. This was replicated in a second study which enrolled university students without ADHD.³⁵ The CAARS, the Barratt Impulsivity Scale (BIS), the IES-Reliance subscale, and the HTT and Go/NoGo tasks were administered. As in the previous study, the Inattention domain of the CAARS was significantly associated with the Reliance subscale of the IES, but the Hyperactivity/Impulsivity domain was not.

The third study enrolled a large community sample of adults, of whom 47% reported clinically significant ADHD symptoms.⁴⁵ Questionnaires included Conner’s Adult ADHD Rating Scale: Short Version (CAARS: SV), the IES-Reliance subscale, the Body Perception Questionnaire-Very Short Form (BPQ-VSF), and the IAS. The Inattentive domain of the CAARS was significantly associated with IAS, while the Hyperactive/Impulsive domain was associated with the BPQ-VSF. The fourth study (a thesis) enrolled 76 adults, of whom 37 reported they had a diagnosis of ADHD.⁵¹ Questionnaires included the Adult ADHD Self-Report Symptom Checklist (ASRS v1), the MAIA, and others. The inattention and hyperactive/impulsive domain of the ASRS were significantly negatively correlated with MAIA total score.

2.1.2. Impulsivity

Four studies evaluated interoception and impulsivity in adults. In the first study, interoceptive sensibility via the BPQ-SF was not associated with 2 lab-task measures of impulsivity.⁴⁷ The second study examined impulsivity via the 3 subscales of the BIS and seven measures of interoception: BPQ score, HTT, HDT, confidence on HTT, confidence on HDT, interoceptive awareness for HTT, and interoceptive awareness for HDT. Only 1 of 3 subscales of the BIS were significantly and negatively associated with HDT score, but no other predictors or subscales were significantly associated.³⁶ The third study examined impulsivity with a Go/NoGo test, interoceptive sensibility with the BPQ, interoceptive accuracy with the HTT, and interoceptive awareness as a calculated score. There was a negative association between interoceptive awareness and impulsivity on the Go/NoGo test; participants with lower interoceptive abilities had more urges to react and greater impulsivity.⁴⁸ The fourth study examined interoceptive accuracy with the HTT and interoceptive sensibility with the SAQ.⁴⁶ Impulsivity was assessed with the three subscales of the BIS, and inhibitory control with the Go/NoGo task. SAQ score was a significant predictor of trait impulsivity (BIS-11 total score) but not inhibitory control (mean reaction time on the Go/NoGo test). HTT score was not a significant predictor of trait impulsivity, or inhibitory control.

2.1.3. Emotion regulation

Two studies evaluated interoception and emotion regulation in adults. One study enrolled participants who completed the HTT and the Emotional Lability subscale of a personality questionnaire.⁴⁹ Based on their performance on the HTT, participants were divided into two categories: “good heartbeat perceivers” with high interoceptive accuracy, and “bad heartbeat perceivers” with low interoceptive accuracy. The good perceivers had significantly higher emotion regulation scores compared to the bad perceivers, suggesting that the participants with higher interoceptive accuracy had emotional experiences which were more frequent and/or more intense. The second study examined interoception via the MAIA Noticing and Body Listening subscales and the HTT, and emotion regulation via the Emotion Regulation subscale of the Profile of Emotional Competence (PEC). ⁵⁶ Participants reported on their confidence regarding their accuracy on the HTT. In a hierarchical regression series predicting emotion regulation with all the interoceptive measures, only Noticing was a significant predictor of Emotion Regulation.

Additional studies of interest

Six studies examined interoception and emotional competence,⁵⁷ emotional awareness⁵⁸ or frequency of emotion regulation strategy use.^59–62 This review focused on emotional dysregulation/lability, which commonly co-occurs with ADHD,^63,64 therefore these studies were outside of the scope of this review but are briefly summarized in Appendix 3.

Study quality

The Newcastle-Ottawa Scale adapted for cross-sectional studies was further adapted for use in individuals with ADHD.^31,32,34 This scale was used to assess the quality of the six studies which specifically enrolled individuals with and without ADHD. Items assessed included adequacy of participant categorization, outcome measures, sample size, and reporting. Studies were assigned a score from 0–7, with a higher score reflecting higher quality. The 12 studies in the general population were not assessed as the Newcastle-Ottawa scale is generally only used to evaluate case-control studies. The two studies in adults with ADHD were assigned a score of 4.5 and 6,^41,42 while the studies in children were assigned scores between 4–6.^37,38,40 The dissertation was assigned a score of 2.5.³⁹ The most common issues were lack of a power or sample size calculation, and failing to report whether or not participants were taking medication in general, and on the day of testing. Further details in Table S8 and Appendix 4.

DISCUSSION

This is the first review to summarize the literature on interoception and ADHD, including the comparison of interoception in samples with/without ADHD, and the association of interoception with symptoms of ADHD in the general population. Studies that were included assessed several different dimensions of interoception, including accuracy, sensibility, and awareness.

Summary of our results

We identified six studies in participants with diagnoses of ADHD: four in children and two in adults. Two of the studies in children included participants with ADHD, ASD, or both, plus comparison groups of children without either.^37,38 These children with ADHD+/−ASD had diminished interoceptive accuracy and sensibility compared to typically developing children as measured by both self-report questionnaires and objective tests. Both studies found the same 3 subscales of the MAIA (Body Listening, Self-Regulation, and Trusting subscales) reduced in the ADHD+/− ASD group compared to the comparison group of children without either. A third study included children with ADHD, ASD, or both, as well as a comparison group of children without diagnoses of ASD but with parent-reported high and low autistic traits.³⁸ This study also found lower interoception accuracy in the children with ADHD+/− ASD than either of the other groups. Additionally, ADHD symptom severity was negatively correlated with interoceptive accuracy, suggesting that children with decreased interoceptive accuracy had more severe symptoms of ADHD. A dissertation did not find differences in interoceptive sensibility between different presentations of ADHD (Inattentive, Hyperactive/Impulsive, or Combination).³⁹

In the two studies in adults with ADHD, one found differences in interoceptive accuracy between ADHD and non-ADHD groups, while the other did not. The one study in adults that assessed interoceptive sensibility found no difference between groups. Combined, these studies suggest that interoceptive accuracy and sensibility may be diminished in children with ADHD with or without ASD, but these differences may not be maintained to adulthood.

Twelve studies in the general population examined the association between interoception and symptoms of ADHD. In two studies of children, interoceptive sensibility was associated with both executive function⁴⁴ and emotional regulation,⁴³ suggesting participants with higher interoceptive sensibility reported higher executive function and emotion regulation. In three studies in adults, interoceptive sensibility was negatively associated with inattention, though only associated with hyperactivity/impulsivity in one of the three.^35,45 Four studies in adults examined impulsivity alone and reported it was positively associated with interoceptive sensibility⁴⁶ and interoceptive awareness.⁴⁸ However, impulsivity was negatively associated with interoceptive accuracy.³⁶ Two studies in adults examined interoception and emotion regulation, both finding participants with higher interoceptive accuracy had higher emotional regulation scores.^49,50 As emotional dysregulation is a common co-occurring symptom in ADHD, this suggests that higher interoceptive accuracy would be negatively associated with emotional dysregulation.

Overall, these results suggest: 1) interoceptive accuracy is lower in participants who report higher symptoms of impulsivity and emotional dysregulation; 2) interoceptive sensibility is lower in participants who report higher symptoms of inattention and hyperactivity/impulsivity; and 3) interoceptive awareness is lower in participants who report higher symptoms of impulsivity. The negative association between all three dimensions of interoception and symptoms of ADHD suggest interventions aimed at increasing interoceptive abilities have the potential to reduce ADHD symptoms. The domains of interoception assessed in each study are shown in Figure 3, and a summary of results in Table 4.

Figure 3:

Domains of interoception assessed in each study

s1, study one, and s2, study 2.

Table 4:

Number of studies reporting statistically significant associations between domains of interoception and ADHD symptoms in the general population

Lack of consistent measures across studies

How to best quantify interoception remains the largest challenge in the field.⁶⁵ This is an issue with objective measures⁶⁶ as well as self-report measures. For example, the most used self-report questionnaires in these studies— the BPQ and the MAIA— have been validated; however, a large study in adults (n=1003) administered multiple self-report questionnaires on interoception, including the BPQ and MAIA, and examined the different constructs being measured with a factor analysis.⁶⁷ The authors determined that each questionnaire measured a distinct dimension of interoception, and cautioned against comparing results across studies which use different questionnaires. Furthermore, in this study, participants’ scores on the BPQ and MAIA were not correlated with each other.⁶⁷

Classic objective measures, such as the heartbeat tracking task, the heartbeat detection task, and an eye-tracking task, also have important psychometric limitations.⁶⁸ For example, growing concerns have been raised regarding the validity of the heartbeat counting task, which is likely the most frequently used measure of interoceptive accuracy. Among healthy participants, accuracy scores largely reflected non-interoceptive processes (i.e., estimating and diffuse detection) and were drastically reduced (~50%) when they asked participants to avoid relying on non-interoceptive signals and to only report the heartbeats they perceive.⁶⁸ Further, when a meta-analytical approach was used to assess the association between two commonly used objective tasks, the heartbeat counting and discrimination tasks, pooled findings (k=22) revealed a small relationship between scores on the measures, questioning the interchangeable use of the two tasks.⁶⁹

Further, it has been suggested that that all interoception measures may be contaminated by signal intensity. This is because researchers are currently unable to precisely control the intensity of internal bodily signals.¹⁷ Stated differently, individuals may differ in task performance because the signal to be detected is weaker or stronger, not because they differ in detection ability, resulting in substantial consequences for our scientific understanding.⁷⁰ Hence, these limitations make it difficult to draw clear conclusions from the literature.

There is agreement in the field that new methods of assessing interoception are needed.⁷¹ First, given the numerous and often confusing terms associated with the dimensions of interoception, some have advocated for more precise terminology, and proposed three alternative constructs: a) Interoceptive detection, the capacity to detect internal bodily states (or changes); b) Interoceptive correspondence, the degree of correspondence between objective and felt bodily states/changes; and c) Interoceptive estimation, the capacity to estimate one’s internal bodily states/changes. Authors have argued that these terms would improve conceptual clarity and allow for each construct to be targeted by different existing measures. Proposed alternative methods of the heartbeat detection task include breathing tasks, adaptations of previous heartbeat discrimination tasks, and physical activity-based paradigms.^57,72–74 Finally, it has been suggested that interoception should be assessed at perceptual levels (e.g., cardiac, gastric, thermosensory), rather than a single construct. For a thorough discussion see Crucianelli et al. 2022 and Murphy 2024.^9,70

Possible mechanisms of altered interoception in ADHD

A recent meta-analysis demonstrated compelling evidence that interoception is of central importance for effective regulation of emotional responses.⁷⁵ Better interoceptive abilities have been linked to greater emotional and attentional regulation.^59,62,76,77 Given that ADHD is characterized by attentional and emotional dysregulation, reduced interoception may contribute to the self-regulation deficits that drive ADHD symptoms. Self-regulation is thought to be strongly dependent on the moment-to-moment awareness of bodily states relayed via interoceptive pathways.^78,79 More specifically, effective self-regulation is thought to require the ability to accurately detect and evaluate physiological and emotional states which is then accompanied by appropriate regulation strategies to temper and influence the emotional response.^11,80 Previous research has demonstrated that greater interoceptive sensibility or accuracy is associated with habitual use of emotion regulation strategies.^59,62

One important regulation strategy is reappraisal, the reinterpretation of the meaning of an adverse stimulus or situation so as to reduce its negative emotional impact.⁸¹ According to Gross’ process model, cognitive reappraisal belongs to the family of antecedent-focused strategies, which aim to regulate emotions before they manifest as a full-scale emotional response.⁸² Multiple studies have documented positive associations between interoception and cognitive reappraisal. Moreover, studies have reported that greater interoception was associated with more flexible and adaptive use of different emotion regulation strategies, that is, the ability to choose an appropriate strategy for the specific situation (distraction, reappraisal, acceptance, etc.).^58,59 Individuals with lower interoceptive sensibility or accuracy tend to choose the same emotion regulation strategy, regardless of the situation, if one is chosen at all.

Thus, interoception may be seen as a precursor to emotional regulation.⁸³ The interoceptive system is responsible for creating homeostatic maps of the body and orchestrating regulatory responses at both a conscious level, through emotions and feelings, and at the autonomic level.⁸⁴ The process of interoception which requires the interplay between perception of, and attention to, bodily sensations and the cognitive-affective appraisal of the bodily sensations,^11,80 may, in fact, underpin the self-regulation deficits that contribute to ADHD symptoms. Fortunately, the bidirectional interactions between the central nervous system and other physiological systems offer suitable opportunities for experimental manipulations of mind–body interactions, leading to promising clinical behavioral interventions.

Interoception in ASD

Though this is the first systematic review on interoception in ADHD, other reviews have assessed interoception in ASD, which shares some clinical features with ADHD.^85,86 A meta-analysis in children and adults (k=15) found reduced interoceptive accuracy on the heartbeat tracking task but increased confidence in their task performance in autistic compared to non-autistic individuals.⁸⁶ However, both objective and subjective assessments of interoception (heartbeat tracking and self-report questionnaires, respectively) in autistic populations present undeniable challenges. There are high rates of co-occurring conditions like anxiety and gastrointestinal disorders in autistic populations, which can impact internal processes such as digestion or heartrate and confound assessments of interoception.^87,88 It may be that autistic individuals have differing patterns of interoceptive disturbances, where some have heightened and some have diminished interoceptive abilities; if so, there is limited utility to the comparison of mean scores, such as in meta-analysis.^25,89

Of the four pediatric studies included in this review, three included participants with ASD, ADHD, or ASD+ADHD, and the fourth excluded participants with ASD,³⁹ aside from one participant who had Asperger’s Syndrome (as of the release of the DSM-5 in 2013, this is now considered part of ASD). In the adult studies, one study did not screen or exclude participants for ASD (Kutschiedt et al.).⁴¹ As interoceptive accuracy is altered in ASD,^85,86 findings could be partially attributable to the presence of ASD in both samples.^37,38 The only study which excluded individuals with ASD was the other study in adults with ADHD (Wiersema et al); they did not find reduced interoception in the ADHD group.⁴² Further research is needed to clarify the role of interoception in participants with ADHD but not ASD.

Interventions aimed at increasing interoceptive awareness in ASD successfully improve anxiety⁹⁰ and emotion regulation.⁹¹ These interventions included mindfulness practices⁹¹ and the heartbeat tracking task with real-time feedback about performance.⁹⁰ ADHD and ASD share some clinical features and up to half of children with ASD also have ADHD.⁹² Given the overlap between the two disorders, interventions that have been used successfully in autistic individuals may also be effective in those with ADHD. Potential interventions include mindfulness meditation, yoga, tai chi, physical therapy, visualization practices, and others.⁹³ These interventions have the potential to increase self-regulatory capacity and decrease impairment caused by symptoms, but further research is needed.

Interoception also plays a role in eating disorders,^94–98 depression,^99,100 anxiety,¹⁰⁰ and chronic pain.¹⁰¹ Atypical interoception may be a risk factor for psychopathology in general⁶⁵ and interventions which target interoception have the potential of benefiting a wide variety of neurological and psychiatric conditions.¹²

Quality of the evidence

The included studies were of moderate quality due to inadequate reporting of information about power, statistical results, and medication status of participants. The impact of medication on interoceptive abilities in ADHD has not yet been investigated in the literature. However, stimulant medication decreases pain sensitivity in adults with ADHD,¹⁰² suggesting it has impacts on sensory awareness. Only one study required participants to abstain from medication for 48 hours prior to the study assessments. If other participants were taking medication at the time of the study, this may have impacted their interoceptive abilities and influenced the results.

Role of culture in emotion and sensation

The six studies in participants with ADHD were conducted in Belgium, China, Germany, Japan, and the US. The differences in culture could have influenced results as conceptions of how the mind, body, and emotions interact are culturally determined.¹⁰³ In some cultures, it is believed that emotions arise from physical sensations and the emotions themselves are located in the body. In others, emotions are mental processes, essentially cognitive appraisals of a situation, entirely separate from the body.¹⁰³ When the MAIA was translated from English into Japanese, for example, a factor analysis identified fewer subscales than the original version, and the subscales were slightly different than those in the original version.¹⁰⁴ Focus groups in Japanese individuals who spoke both Japanese and English revealed marked differences in the way Japanese and US cultures conceptualize body awareness, emotion, and self-monitoring.¹⁰⁵ In the Japanese language, emotions are often framed as occurring in the body; i.e., anger is expressed with the phrase “my stomach is standing up” and excitement with “my heart is dancing.” ¹⁰⁵ In contrast, countries that primarily use the DSM-5 to guide psychiatric diagnoses consider the experience of emotional distress as a physical sensation to be a “somatoform disorder.” The cultural evaluation of the relationship between sensation and emotion plays a critical role in how individuals respond to and evaluate sensations originating inside their body. None of the included studies discussed cultural beliefs about emotion or embodiment, which may have impacted results. Examining the role of culture in research outcomes is complex. A multidisciplinary panel of experts recently identified multiple challenges to including the role of culture in health research, including the dynamic nature of culture itself.¹⁰⁶

Potential confounds

Interoceptive abilities usually increase as children age, peaking in adulthood, then declining in old age.¹⁰⁷ No age limitations were utilized in this review, though child and adult studies were reported separately. All five of the studies which included both ADHD and non-ADHD groups reported no differences in age and sex/gender between groups, so it is unlikely that the observed variation in interoceptive accuracy or sensibility could simply be due to differences in age. Intelligence may play a role in the ability to accurately assess internal sensations or respond to questionnaires. However, in the five studies with both ADHD and non-ADHD groups, there were either no differences in IQ between groups,^38,41 or intellectual disability (IQ < 70) was an exclusion criteria for the study.^37,40 However, in the two studies which did not compare IQ between groups, differences in intelligence could have contributed to observed results. Additionally, only one study screened for potential interoceptive training (yoga, mindfulness, or tai chi) so participants could have had varying levels of experience, which could have skewed results. Future research on interoception should screen for interoceptive training when enrolling participants to avoid this.

Strengths and limitations

Strengths of this review include the prospectively registered protocol, the comprehensive search strategy, and the inclusion of both peer-reviewed and non-peer-reviewed literature (i.e., a thesis and a dissertation). Limitations of the body of literature include the cross-sectional nature of all included studies, limiting the ability to assess changes over time or determine causality. It is not determined whether altered interoception contributes to symptoms of ADHD, or whether biological or psychological processes in ADHD contribute to altered interoception. Some combination of the two is also possible, where deficits in interoception may contribute to ADHD symptoms which then reinforce deficits in interoception. Additionally, attention, executive function, and emotion regulation are complex processes which are impacted by numerous psychological and social factors (i.e., stress, access to resources, etc.). The assessment of ADHD symptoms in twelve studies in the general population may not have truly measured symptoms of ADHD, but some other mechanism impacting psychological function. Though most studies screened participants for neurological and psychological disorders, it is still possible that other phenomena, not ADHD, were contributing to the symptoms in question. Another limitation of the body of literature was that only one of the six included studies in participants with ADHD fully screened and excluded participants on the basis of an ASD diagnosis. Further focused research is needed in participants with ADHD in the absence of ASD to clarify the role interoception may play in ADHD symptoms.

Conclusions

Decreased interoception may be associated with symptoms of ADHD including inattention, hyperactivity, impulsivity, executive function, and emotional dysregulation. Interventions focused on improving interoceptive abilities may provide an avenue for treatment.

Data Availability Statement

Only publicly available data from previously published research was used for this study.