Interoception

BB2 lecture content revision

what is interoception?

* feelings we receive from the body - signals about physical state
* distinct from the 5 senses e.g. HR, trouble breathing, hunger
* contributes to homeostasis
* self awareness and emotion
* restrictive- purely visceral (old definition)
* inclusive- general homeostatic capacity including broader sensations (recent definition)

what does afferent mean?

* sensory info from body → brain/ sensory neurons

what does efferent mean?

* regulatory signals from brain → body/ motor neurons

what is the neural anatomical pathway?

* signals about physiological state of the body , categories of afferent nerves
* ascending pathway: afferents → lamina I → brainstem / lateral spinothalamic tract→ thalamus, hypothalamus and amygdala → cortex (insula and anterior cingulate)
* signals that travel on this pathway are classed as interoception

fibres in the neural anatomical pathway

* different fibres carry different info and follow different pathway s
* afferent nerves tend to travel along the 2 smaller ones - A delta (for pain) and C (for affective touch)
* Small diameter A-delta and C afferent nerves, including Nociceptors (pain), thermoreceptors (temperature), osmoreceptors (thirst), baroreceptors (heart rate).
* C afferents are unmyelinated - slower

afferent projections from the dorsal horn provide …

… the somato-autonomic reflexes crucial for maintaining homeostasis

* reflexes don’t need to go to the cortex- not conscious

scratching and itching - the cycle

* Scratching inhibits itch via inhibitory interneurons in the dorsal horn of the spinal cord
* the pain caused by scratching inhibits the itching sensation
* but can make it worse e.g. if its a rash

interoception and emotion

* emotions can be felt in the body
* panic or excitement → increase HR
* fear or surprise → muscle tension
* can be mapped by colouring in how emotions feel , patterns are universal across cultures

gut feelings

* interconnection between the gut and the brain is deeply rooted in our language
* Neural network communicates with the brain about gastrointestinal homeostasis. ALSO emotion motivation and decision making
* Bidirectional – emotions can influence digestive system

where is the self? - different views

* suggested interceptive signals are intrinsic to the self, everything we do is in the context of the body - difficult to seperate body and self
* Aristotle - self in the heart
* Limanowski & Hecht (2011) – Head and heart
* Alsmith & Longo, 2014 – between the head and the torso
* study had people point to where they thought the self was, generally between the head and the heart

does the body shape behaviour?

* Affective experience influenced by the current body state
* Drive behaviour = achieve homeostasis (desirable physical state)
* Interoception drives emotion and helps maintain homeostasis

Garkinkel et al - Attentional Blink study- study 1

* rapid images, 2 targets you need to identify
* gap so that when you detect target 1 its difficult to detect target 2
* identification of T1 during RSVP of stimuli impairs ability to detect T2- T2 was a kind of facial expression
* they presented the stimuli at different stages of the heart beat, either when it contracts (systole) or relaxed (diastole)
* systole presentation led to greater proportion of fear faces detected, whne heart working harder it recognised more

Garkinkel et al - Attentional Blink study- study 2, MRI

* Neutral or fear faces presented briefly (100ms).


* Time locked to heartbeat (Diastole or Systole)
* asked ppts to give Emotion intensity ratings of stimuli
* systole stimuli lead to greater intensity for fear faces
* when heart is contracting it effects how easily you detect something and the intensity of what you detect
* related to individual differences in state anxiety
* anxiety → greater arousal → more time in systole → higher intensity perception state

Azevedo et al (2017)- study 1 of more complex behaviours and effect of cardiac cycle on racial stereotyping

* Comparing stimulus presentation when the muscle is contracting (Systole) with when it is relaxing (Diastole).
* Identify targets (tool or a weapon)


* Primed by Black or White males faces
* Stimuli were presented at Systole or Diastole
* systole exacerbated racially stereotyped responses
* more accurate detecting tools with white primes compared to black primes but only when in systole
* more accurate with weapon with a black face prime compared to white but again only in systole

\

Azevedo et al (2017)- study 2 , 1st person shooter task

* task to decide when to shoot of not- more active behaviour than study 1
* all ppts are white
* task: correct to shoot if they had a weapon and incorrect when they were unarmed
* either black or white unarmed man with a tool or black or white man with a weapon, presented in systole or diastole
* Participants chose to shoot unarmed Black men more often than unarmed White men
* about the same when armed, but unarmed changes when in systole and heart is working hard

\

interceptive accuracy

* How accurate an individual is at detecting and interpreting interoceptive signals when compared with objective measures. (heartbeat monitoring tasks)
* Measure what is happening and compare to what they think is happening to see how accurate they are
* HR monitoring
* high mechanoreceptive sensitivity may be able to feel and count their own heartbeats and estimate their heart rate.- high mechanoreceptive sensitivity → better interoceptive accuracy

interoceptive sensibility

* The subjective experience of interoception - how interoceptive signals are experienced irrespective of their objective reality. (Questionnaires: e.g. Multidimensional Assessment for Interoceptive Awareness, Body Awareness Questionnaire)
* ask people about their body awareness

interceptive awareness

* The correspondence between objective interoceptive accuracy and subjective reports.
* A metacognitive awareness of one’s own interoceptive accuracy.
* How good your are at detecting signals
* How good you think you are at detecting signals
* meta cognition about own interoceptive state

what are the 3 ways of measuring neural sigature of intercpetion?

* fMRI- can locate insula
* EEG- can’t locate insula but can time with heartbeats
* Heartbeat evoked potential (HEP)- time-locked to R wave

Pollatos & Schandry, (2004)- EEG and HEP study

* Participants counted their felt heartbeats during EEG
* Objectively measure heartbeats with ECG
* Use recorded actual heartbeats for examining interoceptive accuracy and identifying HEP
* Participants divided into good and poor perceivers
* HEP magnitude was larger than poor perceivers
* HEP magnitude (C4) positively correlated with accuracy of heartbeat detection
* good HR perceivers → higher HEP magnitude
* suggetss this is a neural signature of neural processing of heart beats in the brain

neural correlates of interoception - Critchley et al study

* HR detection task vs external task whilst in an fMRI
* presented with audio tones either synched to their HR or not and then asked them if they were synched (testing interoception) and if the notes were the same (testing exteroception)
* when attending to HR there was greater activation in the anterior insula when detecting a mismatch between notes and HR
* insula activity correlated with accuracy of HR detection and anxiety
* grey matter vol in the insula correlated with accuracy of HR detection and interoceptive sensibility

the insula cortex

* a fold within the cortex below the lateral sulcus, divided into anterior and posterior parts by the central insula sulcus
* associated with many cognitive and emotional task e.g. disgust, salience and bodily-related tasks e.g. perception, action awareness , internal body signals
* posterior connectivity- to thalamus
* anterior connectivity - to limbic system and amygdala
* its an anatomical hub
* involved in pain, emotion, cognition, social

How does interoceptive experience map onto the insula?- Avery et al fMRI study

* did an interoceptive task - attend to viscera
* and an exteroceptive task - asked to distinguish sweet and neutral taste
* examined the neural maps
* exteroceptive = more anterior insula activation
* interoceptive= more posterior insula activation
* found regions where exteroceptive and interoceptive overlapped- these were in the mid to posterior regions

\

If insular integrates sensory information in to a coherent experience how does this relate to behaviour?- Accolla & Carleton (2008) study

* Conditioned Taste Aversion (CTA) for sweet stimuli in rats
* uses conditioning to create aversions to certain tastes
* Adding malaise (induced sickness) to saccharin (artificial sweetener) induce aversion of saccharin.
* looked at the sweet and bitter maps in the rats gustatory cortex (insula) before, and after conditioning and after extinction

results:

* becomes a greater overlap once the aversion was created
* then goes back to separated after extinction (removal) of the taste aversion

this means:

* The insula integrates experience into valence maps.
* Insula integrates taste perception with physiological state and hedonic value.
* Changes in insular cortex maps corresponded with behaviour

cravings

* integrated with key reward circuit (ventral striatum)
* links physiological state and hedonic response
* cravings = physiological urges . desire to maintain homeostasis . links to addiction
* study compared smokers with insula lesions to smokers with other brain damage, those with insula damage >100 times more likely to quit smoking without relapse!

insula, cravings and addiction

* urges are associated with memories of body states
* links to reward network
* withdrawal disturbs homeostasis - intensifying cravings and an unpleasant physiological state

Interoceptive accuracy meta analysis with anxiety

* 55 studies using different measures of IA
* Clinical and subthreshold (healthy controls) samples
* found no relationship between IA and anxiety
* No effect of task or sample
* might suggest anxiety isn’t associated with actual detection of these signals but instead the way in which they are interpreted and thought about

interoceptive awareness and anxiety study

* 21 anxiety patients with panic attacks, 13 HC
* Heartbeat monitoting task
* measured metacognitions about interoception


* No difference in interoceptive accuracy
* Difference in metacognition (threatening beliefs about the body)
* Not detection of symptoms but how they are interpreted

HEP and anxiety study

* Generalised Anxiety Disorder (GAD) and health controls (HC)
* Examined HEP during resting state in 2 conditions:
* Eyes open (exteroception)
* Eyes closed (interoception) thought would be more in tune with interoceptive symptoms
* In HC HEP greater in eyes closed compared to eyes open rest- suggest better at interocpetion when the eyes were closed fior HC but no difference for people with GAD
* Greater focus in interoception with eyes closed
* In GAD, no difference between condition
* Cortical response to heartbeats greater in GAD
* Suggests people with GAD have heightened interoception compared to HC and thats why there isn’t a significant difference

the insual cortex and anxiety

* found to have abnormalities in structure and function across different forms of anxiety disorders
* Etkin et al (2007)- meta analysis of brain activation for emotional processign across different anxiety disorders
* Social Phobia
* Specific Phobia
* PTSD


* Patients with all three disorders demonstrated hyperactivity (patients > comparison subjects) in the amygdala and insula
* This pattern of activation was also noted for healthy subjects experiencing anticipatory anxiety during fear conditioning
* An exaggerated fear response—might be reflected in shared neurobiology

experience of body state

* Inaugural moment – initial sensory map from the body without prior experience, when you are first born have no experience to make predictions from
* Map then modified by life experience – predicted sensory experience of encountering stimuli


* Insular cortex, amygdala, anterior cingulate, ventral medial prefrontal cortex

Geuter et al - heat pain paradigm

* gave a cue of whether to expect warm or painful stimuli: low medium or high intensity
* gave them a stimuli- warm or painfully hot and asked if they thought it was painful or not



* Posterior insula exhibited neural patterns in line with detecting pain intensity – receives afferents from body (interoception)
* Anterior insula exhibited neural patterns in line with predictive coding of pain - integrates information with other modalities
* These predictions are modality specific (pain) rather than a general expectation of an aversive outcome

psychiatric disorders and interoception

* proposed individuals who focus on their bodily sensations more may exhibit greater interoceptive prediction signals
* prediction of aversive states may trigger aversive states
* if someone experiences a panic attack in a lecture they then predict that will happen again due to the association, so then they worry and might lead them to avoid lectures
* might be fine walking into the lecture and HR is normal but because you’re predicting that it isn’t you have this big mismatch
* in order to rectify the mismatch your insula cortex might increase your heart rate to meet the prediction
* you basically predict yourself into the panic

Uchida et al (2008)- study comparing HC and panic disorder

* Increased grey matter volume in the insula in PD compared to HC
* Misattribute symptoms as a heart attack despite physiologically heart is healthy
* Catastrophising normal things, you’re actually fine physiologically but think you’re having a heart attack
* Misattribute events occurring at the same time as associated (triggers) – avoidance behaviours
* Anxiety over bodily signals, maladaptive behaviours and cognitions

Pauli et al (1991)- study of patients with panic attacks compared to HC

* monitored 28 patients with panic attacks and 20 healthy controls
* 24 hours ECG monitoring (HR/heart rate)
* asked ppts to note cardiac perceptions, anxiety and activities (not sports)

results:

* found similar variations in actual heart rate, everyone has these variations in HR it wasn’t that people with PD had varying HR more
* however, Panic attacks patients rated higher anxiety associated with cardiac perceptions
* Anxiety experienced in relation to heart perceptions related to subsequent shortening between heartbeats- so difference in HR once the person with PD because aware of it
* Heartbeat variations prior to cardiac perception not related to anxiety level
* supports notion that its a misattribution of a normal change in HR
* vicious cycle

beliefs in panic disorder

* all begin with maladaptive beliefs
* A change in the internal state of the individual, which may be due to an increased attentional biases toward threat
* Beliefs used to interpret internal body signals
* sets up a bodily prediction error
* External cues or internal thought generate anticipation of aversive body states that - sets up a body prediction error

self processing in panic disorder

* relating these maladaptive states to insula activity
* Worrying is associated with increased activity in prefrontal cortex, striatum, and insula (Hoehn-Saric et al. 2004)
* Guilt was linked to left anterior insular cortex (Shin et al. 2000)
* Sad self-relevant autobiographical memories linked to activation the ventral insula (Liotti et al. 2000)

Somatic Error Hypothesis

* Psychiatric disorders, such as anxiety, may be a result of mismatches between anticipated and incoming bodily signals
* **Somatic error** – the body state predicted by the brain differs from the afferent signals of the body state
* Compensatory behaviour = psychiatric symptomology
* how we compensate for the error are the symptoms of disorders
* Actions to reduce the discomfort and distress (e.g. agoraphobia and other maladaptive behaviours )

Dresler et al (2011)- fMRI from when patients had a panic attack in the scanner

* 2 Patients suffering panic attacks in the fMRI scanner leading to termination of testing (must have happened in another study)
* One removed beginning of discomfort
* One removed once in full panic attack


* Patient removed from the scanner first only observed a decrease in prefrontal cortex activity during their discomfort- top-down onset of panic attacks


* ppt who has the Full panic attack demonstrated significant insula and amygdala activity
* Insula activity linked to feelings of discomfort
* Amygdala related to feelings of panic

DSM-5 criteria for anorexia nervosa

* restriction of energy intake to be less than what the body needs leading to sig low body weight relative to age, sex and developmental trajectory
* intense fear of gaining weight
* disturbance in they way the body weight or shape is experienced, self objectification
* reduced capacity to soothe self or empathise with others
* emotionally inhibited- flattening of affect
* depression
* alexithymia- can’t describe or recognise emptional distress

interoception and AN

* Intuitive link between interception and eating disorders in general like AN (through hunger association )
* eating dirsorder inventory asks about intercopetive asepcts e.g. “ I get confused about what emotion I am feeling” or “I feel bloated after a normal sized meal”
* however, Interoceptive sensibility and Interoceptive accuracy were not correlated in patients with AN

interceptive sensibility and AN

**Jenkinson et al 2018**

* Meta-analysis across all EDs
* EDI interoceptive scale
* anorexia and bulimia - significant deficits in interoceptive sensibility
* Lower BMI = greater deficits

interoceptive accuracy and AN

**Pollatos et al (2008)**

* The heartbeat perception task was performed using four intervals of 25 s, 35 s, 45 s and 100 s
* asked to count each heart beat across these different intervals
* AN performed worse than HC
* Patients with AN exhibit a generally reduced capacity to accurately perceive bodily signals
* Less intense emotional experiences in many everyday situations
* Potential importance of interoception in the pathogenesis of AN

interoceptive awareness and AN

**Lernia et al (2018)**

* Case study 1 AN, against 4 HCs
* Heart beat counting task (interoceptive accuracy)
* Confidence measure (Interoceptive Awareness – when compared to accuracy score)
* MAIA (Interoceptive sensibility)
* results found a trend towards lower interoceptive accuracy
* but they had an enhanced confidence of interoception- 93/100 in their belief
* Less able to regulate distress and distract from bodily signals, reduced body trust
* Detachment between ability to perceive the body and the awareness- think they are aware of their bodily sensations but clearly they are not

neural indicators of interoception and AN

lutz et al (2019)

* Heartbeat counting task during EEG HEP
* interoceptive sensibility- found no diff in confidence but difference in interpretation
* interoceptiove accurary (heartbeat counting)- non sig trend for better IA in AN
* higher HEP amplitude
* \

evidence for insula involvement in ED

Murialdo et al (2007)

* regulatory control of HR variability in ED using tilt table test


* Lie on a table that adjusts your body position from horizontal to vertical to simulate standing up
* Monitor changes in heart rate and blood pressure
* patients had lower blood pressure than HC
* sympathetic cardiac activity did not increase from lying to standing as it did in HC
* cardian abnormalities are simialr in AN and BN even though only AN are emaciated (so not linked to BMI)

Fisher et al (2016) - AN and CBT

* 15 AN patients
* given CBT with special attention to maladaptive emotional processing
* tested at the beginning, 4-6 weeks in and then again at the end of therapy, measuring interoceptive accuracy
* found BMI increased and depression decreased
* but interceptive accuracy and sensibility didn’t improve so recovery was small and inconsistent
* interception doesn’t improve with other symptoms which makes ED so difficult to recover from

Kerr et al (2016)- AN patient study with interoceptive trials, anxious rumination trials and exteroceptive trials

* Interoceptive attention task - focused on sensations in their heart, stomach, and bladder.
* Anxious Rumination- given primes to trigger ruminations
* Exteroceptive trials- rate intensity of colou


* Activity in the dorsal mid-insula was reduced in AN patients during stomach interoception
* Activity in the anterior insula was relatively higher in AN patients during heart interoception
* shows dissociation- can’t use HR tasks for everything
* increased activation during anxious rumination in the dorsal mid-insula, in stomach interception correlated with anxiety
* abnormal dorsal mid-insual activity during gastric interception

prediction mechanisms in AN

* Somatic error hypothesis and other prediction theories suggest inaccurate interoceptive predictions in psychiatric disorders
* Gastric abnormalities not sensory input but interpretation
* Unclear mechanisms for how interoception underlies a complex disorder such as AN

interoception and eating - HC vs AN

* double blind trial either given saline of isoproterenol (an adrenoceptor agonist)
* measured pre and post meal HR and breathing and their interoceptive detection
* AN experienced more intense cardiorespiratory sensations before consuming a meal
* AN more likely to report interoceptive experiences in all conditions, particularly in low arousal condition
* AN could be anticipating meal eating → anxiety

Interoception and restrictive symptoms of AN

* AN patients can’t use internal signals to perceive physical changes of weight loss
* don’t recognise satiety or hunger
* continue to restrict food despite emaciation- not getting signals from their body that something is wrong
* intense fear of gaining weight
* can’t self regulate emotions

Objectification theory of AN

**Fredrickson & Roberts (1997)**

* Bodies are viewed and evaluated based on appearance
* The body as an object (sexual)
* The body and sexuality are separate from the person
* Objectifying others and objectifying the self
* The more you self objectify, the more important body appearance and deviations form social ideas is important for self worth

AN and the insula

* Insular cortex is a neural hub connecting interoceptive signals with other modalities including exteroception, cognition and emotion
* Potential mechanism in AN of interoception underpinning many of the key symptomology in AN
* As it isn’t addressed in treatments high chance of relapse

interoception and anorexia relapse

* Interoceptive prediction errors may lead to further errors and negative affective that is unresolved lead to risk of relapse particularly with certain triggers
* Weight restored AN patients to show differences in insula activity when anticipating food as well as decreased response to the taste of food stimuli
* Both exteroceptive and interoceptive bodily symptoms continue following weight restoration

\