Sensitivity as the common denominator of ADHD, autism and PTSD — an overview of the concept, the evidence and the diagnostic consequences

1. Why the question arises

Anyone who spends enough time assessing people with ADHD, with an autism spectrum disorder and with post-traumatic stress disorder makes an observation that appears in no diagnostic manual: they report similar things. Light that is too bright. Sounds that refuse to recede into the background. Surroundings that do not arrive filtered but unbraked. And, less often noticed, the reverse: stimuli that do not arrive at all, pain that is noticed late, a bodily signal that fails to appear.

Three diagnoses, three aetiologies, three sets of guidelines — and one recurring motif. It is no accident that the Biomarker Workshop in December places these three conditions alongside a fourth term that is not a diagnosis at all: sensitivity. The question at stake is this: are we describing the same phenomenon three times under different names — or are we mistaking a superficial resemblance for a common cause?

The answer, to say it upfront, is not settled. But the evidence is richer than the everyday routine of categorical diagnosis would suggest, and it is contradictory enough to deserve a careful account. The following text attempts that account: what is meant by sensitivity, what is empirically established, where the evidence is thin, and what follows from it for a diagnostic approach based on measurement.

2. Two research traditions that use the same word

The first difficulty is conceptual, and it is rarely named. Two largely separate literatures lead to the word “sensitivity”, and they hardly cite each other.

2.1 Personality research: environmental sensitivity

The first strand comes from personality and developmental psychology. It describes sensitivity as a temperament trait of the healthy population: the capacity and the tendency to process environmental stimuli more deeply. In the literature the construct is called Sensory Processing Sensitivity (SPS) and is understood as part of a broader framework of Environmental Sensitivity.

The authoritative critical review is by Greven and colleagues (Neuroscience and Biobehavioral Reviews, 2019). Its central claim is remarkably symmetrical: sensitivity increases the risk of stress-related problems in adverse environments — and it increases the benefit drawn from positive, supportive experiences. In this understanding, sensitivity is not a vulnerability but a heightened receptivity in both directions. The same paper, however, also names the field’s main problem without embellishment: there is a lack of reliable, objective measurement, and the boundary to neighbouring traits remains unresolved.

On the distribution in the population, data from latent class analyses are available. Lionetti and colleagues (Translational Psychiatry, 2018) found not two groups — sensitive and not sensitive — but three, with roughly one third each in the low- and the high-sensitivity groups and around forty percent in the middle. The common images for this — dandelion, tulip, orchid — are catchy and at the same time carry an error: they suggest sharply defined types where what is actually at hand are segments of a continuous distribution.

That there is something biological about the trait is supported by twin data. Assary and colleagues (Molecular Psychiatry, 2021) found a heritability of around 47 percent in more than 2,800 seventeen-year-old twins — a value in the same range as other personality traits. Sensitivity is thus neither purely acquired nor purely innate, but both, roughly in equal halves.

2.2 Clinical research: sensory reactivity

The other strand comes from clinical developmental research and is interested not in a temperament but in a symptom. Here the subject is sensory over- and under-reactivity: the child who cannot bear the label in the collar of a shirt, the adult who cannot work in an open-plan office, the person who responds to a slammed car door with a startle reaction out of all proportion to the trigger.

In this tradition, sensitivity is anchored diagnostically. Since DSM-5, hyper- or hyporeactivity to sensory input — or an unusual interest in sensory aspects of the environment — has been among the criteria for autism spectrum disorder. In PTSD the same motif appears under a different name: hyperarousal, exaggerated startle, hypervigilance. And in ADHD it does not formally appear in the criteria at all, yet has long been described clinically.

2.3 Why the distinction matters

The two traditions measure things that are related but not identical. One uses self-report questionnaires to assess a personality trait in the general population; the other assesses, usually by informant report, a clinically abnormal feature. Anyone who lumps both together under the word “sensitivity” produces spurious agreement.

Hence the convention used below: environmental sensitivity for the temperament trait, sensory reactivity for the clinical feature, and sensitivity as the umbrella term for the presumed shared dimension — with the explicit caveat that this shared dimension is a hypothesis and not a finding.

3. What speaks for a common denominator

Three recent observations make the hypothesis attractive.

First: hyperreactivity does not respect diagnostic boundaries. Dwyer and colleagues (Journal of Attention Disorders, 2025) studied 492 adults divided into four groups: ADHD only, autistic only, autistic with ADHD, and a comparison group. Auditory hyperreactivity was markedly elevated in both neurodivergent groups — not in one or the other, but in both. The study also shows a circle familiar from clinical practice: stimulus sensitivity is linked to anxiety, anxiety to hypervigilance, and hypervigilance in turn to stimulus sensitivity. A second finding from the same paper is noteworthy: psychoacoustic measurements agreed only moderately with self-report. What people experience as an unbearable stimulus is evidently more than a property of the stimulus itself.

Second: sensory profiles form groups of their own, cutting across the diagnoses. Brandes-Aitken and colleagues (Scientific Reports, 2024) studied 117 children aged eight to twelve with mixed neurodevelopmental conditions — autism, ADHD, anxiety disorders, sensory processing difficulties. A latent profile analysis yielded five clusters: typical processing, a mixed profile, over-reactivity, sensory seeking and under-reactivity. These clusters could be linked to behavioural characteristics — the over-reactive group showed elevated anxiety scores, the seeking and the under-reactive groups elevated ADHD scores — but they did not coincide with the diagnoses. The sensory axis cuts across the diagnostic one.

Third: the genetic components overlap. Assary, Oginni and colleagues (Molecular Psychiatry, 2024) examined, in more than 2,800 twin pairs aged 15 to 17, the relationship between environmental sensitivity on the one hand and emotional problems, autistic traits and wellbeing on the other. The associations were moderate, and around two thirds of them could be attributed to shared genetic factors. Decisive for the present question, however, is one detail: the aesthetic sub-dimension of sensitivity — receptivity to music, art, natural beauty — correlated positively with wellbeing and not with emotional difficulties.

This last finding is the strongest argument against hastily equating sensitivity with disorder. Sensitivity is evidently not a single block but breaks down into sub-aspects that point in opposite clinical directions.

4. The special case of PTSD: innate or acquired?

In ADHD and autism, the subject is a development that takes a different course. In PTSD, the subject is an event that leaves something behind. If sensitivity is to be the connecting dimension, it has to explain both — and this is precisely where it becomes interesting.

The obvious assumption would be that PTSD raises sensitivity: that after a trauma, a nervous system tips permanently into a state of heightened alertness. There is electrophysiological evidence for this assumption, but it is weaker than its popularity suggests. The most frequently cited measure is sensory gating: if two identical click stimuli are presented in quick succession, a healthy brain suppresses its response to the second — a simple measure of the ability to filter out redundancy. Reduced suppression in PTSD has been reported repeatedly, among others by Neylan and colleagues (Biological Psychiatry, 1999) and, in a magnetoencephalographic variant, by Hunter and colleagues (Psychiatry Research, 2011). The second study included seven patients. That is the state of affairs: individual, consistently small studies pointing in the same direction, without a robust meta-analysis to rely on.

The situation with resting EEG is similar: there are replicated group differences — changes in the alpha band, reduced signal complexity, altered microstate dynamics — but no classification accuracy that would suffice for a statement about an individual case. [Insert cross-reference to the PTSD post here once it is online.]

The theoretically more exciting question runs in the opposite direction: not whether the trauma raises sensitivity, but whether a pre-existing sensitivity raises the likelihood of developing PTSD after an event. That is exactly what the model of differential susceptibility would predict, as formulated by Belsky and Pluess (Psychological Bulletin, 2009): sensitive people respond more strongly to their environment — for the worse under adversity, for the better under favourable conditions. Applied to PTSD, this would mean that the same trait that raises the risk of illness after a trauma would also raise responsiveness to good treatment.

That is a testable hypothesis and, as far as we know, one that has not been convincingly tested for PTSD. It would be worth it.

5. Process models

Why should a brain take in too much or too little in the first place? Two families of models are under discussion; both are theories, not established mechanisms.

The Intense World Theory of Markram and Markram (Frontiers in Human Neuroscience, 2010) describes autism as the consequence of locally hyperactive and excessively plastic neural circuits. The world would then arrive not attenuated but too intensely; withdrawal and routines would not be a core deficit but a sensible response to being overwhelmed. The model has noticeably changed the debate and is at the same time empirically contested — it rests in essential parts on an animal model and accounts only inadequately for under-reactivity, which is just as common clinically.

Predictive processing sets in at a more abstract level. Van de Cruys and colleagues (Psychological Review, 2014) propose that it is not stimulus intensity itself that is altered, but the weighting of prediction errors: a brain that assigns too much significance to every deviation from expectation cannot file the inessential as inessential. The advantage of this framework is that it can capture over- and under-reactivity as two manifestations of the same regulatory problem and that it can in principle be tested electrophysiologically — prediction errors have measurable correlates. The disadvantage: the model is flexible enough to accommodate almost any finding after the fact. That kind of explanatory power is a warning, not evidence.

6. What follows for diagnostics

If one assumes that a shared dimension lies beneath the four labels, three consequences follow for diagnostic practice.

First, comorbidity becomes partly an artefact. If a continuous dimension is cut by several categorical grids, a person at the upper end of that dimension will very probably receive several diagnoses — not because they have several illnesses, but because several grids respond at the same place. The high comorbidity between ADHD, autism and anxiety disorders is compatible with this reading. It does not prove it.

Second, the questionnaire as the sole instrument is overstretched. All of the evidence reviewed in section 3 rests on self-report or parent report. The finding by Dwyer and colleagues that psychoacoustic measurement and self-report agree only moderately is therefore not a marginal result but a fundamental methodological problem: so far we mostly measure how people describe their sensitivity.

Third, this creates the need for a second, independent approach. This is exactly where brain function analysis comes in. From a resting EEG and from event-related methods, indices can be computed that describe aspects of stimulus processing without anyone having to report them: arousal and vigilance trajectories, measures of stimulus suppression, complexity and entropy measures, the dynamics of microstates. Several of these quantities are available in the HBImed system as database-referenced indices.

What is expressly not being claimed here: that any of these indices is a validated biomarker for sensitivity. No such biomarker exists. What they achieve is more modest and still not little — they describe where a particular brain lies in comparison with many others, and they do so by a route that is independent of the examined person’s own narrative. Where this route and the clinical assessment agree, the judgement becomes more robust. Where they diverge, that is not a contradiction but a lead.

7. What the data do not support

Four limitations are serious enough not to be relegated to a footnote.

The constructs are not cleanly separated. Environmental sensitivity correlates with neuroticism, with introversion, with anxiousness. Whether it is a distinct trait or a renaming of known dimensions has not been conclusively resolved — the review by Greven and colleagues names precisely this as an open question. Anyone who declares sensitivity to be the connecting dimension must first show that it is more than a new word.

The samples are small and the assessment is subjective. 117 children in a profile analysis, 492 adults in a questionnaire study, seven patients in a gating study: these are leads, not foundations for clinical decisions. Almost all findings, moreover, come from Western, predominantly well-educated samples.

Group differences are not individual diagnoses. For all the electrophysiological findings reviewed here, what was set out in detail for PTSD applies: the distributions overlap, and a difference in means says little about where a single person lies.

And the yardstick remains circular. Any procedure that is validated against the clinical diagnosis cannot, by construction, be better than that diagnosis. That is the usual first step — but the added value of measurement-based diagnostics only shows where measurement and judgement diverge and it can be established who turns out to be right. For sensitivity as a dimension, that demonstration is still outstanding.

The foundation’s working hypothesis, expressly unproven

We consider it plausible that a substantial share of the burden of psychiatric illness stems from a mismatch between individual sensitivity and environmental demands — that is, that it is not the sensitivity that is the disorder, but the environment that makes no provision for it. We have no quantitative basis for this assessment. It is formulated as a research hypothesis and not as a finding; anyone wishing to test it would need epidemiological data that do not exist in this form.

8. Outlook

The Biomarker Workshop on 10 December 2026 in Zurich places ADHD, autism, PTSD and sensitivity side by side because the four poles overlap in lived experience and because categorical diagnostics cannot represent that overlap. The path “from category to dimension” is not a rhetorical one but a methodological one: it demands measures that are continuous, that can be repeated, and that do not depend on how well someone can give an account of themselves.

Whether sensitivity is the dimension that carries this, we do not know. That the existing categories cut apart something that belongs together, we consider well founded.

References

Assary, E.; Zavos, H. M. S.; Krapohl, E.; Keers, R.; Pluess, M. Genetic architecture of Environmental Sensitivity reflects multiple heritable components: a twin study with adolescents. Molecular Psychiatry, 2021 (online 2020). doi:10.1038/s41380-020-0783-8
Assary, E.; Oginni, O. A. et al. Genetics of environmental sensitivity and its association with variations in emotional problems, autistic traits, and wellbeing. Molecular Psychiatry, 2024, 29(8). doi:10.1038/s41380-024-02508-6
Belsky, J.; Pluess, M. Beyond diathesis stress: differential susceptibility to environmental influences. Psychological Bulletin, 2009.
Brandes-Aitken, A. et al. Sensory processing subtypes relate to distinct emotional and behavioral phenotypes in a mixed neurodevelopmental cohort. Scientific Reports, 2024, article 29326.
Dwyer, P.; Williams, Z. J.; Lawson, W.; Rivera, S. M. A Trans-Diagnostic Investigation of Attention and Diverse Phenotypes of “Auditory Hyperreactivity” in Autism, ADHD, and the General Population. Journal of Attention Disorders, 2025.
Greven, C. U.; Lionetti, F.; Booth, C.; Aron, E. N. et al. Sensory Processing Sensitivity in the context of Environmental Sensitivity: A critical review and development of research agenda. Neuroscience and Biobehavioral Reviews, 2019, 98, 287–305. doi:10.1016/j.neubiorev.2019.01.009
Hunter, M. et al. Lateralized abnormalities in auditory M50 sensory gating and STG cortical thickness in PTSD. Psychiatry Research, 2011, 191(2).
Lionetti, F.; Aron, A.; Aron, E. N.; Burns, G. L.; Jagiellowicz, J.; Pluess, M. Dandelions, tulips and orchids: evidence for the existence of low-sensitive, medium-sensitive and high-sensitive individuals. Translational Psychiatry, 2018, 8(1).
Markram, K.; Markram, H. The Intense World Theory — A Unifying Theory of the Neurobiology of Autism. Frontiers in Human Neuroscience, 2010, 4:224.
Neylan, T. C. et al. Sensory gating in chronic posttraumatic stress disorder: reduced auditory P50 suppression in combat veterans. Biological Psychiatry, 1999.
Van de Cruys, S.; Evers, K.; Van der Hallen, R.; Van Eylen, L.; Boets, B.; de-Wit, L.; Wagemans, J. Precise Minds in Uncertain Worlds: Predictive Coding in Autism. Psychological Review, 2014, 121(4), 649–675.