The Unreliable Future Hypothesis

Childhood Adversity, Interruption-Calibrated Present-Focused Reward Bias, and ADHD-Inattentive Phenotypic Overlap

Author: Mik Idrizovic
Date: July 2026 (revised August 2026)
Status: Hypothesis-generating theoretical proposal; not clinical guidance

No empirical work is reported in this paper. The framework is offered as a falsifiable research proposal for improving mechanistic precision in ADHD-I assessment among trauma-exposed adults.

Abstract

Attention-Deficit/Hyperactivity Disorder, Inattentive Presentation (ADHD-I), is a clinically useful but mechanistically heterogeneous phenotype. Current diagnostic criteria identify recurring patterns of impairment; they do not, by themselves, identify the developmental computation that produced those patterns. This paper proposes the Unreliable Future Hypothesis: in a subset of individuals who meet full ADHD-I criteria, particularly those with substantial childhood adversity, inattentive symptoms may be partly generated by an adversity-calibrated present-focused reward bias. In such individuals, the future may be implicitly encoded as unreliable, unsafe, or insufficiently bankable, making delayed goals less motivationally potent than immediate cues, relief, threat signals, or short-horizon control opportunities.

The proposal is positioned as a developmental specification of the motivational pathway already described in dual- and triple-pathway models of ADHD (Sonuga-Barke, 2002, 2003; Sonuga-Barke et al., 2010), not as a rival to them. It does not claim that trauma generally causes ADHD, that primary neurodevelopmental ADHD is misdiagnosed trauma, or that stimulant response can neatly distinguish “real” from “acquired” ADHD. Rather, it argues that the ADHD-I phenotype may be reachable through multiple developmental routes, and that one such route is calibration to interruption risk: the learned probability that a promised reward will fail to arrive. Drawing on formal modelling work showing that present-oriented behaviour is adaptive under some environmental conditions and not others (Fenneman et al., 2022), the framework replaces the loose construct of “unpredictability” with a specific, measurable contingency.

The framework generates testable predictions involving delay discounting, salience-gated attention, uncertainty sensitivity, subjective stimulant effects, treatment moderation, and — most decisively — experimentally manipulated environmental reliability. It further proposes a staged empirical program beginning with secondary analysis of existing prospective cohorts, followed by mechanistic phenotyping, a carefully controlled stimulant-challenge pilot, and a treatment-matching study. Throughout, the proposal is constrained by three known threats to inference that any adversity-and-ADHD claim must survive: poor agreement between prospective and retrospective adversity measures, substantial familial and genetic confounding of adversity-ADHD associations, and elevated symptom over-reporting indices in high-adversity ADHD referrals. The hypothesis is offered as a falsifiable framework for increasing diagnostic precision and reducing mechanism-blind treatment matching.

Keywords: ADHD-I; childhood adversity; interruption risk; delay discounting; delay aversion; reward processing; present-focused reward bias; mechanistic heterogeneity; PTSD; stimulant response; precision psychiatry

1. Introduction: The Future Must Be Credible

ADHD-I is often described as difficulty sustaining attention, organizing behavior, completing tasks, and following through on delayed goals. That description is clinically useful. It is also mechanistically thin. A person can fail to sustain attention for many reasons: primary neurodevelopmental differences in executive control and reward processing; sleep disruption; anxiety; depression; dissociation; current threat monitoring; learned helplessness; low perceived agency; or a history in which delayed rewards repeatedly failed to arrive. Similar behavioral outputs can have different developmental causes.

The claim advanced here is simple but potentially consequential: the future is not equally motivationally real for everyone. Future reward only competes with immediate relief when the nervous system implicitly expects that future to be reachable, safe, and worth investing in. In environments marked by threat, scarcity, inconsistent caregiving, abandonment, humiliation, or chronic unpredictability, a child may learn that delayed reward is a weak bet. The adaptive move is to prioritize the immediate field: danger, opportunity, relief, emotional shifts, and controllable short-horizon actions.

In adulthood, that same calibration can look like ADHD-I. The person may appear distractible, avoidant, inconsistent, forgetful, unmotivated, or unable to initiate effort for abstract goals. But the deeper problem may not be that attention is globally broken. It may be that attention and motivation have been optimized around a world in which the future did not reliably keep its promises.

This paper names that possibility the Unreliable Future Hypothesis. The proposed mechanism is Adversity-Calibrated Present-Focused Reward Bias (APRB): a relatively stable but potentially modifiable attentional-motivational pattern in which immediate, salient, safety-relevant, or relief-producing cues dominate delayed, abstract, or low-salience goals. The framework is designed to be testable, not merely evocative. Its value depends on whether it can predict measurable differences within people who already meet ADHD-I criteria.

2. ADHD-I as Phenotype, Not Mechanism

ADHD is a well-supported neurodevelopmental disorder. The World Federation of ADHD consensus statement summarizes extensive evidence that ADHD is associated with genetic and environmental risks, measurable impairment, clinically meaningful outcomes, and evidence-based treatments (Faraone et al., 2021). Preserving that foundation matters. The Unreliable Future Hypothesis does not argue that ADHD is culturally invented, reducible to trauma, or usually misdiagnosed. Those claims are too broad, too blunt, and not supported by the available literature.

The more defensible claim is that ADHD is heterogeneous. Reviews of ADHD heterogeneity emphasize variability in etiological risk, developmental course, cognitive profile, comorbidity, treatment response, and neural markers (Luo et al., 2019). In other words, the same diagnostic label can collect people whose outward symptoms overlap while their causal pathways differ. That is not a failure unique to ADHD. It is a recurring problem in psychiatry: diagnoses often begin as reliable descriptions before becoming mechanistic explanations. The description then starts pretending it is the mechanism. It is not.

2.1 Where this framework sits

Heterogeneity is not a new observation, and this proposal should be read as a specification within an existing research tradition rather than a challenge to it. Against the dominant executive-dysfunction account (Barkley, 1997), Sonuga-Barke and colleagues recast a subset of ADHD behavior as a motivational style: delay aversion, an active tendency to escape or avoid the negative affect evoked by waiting (Sonuga-Barke et al., 1992; Sonuga-Barke, 2002). The dual pathway model, and its later triple pathway extension incorporating temporal processing, hold that executive dysfunction, delay aversion, and timing constitute dissociable routes to the same diagnostic label, with different individuals affected in different domains (Sonuga-Barke, 2002, 2003; Sonuga-Barke et al., 2010; Coghill et al., 2014). Notably, Sonuga-Barke’s (2003) elaboration explicitly placed these pathways in developmental context, invoking person–environment correlation and individual adaptation to developmental constraint.

What those models do not specify is where a delay-averse motivational style comes from in any given person. The neurobiological account — altered signalling of delayed reward in ventral fronto-striatal circuitry — describes the machinery without describing what the machinery learned. The Unreliable Future Hypothesis proposes one candidate answer for a subset of cases: the motivational pathway is populated in part by people whose reward-delay expectancies were calibrated by environments in which delivery of promised reward was genuinely unreliable.

This is a narrower and considerably more falsifiable claim than “trauma can look like ADHD.” It also carries a cost the broader claim avoids: it must show that the calibration story explains variance the existing motivational pathway does not already explain on its own.

ADHD-I, in particular, may be vulnerable to mechanistic compression because the phenotype is defined largely by failures of sustained attention, organization, working memory, task completion, and persistence. Those behaviors are downstream of many systems: cognitive control, reward valuation, sleep, arousal, affect regulation, threat detection, self-concept, agency, and time horizon. A person can look inattentive because delayed reward does not hold sufficient motivational weight. That possibility is central to the present framework.

3. The Core Construct: Adversity-Calibrated Present-Focused Reward Bias

Adversity-Calibrated Present-Focused Reward Bias refers to a developmental pattern in which attention and motivation become preferentially organized around present-moment salience. The term “reward” is used broadly here. It includes pleasure, relief, safety, emotional discharge, control, certainty, social reassurance, novelty, and avoidance of anticipated pain. In adversity, relief can become as powerful a reward as pleasure.

APRB is proposed to include five interlocking components:

Temporal compression: delayed rewards are devalued because the future is implicitly modeled as unreliable, unsafe, or inaccessible.

Salience-gated attention: attention is captured by cues that feel immediate, emotionally charged, threatening, novel, or capable of producing rapid state change.

Uncertainty vigilance: ambiguous or unstable conditions trigger scanning, rumination, checking, avoidance, or emotional reactivity.

Relief learning: behaviors that quickly reduce distress are reinforced, even when they undermine long-term goals.

State-contingent performance: functioning improves under urgency, danger, social accountability, novelty, or immediate payoff but collapses during low-salience, delayed-reward tasks.

This construct reframes a familiar clinical observation. Some high-adversity adults can perform impressively in crisis, respond quickly to emotionally meaningful problems, and show intense focus under urgency, yet struggle to answer an email, complete paperwork, maintain routines, or advance a goal whose payoff is distant and abstract. The issue is not simply capacity. It is motivational credibility. The nervous system allocates resources toward what it has learned matters now.

In childhood adversity, that allocation may be rational. If caregiver responses are inconsistent, if safety changes without warning, if punishment arrives unpredictably, if promised rewards disappear, or if long-term planning is repeatedly overturned by family chaos, the child learns that long-horizon investment is risky. Under those conditions, present-focus is not pathology. It is local intelligence. The pathology emerges later, when an adaptation to volatility is carried into environments that demand trust in delayed, invisible, bureaucratic, or self-generated reward.

3.1 Which unpredictability: the interruption-risk specification

The intuition that present-focus is a rational response to an unpredictable world is widely shared and, in its loose form, wrong often enough to be dangerous. Fenneman, Frankenhuis, and Todd (2022) reviewed and integrated thirty formal models of impulsive behavior across disciplines. Their conclusions were less flattering to the folk version than the folk version assumes. Impulsive behavior is not uniformly adaptive in harsh or unpredictable environments; whether it pays depends on exactly how harshness, unpredictability, and impulsivity are each defined. Impulsivity may be adaptive when resource quality is low or high but less so when it is moderate. Extrinsic events — the mortality-risk framing that dominates much of the life-history literature — turn out to exert only limited influence on whether impulsivity pays. And impulsive behavior is adaptive when resource encounters are likely to be interrupted: when the chance that a resource becomes unavailable before it can be obtained is high.

That third condition is the one that matters here, and it is the formal analogue of a promise not kept. A caregiver who says “later” and does not return, a household in which a plan can be cancelled without warning, a reward schedule in which waiting is frequently punished by disappearance rather than by mere delay — these are interruption risk, not generic chaos.

APRB should therefore be specified as calibration to interruption risk rather than to adversity in general, to stress in general, or to unpredictability in the loose sense of “things were chaotic.” The distinction is not cosmetic; it generates different predictions. It implies that ACE count — which aggregates threat, deprivation, loss, and household dysfunction without regard to reward-delivery contingency — should be a comparatively weak predictor of APRB markers, while measures indexing broken promises, inconsistent contingency, caregiver unreliability, and plan disruption should be comparatively strong ones. It also predicts a dissociation that a purely stress-based account does not: harshness indexed by danger or scarcity should matter less for reward-timing calibration than unreliability indexed by non-delivery.

This specification is what gives the framework something to be wrong about. A model that predicts “adversity makes people present-focused” is nearly unfalsifiable and largely redundant with existing literature. A model that predicts which dimension of adversity does the work, and which does not, can fail cleanly.

4. Existing Plausibility: What the Literature Already Supports

4.1 Adversity and ADHD symptom entanglement

Associations between adverse childhood experiences and ADHD diagnosis or symptom reporting are well documented. Brown et al. (2017) found a graded association between ACE exposure, ADHD diagnosis, and ADHD severity in children. Zhang et al. (2022), in a systematic review and meta-analysis, likewise reported an association between ACEs and subsequent ADHD. These findings do not prove that adversity causes ADHD. They do show that the boundary between developmental adversity and ADHD symptom expression is clinically important enough to require direct measurement rather than casual assumption.

More recent evaluation work sharpens the problem, and does so in a way that partly cuts against a naive reading of the present framework. Acabal et al. (2026) examined 264 adults referred for ADHD evaluation and found that ACE exposure significantly predicted retrospectively reported childhood ADHD symptoms — particularly inattention and impulsivity — but did not predict current symptom reporting, while active PTSD symptoms showed a different pattern. Their conclusion supports a basic premise of this paper: ADHD assessments should incorporate trauma history and multi-method evaluation. But their specific finding is a caution as much as a support. If adversity primarily shapes how adults narrate their childhood rather than how they function now, then part of the observed adversity–ADHD association is a recall-and-report phenomenon rather than a mechanism phenomenon. Any framework proposing an adversity-calibrated mechanism must be able to distinguish itself from that alternative, and must do so with measures that do not rely on the same retrospective self-report channel.

4.2 Threat, deprivation, and developmental calibration

The Unreliable Future Hypothesis is not a generic “stress causes inattention” proposal. It is a calibration proposal. McLaughlin, Sheridan, and Lambert’s dimensional model distinguishes threat from deprivation as separable dimensions of early adversity with distinct implications for neural development (McLaughlin et al., 2014). Teicher and Samson (2016) similarly review evidence that childhood maltreatment can produce enduring neurobiological effects and may create clinically meaningful ecophenotypic variation within psychiatric categories.

Adaptive calibration models of stress responsivity provide a broader developmental logic: organisms tune stress, attention, and behavioral systems to local environmental demands (Del Giudice et al., 2011; Ellis et al., 2017). In harsh or unpredictable environments, shorter time horizons and heightened sensitivity to immediate contingencies may be adaptive. This does not romanticize adversity. It simply refuses to treat every post-adversity behavior as irrational noise.

4.3 Unpredictability, delay discounting, and time horizon

A central prediction of APRB is steeper delay discounting: future rewards lose motivational value more rapidly relative to immediate rewards. Early research linked family unpredictability with future discounting and risk-related behavior (Hill et al., 2008). Martinez et al. (2022) more directly connected perceived childhood unpredictability with delay discounting and adult externalizing behavior, while controlling for perceived harshness and demographic factors. The important point for this framework is specificity: unpredictability may matter not only because it is unpleasant, but because it teaches a model of time in which delayed payoff is structurally suspect.

Acheson et al. (2019) found that early life adversity was associated with increased delay discounting in the Family Health Patterns Project. That finding gives APRB a measurable behavioral anchor. The proposed construct is not simply a metaphor about living in the present. It predicts quantifiable differences in intertemporal choice.

The construct-validity caveats here are severe and are treated at length in Section 12. Delay discounting is transdiagnostic rather than disorder-specific (Amlung et al., 2019), the ADHD–discounting effect is modest in absolute terms (Jackson & MacKillop, 2016; Marx et al., 2021), and standard intertemporal-choice tasks have been criticized for abstracting away exactly the contextual parameters this framework claims are doing the work (Bailey et al., 2021; Lempert et al., 2019). Delay discounting can serve as one anchor for APRB. It cannot serve as its definition.

4.4 Reward processing after early adversity

Reward-system research also supports the plausibility of the model. Boecker et al. (2014), in a prospective EEG-fMRI study, reported that early life adversity was associated with altered reward processing in young adulthood, including reduced activation during reward anticipation and increased activation during reward delivery. Hanson et al. (2021) review broader human and animal evidence that early life stress can alter reward-circuit development and motivated behavior. Kasparek et al. (2023) likewise examine reward processing as a potential mediator or moderator linking childhood adversity with later psychopathology.

These literatures do not establish APRB as a validated subtype. They do, however, converge on a plausible mechanism: early adversity may alter how the brain anticipates, values, pursues, and receives reward. If ADHD-I symptoms are partly driven by failures to sustain behavior toward delayed reward, then adversity-related reward calibration becomes directly relevant to ADHD-I phenotypic overlap.

4.5 Experimental evidence that reliability, not just adversity, moves the behavior

The strongest evidence for the core claim is not correlational. Kidd, Palmeri, and Aslin (2013) manipulated environmental reliability directly before administering the classic delay-of-gratification task to preschoolers. Children who had just experienced an adult break a promise waited roughly three minutes; children who had experienced the same adult keep one waited roughly twelve. The manipulation was brief and the effect was large. Michaelson and Munakata (2016) extended the finding, showing that merely observing how an adult treats a third party shifts children’s willingness to wait; Moffett, Flannagan, and Shah (2020) replicated the reliability effect in a more familiar setting.

The interpretive weight of this work is considerable. It demonstrates that behavior which looks like a stable deficit in self-control is partly a decision computed over beliefs about delivery probability, and that those beliefs can be moved by evidence in minutes. This is the experimental proof-of-concept that the Unreliable Future Hypothesis requires, and it is the reason the framework proposes an adult analogue as its most decisive test (H6, Section 8).

It is also, appropriately, a limit on the framework’s ambition. Kidd et al. manipulated a state, not a developmental history. Showing that reliability beliefs are causally efficacious in a laboratory does not show that childhood interruption risk produces a durable adult trait. That inference is exactly what remains to be earned.

4.6 Convergent frameworks and what APRB would add

Several established frameworks describe overlapping territory. The Unreliable Future Hypothesis is not an alternative to any of them, and a reviewer is entitled to ask what it adds. The honest answer is: a specific exposure, a specific population, and a specific clinical decision.

The value proposition is therefore modest and specific: not a new theory of adversity, but a mechanistically specified subgroup hypothesis inside a diagnostic category where treatment is currently allocated without reference to mechanism.

5. What the Hypothesis Is and Is Not Claiming

The central hypothesis is this: in an unknown but clinically meaningful subset of individuals meeting DSM-5-TR criteria for ADHD-I, especially those with significant childhood adversity and high perceived childhood interruption risk, inattentive symptoms may be partly driven by adversity-calibrated present-focused reward bias rather than by the same dominant mechanisms seen in lower-adversity primary ADHD-I.

This claim has strict boundaries. A high ACE score does not mean a person does not have ADHD. Stimulant euphoria does not prove the absence of ADHD. Trauma therapy should not be assumed to replace ADHD medication. Primary ADHD can precede adversity, increase risk for adversity, coexist with adversity, or be intensified by adversity. The hypothesis is not a diagnostic shortcut. It is a mechanistic research question.

The model is strongest when it treats primary ADHD and APRB as potentially overlapping pathways rather than mutually exclusive boxes. Four relationships are plausible:

Independent mimicry: APRB generates ADHD-I-like impairment without primary neurodevelopmental ADHD being the main driver.

Amplification: primary ADHD is present, but adversity increases impairment, emotional reactivity, and delay-discounting behavior.

Subtype or modifier: genetic or neurodevelopmental ADHD liability interacts with adversity to produce a distinct clinical profile.

Reciprocal cascade: early ADHD symptoms increase exposure to adversity through conflict, rejection, academic failure, punishment, or family stress, and the adversity then feeds back into attention and reward calibration.

5.1 The confounding problem, stated at full strength

The reciprocal-cascade possibility is not cosmetic, and the evidence for it is stronger than the evidence for the simple environmental reading. Lugo-Candelas et al. (2021) found that ADHD predicted subsequent risk for ACEs, with inattentive presentation potentially conferring particular vulnerability. Stern et al. (2018), analyzing the E-Risk longitudinal twin cohort, found that childhood abuse and neglect did not predict young-adult ADHD once childhood ADHD was adjusted for, while childhood ADHD did predict later abuse and neglect. Their stated conclusion was that the data do not support a causal link from maltreatment to ADHD.

Genetically informative designs push in the same direction. Capusan et al. (2016), in a large adult twin sample, found the maltreatment–ADHD symptom association to be partly unique-environmental and partly familially confounded. Dinkler et al. (2017), using co-twin control analysis in over eight thousand Swedish nine-year-olds, found that maltreatment was associated with only a small increase in neurodevelopmental symptoms among maltreatment-discordant monozygotic twins, with most of the covariance explained by shared familial factors. The same critique — that correlational adversity findings carry a serious genetic confound — was leveled directly at Pepper and Nettle’s target article by commentators, and it has not been fully answered.

A framework that ignores this literature is not a framework; it is an advocacy document. Two features of the present proposal make the confound survivable in principle rather than merely inconvenient.

First, the model does not require adversity to cause ADHD. It requires that within people who already meet ADHD-I criteria, reward-timing calibration explains variance in mechanism, phenomenology, and treatment response. That is a within-diagnosis moderation claim, not a between-group etiological one. Genetic confounding of the adversity–ADHD correlation does not by itself refute a claim about heterogeneity among diagnosed cases.

Second, the specific exposure proposed here is not the exposure those studies measured. Maltreatment, abuse, and neglect overlap with interruption risk but are not identical to it. A household can be materially safe and highly unreliable; it can also be dangerous and perfectly predictable. The interruption-risk specification is testable against maltreatment measures, and if it explains no variance beyond them, that is informative.

Neither of these defenses is a license. If genetically informative designs show that reward-timing calibration in ADHD-I is largely familially confounded, the environmental component of this model should shrink to whatever remains, and the paper’s framing should shrink with it.

6. Phenotypic Overlap and Potential Differentiators

APRB may phenotypically overlap with ADHD-I because both can produce inconsistent attention, task initiation failure, procrastination, forgetfulness, disorganization, and difficulty persisting toward delayed goals. The difference lies in the proposed source of those behaviors. In APRB, the attention system is not merely underpowered; it is preferentially recruited by immediacy, salience, threat, relief, and perceived reliability.

The following contrasts are hypotheses, not diagnostic rules.

The table deliberately avoids clean binaries. If the framework has value, it will not be because clinicians can eyeball a patient and declare “this is trauma, not ADHD.” It will be because measured APRB variables improve prediction beyond symptom severity alone.

6.1 Competing accounts that must be excluded, not assumed away

Four alternative explanations can generate the same pattern of findings and must be measured rather than acknowledged in passing.

Cognitive disengagement syndrome. Formerly termed sluggish cognitive tempo, CDS is empirically distinct from ADHD inattention, is more strongly associated with internalizing psychopathology, and predicts functional impairment above and beyond ADHD symptoms (Becker et al., 2023). Because CDS is concentrated precisely in the ADHD-I space this framework targets, failing to measure it invites the objection that APRB is CDS with better branding.

Current PTSD, depression, and anhedonia. Each independently alters reward anticipation and effort allocation. The framework’s claim is about developmental calibration, not current state; if APRB markers vanish once current symptoms are modelled, the developmental claim loses its point.

Sleep and circadian disruption. Chronically shortened or delayed sleep degrades sustained attention and steepens effortful-choice avoidance, and is elevated in both adversity-exposed and ADHD populations.

Response style. Discussed in Section 9, this is the most underappreciated of the four and the one most capable of producing the predicted results in the absence of any mechanism at all.

7. Stimulant Response: Signal, Not Sorting Hat

The original intuition behind this paper included a provocative clinical observation: some high-adversity adults diagnosed with ADHD-I report pronounced euphoria or unusually strong affective activation from low-dose stimulant medication, rather than the calmer “return to baseline” experience commonly described by many ADHD patients. That observation may be useful, but it must be handled carefully.

Subjective stimulant response is not a diagnostic truth serum, and the field established this early. Rapoport et al. (1978, 1980) administered dextroamphetamine to normal prepubertal boys, hyperactive boys, and normal adult men under double-blind conditions and found broadly similar effects across groups: reduced motor activity, improved vigilance, better learning performance. The finding cast doubt on the then-standard assumption that stimulant response was diagnostically specific or “paradoxical,” and it has held up (Rapoport & Inoff-Germain, 2002). One detail of the 1980 study is directly relevant to the present design: the adult men reported euphoria while the boys did not, indicating that developmental stage moderates subjective response independently of diagnosis.

Methylphenidate can produce subjective and reinforcing effects, and those effects vary by dose, formulation, pharmacokinetics, prior exposure, expectancy, sleep, mood, route, substance-use history, cardiovascular state, and individual biology. Kollins et al. (2009) found subjective methylphenidate effects in adults with and without ADHD, while reinforcing effects differed by group. Kollins et al. (2001) reviewed methylphenidate abuse liability and emphasized the importance of context and pharmacological parameters.

The stronger prediction is therefore probabilistic: high-APRB participants may show greater placebo-corrected positive subjective activation, drug liking, euphoria, or affective lift during a standardized stimulant challenge than low-APRB ADHD-I participants matched on symptom severity and relevant confounds. This would not prove “not ADHD.” Given Rapoport’s findings, it would not even prove a diagnostic distinction. It would suggest that subjective stimulant response may be one marker of mechanistic heterogeneity within ADHD-I, especially when combined with delay discounting, adversity history, uncertainty reactivity, and treatment response.

This component of the framework is also the most expendable. It is the most expensive to test, the most ethically constrained, and the most likely to dissolve under adequate control. It should be treated as a downstream question, not a load-bearing one.

8. Testable Predictions

The framework is only useful if it produces predictions that could be wrong. The following hypotheses represent the minimum empirical commitments of the Unreliable Future model. Each should be preregistered with a smallest effect size of interest specified in advance, given that the relevant baseline effects are modest — the ADHD-versus-control difference in monetary delay discounting is approximately d = 0.43 (Jackson & MacKillop, 2016), and any adversity-related increment within an ADHD-I sample should be expected to be smaller than that.

H1: Among adults meeting full ADHD-I criteria, high-adversity participants will show steeper delay discounting than low-adversity participants matched on ADHD symptom severity, age, sex, estimated IQ, socioeconomic background, sleep quality, depression, anxiety, substance-use history, and current stress.

H2: Perceived childhood interruption risk will predict APRB markers more strongly than ACE count, perceived harshness, or maltreatment severity, and APRB markers will explain unique variance in ADHD-I-related functional impairment beyond total ADHD symptom severity and beyond cognitive disengagement syndrome symptoms.

H3: High-APRB participants will show greater attentional capture by immediate reward, threat, uncertainty, or emotionally salient cues on laboratory and ecological measures, with the enhancement–impairment pattern varying by whether task content matches the adversity dimension experienced (Frankenhuis et al., 2020).

H4: During a double-blind, placebo-controlled stimulant-challenge pilot, high-APRB participants will show greater placebo-corrected positive subjective activation or drug liking than low-APRB ADHD-I controls, after controlling for expectancy, prior stimulant exposure, current PTSD symptoms, sleep debt, and substance-use risk.

H5: APRB status will moderate treatment outcome, such that participants high in APRB receive greater incremental benefit from integrated ADHD care plus trauma-focused or future-reward-recalibration intervention than from standard ADHD care alone.

H6 (experimental, and the most decisive): Within-subject manipulation of environmental reliability — a demonstrated-reliable versus demonstrated-unreliable experimenter or task context, adapted for adults from Kidd et al. (2013) — will produce larger shifts in delay discounting, persistence, and task initiation among high-APRB than low-APRB ADHD-I participants.

H6 deserves emphasis because it inverts the usual epistemic position of this kind of proposal. H1 through H5 are correlational or quasi-experimental and inherit every confound described in Section 5.1. H6 manipulates the proposed active ingredient directly. A null result on H6 is more damaging to the model than a null on H1, because H1 could fail for measurement reasons while the mechanism holds, whereas H6 failing means reliability beliefs are not differentially recruited in the proposed subgroup. It is also, by an order of magnitude, the cheapest study in the program. It should be run first.

These hypotheses are intentionally narrower than the cultural claim that trauma is often misdiagnosed as ADHD. The framework does not need to win that argument. It only needs to show that APRB variables identify a subgroup whose impairment, stimulant phenomenology, and treatment response are predicted better by interruption-calibrated reward/time-horizon models than by ADHD symptom severity alone.

9. Proposed Empirical Program

Stage 0: Secondary analysis of existing prospective cohorts

Before any new data collection is funded, the cheapest and most informative move is secondary analysis of cohorts that already exist. The ABCD Study, the Environmental Risk (E-Risk) Longitudinal Twin Study, ALSPAC, and the Dunedin cohort contain prospectively measured adversity, repeated ADHD assessment, and in several cases reward and intertemporal-choice data. Genetically informative subsamples permit discordant-sibling and co-twin analyses of exactly the confound described in Section 5.1.

These datasets can address the direction-of-effect and familial-confounding questions that no cross-sectional adult sample can settle, and they can do so before money is spent on stimulant challenges. A framework that skips this step and proceeds directly to a novel laboratory paradigm is inviting the response, three years and one grant later, that its central association was confounded all along.

Stage 1: Mechanistic phenotyping study

The next study should be a mechanistic phenotyping study of adults meeting full ADHD-I criteria. A feasible initial sample would include approximately 240 participants, with high-adversity and low-adversity groups matched on ADHD symptom severity and core demographic variables.

Adversity should not be measured only by ACE count. ACE scores are useful as a screening tool but too blunt to carry the construct. The study should include the Childhood Trauma Questionnaire, dedicated measures of perceived childhood unpredictability and interruption risk (Young et al., 2020), threat/deprivation ratings, current stress load, current PTSD symptoms, depression, anxiety, sleep, substance-use history, socioeconomic background, medication history, and collateral developmental history when possible.

ADHD diagnosis should be confirmed using a structured adult ADHD interview such as DIVA-5 or an equivalent instrument, plus validated symptom scales such as the ASRS and CAARS. Cognitive disengagement syndrome should be measured with a validated CDS scale. PTSD should be assessed with the PCL-5 and, where feasible, CAPS-5. The primary behavioral outcomes should include monetary delay discounting, effort discounting, reward learning, attentional bias toward threat/reward cues, reaction-time variability under predictable versus unpredictable reward schedules, and ecological momentary assessment of immediate-relief behavior.

The key analysis is not simply whether high-ACE participants are more impaired. That would be unsurprising and nonspecific. The key analysis is whether perceived interruption risk and APRB markers predict ADHD-I impairment after matching and covariate control. If the model is right, the high-adversity group should not merely be “more symptomatic.” It should be differently organized around time, uncertainty, and reward reliability.

Design cautions that are not optional

Four methodological issues could produce or destroy the predicted results independent of any mechanism.

Symptom validity is a first-order confound here, not a hedge against malingering. Gonzalez et al. (2024), in a multiracial sample presenting for ADHD evaluation, found that high-ACE participants scored higher on symptom validity scales indexing over-reporting and infrequent symptom endorsement, while showing no corresponding difference on performance validity tests. Related work from the same group finds perceived stress associated with lifetime ADHD symptom reporting over and above anxiety and depression (Woloszyn et al., 2026). This is precisely the mechanism that could manufacture H1 and H2 out of nothing: if adversity exposure is associated with a response style that inflates symptom and impairment ratings, then a finding that adversity predicts self-reported impairment beyond symptom severity is uninterpretable. The design implication is twofold. Primary APRB outcomes must be behavioral rather than self-reported, and symptom and performance validity measures should be administered, reported, and modelled rather than used silently as exclusion criteria.

Retrospective adversity measurement is weakly related to what actually happened. Baldwin et al. (2019), meta-analyzing sixteen studies and over 25,000 participants, found poor agreement between prospective and retrospective measures of childhood maltreatment (κ = 0.19); roughly half of those with prospectively documented maltreatment did not retrospectively report it, and roughly half of retrospective reporters had no concordant prospective record. The two measures identify largely different people and are not interchangeable (see also Baldwin et al., 2024). This is simultaneously a threat and an opportunity for the present framework. It is a threat because a study built on retrospective report is not studying documented adversity. It is an opportunity because Danese and Widom (2020) found that psychopathology tracks subjective experience of maltreatment more closely than objective records — which is what a calibration model would predict, since the nervous system calibrates to what it encoded, not to what a case file recorded. The framework should say this explicitly and own the consequence: APRB is a claim about encoded expectancy, and should be measured and defended as such.

Matching and covariate choice can silently destroy the effect. Matching high- and low-adversity groups on ADHD symptom severity is defensible for isolating mechanism but introduces collider stratification: conditioning on a common downstream variable can induce spurious associations among its causes. Similarly, controlling for depression, anxiety, and sleep is appropriate if they are confounders and inappropriate if they are mediators on the adversity-to-impairment path, where adjustment removes part of the effect of interest. Both matched and unmatched models should be reported, mediator-adjusted and unadjusted analyses should be presented side by side, and the analytic decisions should be preregistered with a specification-curve or multiverse analysis rather than defended after the fact.

Moderation is underpowered at these sample sizes. Interaction effects typically require substantially larger samples than main effects to detect at equivalent power. A 240-participant Stage 1 is adequate for main effects and marginal for the moderation tests that carry the framework’s clinical claim. Pilot studies should not be used to generate power calculations for subsequent trials (Kraemer et al., 2006); Stage 3 in particular should be described as a feasibility and signal-detection study, and its effect estimates should not be treated as targets.

Stage 2: Stimulant-challenge pilot

A subsequent pilot should test subjective stimulant response under rigorous safeguards. A subset of 80 to 120 participants could complete a double-blind, placebo-controlled, crossover low-dose methylphenidate challenge. The study should measure placebo-corrected changes in drug liking, euphoria, stimulation, calmness, anxiety, focus, craving, and perceived agency using validated measures such as the Drug Effects Questionnaire, Addiction Research Center Inventory subscales, and visual analog scales. Physiological measures should include heart rate, blood pressure, heart-rate variability, and pupillometry. Expectancy ratings should be collected before dosing.

The prediction is not that stimulant response cleanly separates trauma-shaped ADHD-I from primary ADHD-I. Rapoport’s findings make that prediction untenable at the outset. The prediction is that stimulant phenomenology, when combined with APRB markers, may identify a high-adversity subgroup whose reward system responds differently to catecholaminergic enhancement. If no such pattern appears after appropriate controls, the stimulant-response component of the hypothesis should be weakened or discarded without prejudice to the rest of the model.

Stage 3: Treatment-moderation pilot

A third-stage pilot should test treatment moderation. The cleanest design would randomize high-APRB ADHD-I adults to either standard ADHD care or integrated care. Standard care could include medication optimization, ADHD-focused CBT, psychoeducation, skills support, and accommodations where appropriate. Integrated care would include standard care plus evidence-based trauma-focused psychotherapy when PTSD or trauma symptoms are present, or a structured safety/reward-recalibration protocol when the main target is future unreliability rather than explicit PTSD. Evidence-based PTSD treatments with strong support include Prolonged Exposure, Cognitive Processing Therapy, and EMDR, according to major clinical guidance and research summaries.

The integrated protocol should not be framed as “therapy instead of medication.” That is the wrong fight. The real test is additive and moderating: does APRB identify people for whom ADHD care works better when it also targets threat expectancy, future self-continuity, delayed-reward credibility, emotional safety, and uncertainty tolerance? Candidate adjunctive components could include behavioral activation, episodic future thinking, future-self continuity exercises, graded commitment practice, trauma-focused work where indicated, and — most directly implied by the interruption-risk specification — reinforcement schedules explicitly designed to deliver promised rewards reliably and visibly, so that the environment supplies disconfirming evidence rather than the patient supplying willpower. Episodic future thinking and related interventions have been shown in other contexts to reduce delay discounting (Peters & Büchel, 2010; Rung & Madden, 2018), making them plausible candidates for adaptation.

Primary outcomes should include ADHD-related functional impairment, task initiation, procrastination, delay discounting, emotional reactivity to uncertainty, work/school functioning, quality of life, and medication stability. Secondary outcomes should include PTSD symptoms, depression, anxiety, sleep, stimulant misuse indicators, and ecological measures of follow-through on delayed goals.

10. Falsification Criteria

The Unreliable Future Hypothesis should be considered weakened if any of the following occur. Where a criterion turns on a null result, it should be evaluated with equivalence testing against a prespecified smallest effect size of interest rather than by failure to reject the null.

High-adversity and low-adversity ADHD-I groups show no meaningful differences in delay discounting, attentional salience, uncertainty sensitivity, or reward-learning measures after matching and covariate control.

Perceived childhood interruption risk does not predict APRB markers beyond ACE count, maltreatment severity, perceived harshness, current stress, PTSD symptoms, depression, anxiety, sleep disruption, or socioeconomic background. This is the criterion most specific to the present formulation: if the interruption-risk specification adds nothing over generic adversity, the model reduces to existing accounts and should be retired in their favor.

APRB markers fail to predict ADHD-I-related impairment beyond standard ADHD symptom severity and beyond cognitive disengagement syndrome symptoms.

Group differences in self-reported impairment are accounted for by symptom validity indices or response style rather than by behavioral measures.

Experimental manipulation of environmental reliability produces no differential effect on intertemporal choice or persistence in high-APRB participants (H6). This is the most direct disconfirmation available and should be weighted accordingly.

Subjective stimulant-response differences disappear after controlling for expectancy, prior stimulant exposure, sleep, dose, formulation, current affective state, and substance-use risk.

APRB status does not moderate treatment outcome in integrated-care trials.

Low-adversity ADHD-I participants show equivalent or stronger APRB markers than high-adversity participants.

Genetically informative designs show that the association between adversity-related reward-timing calibration and ADHD-I impairment is largely accounted for by familial confounding.

Longitudinal data show that ADHD symptoms consistently precede adversity exposure and that adversity-related reward/time-horizon variables contribute little independent variance to later inattentive impairment.

The last two criteria are the most important, and current evidence already leans against the framework on both. Stern et al. (2018) and Dinkler et al. (2017) represent the state of play, and the state of play is not favorable. If ADHD causes much of the later adversity, and adversity adds little mechanistic value beyond worsening the life of someone already struggling, the model should shrink accordingly. The goal is not to protect an elegant idea. The goal is to find the true size of the effect, including the possibility that it is near zero.

11. Ethical and Clinical Boundaries

This framework has clinical implications, but it should not be treated as clinical guidance. No person should discontinue stimulant medication, reject an ADHD diagnosis, or self-classify as “trauma ADHD” on the basis of this hypothesis. The model is meant to generate research, not replace assessment.

Empirical studies involving trauma-exposed participants and stimulant administration would require full ethics review, medical oversight, careful exclusion criteria, and trauma-informed protocols. Participants should be screened for current substance-use disorder, bipolar disorder, psychosis risk, unstable cardiovascular conditions, stimulant contraindications, severe sleep deprivation, and acute suicidality. Stimulant-challenge procedures should include informed consent regarding possible subjective effects, medical monitoring, adverse-event procedures, and follow-up support. Trauma-history assessment should avoid unnecessary detail unless clinically or scientifically required, and participants should have access to support if distress is triggered.

One additional ethical consideration applies specifically to the symptom-validity component of Stage 1. Validity indices can carry stigmatizing implications, and the finding that high-ACE participants score higher on over-reporting scales is open to a benign interpretation — that people with more severe histories endorse more unusual symptoms because they have more unusual symptoms — as well as a pejorative one. Validity data should be used to model measurement, not to impugn participants, and any reporting should make that distinction explicit.

The clinical danger of the hypothesis cuts in both directions. Overstating it could lead clinicians to dismiss real ADHD in trauma survivors. Ignoring it could lead clinicians to treat an adversity-calibrated motivational system as if the only relevant mechanism were catecholamine availability. Both errors are avoidable. The framework asks for better differential assessment, not ideological conversion.

12. Limitations

Several limitations should constrain interpretation from the beginning.

ACE scores are blunt. They miss timing, severity, chronicity, subjective meaning, protective factors, developmental stage, and distinctions between threat, deprivation, and unpredictability.

Retrospective and prospective adversity measures identify substantially different people (Baldwin et al., 2019). Any study of adults relying on retrospective report is studying encoded and reconstructed experience, not documented history. This framework can absorb that fact but should not pretend otherwise.

Delay discounting is transdiagnostic, not specific. It is elevated across substance use disorders, depression, bipolar disorder, borderline personality disorder, schizophrenia, and disordered eating (Amlung et al., 2019). Standard intertemporal-choice tasks have also been criticized for placing participants in a context unlikely to generalize, and for measuring a trait abstracted from exactly the environmental parameters this framework claims matter (Bailey et al., 2021; Lempert et al., 2019). A steeper discount rate is a weak signature on its own.

APRB is probably transdiagnostic in the same way. It may appear in PTSD, complex trauma, depression, anxiety, substance-use risk, borderline traits, poverty-related stress, and chronic relational insecurity. Its presence would not be specific to ADHD-I.

Causality is difficult, and current evidence is not encouraging. ADHD may increase later adversity exposure; adversity may produce ADHD-like symptoms; shared genetic and environmental risks may contribute to both; and all three may occur in the same person. Genetically informative studies to date suggest the environmental component is smaller than the correlational literature implies.

Construct proliferation is a real risk. APRB overlaps substantially with delay aversion, latent vulnerability, the behavioural constellation of deprivation, and life-history fast strategies. If it does not earn incremental predictive validity over those constructs, it is a new name for an old thing, and naming is not explanation.

Subjective stimulant response is noisy and non-specific. Euphoria, relief, calm, drug liking, reduced anxiety, novelty, overfocus, and normalization are not the same phenomenon, and stimulants produce similar cognitive effects in people with and without ADHD (Rapoport et al., 1978, 1980).

Treatment response is nonspecific unless carefully measured. Improvement after trauma-focused work may reflect therapeutic alliance, expectancy, depression improvement, improved sleep, reduced avoidance, or better structure rather than specific recalibration of future reward.

The construct may identify a modifier rather than a subtype. That distinction changes what the framework licenses clinically, and it is an empirical question, not a framing choice.

These limitations do not weaken the reason to study the hypothesis. They define the conditions under which it would become scientifically useful rather than merely compelling.

13. Implications

If supported, the Unreliable Future Hypothesis would shift a portion of ADHD-I assessment from symptom counting toward mechanism mapping. Clinicians and researchers would ask not only whether the person is inattentive, but when attention fails, what the nervous system treats as urgent, how delayed reward is valued, whether uncertainty collapses task engagement, and whether the person’s future feels motivationally credible.

The framework could also improve treatment matching. Some individuals may benefit primarily from stimulant medication, ADHD-focused CBT, skills training, environmental modifications, and accommodations. Others may need those same supports plus interventions that rebuild trust in delayed payoff, reduce threat-driven salience capture, increase tolerance of uncertainty, and update learned models of safety and agency. In this view, procrastination is not merely a time-management problem. It may be the behavioral surface of a deeper prediction: long-term investment is not safe enough to spend energy on.

The interruption-risk specification carries a specific and somewhat inconvenient clinical implication. If the problem is a learned delivery probability, the treatment target is not only the patient’s beliefs but the reliability of the environment supplying the evidence. Interventions that ask a person to trust the future while the systems around them continue to break promises are asking for belief revision without data. This points toward structured, verifiable follow-through as an active ingredient rather than a logistical nicety.

The broader implication is humane as well as scientific. People with high-adversity histories are often told, implicitly or explicitly, that they lack discipline, motivation, or maturity. The present framework offers a different reading, one consistent with a growing literature that resists reading every post-adversity difference as damage (Frankenhuis & Nettle, 2020; Ellis et al., 2022): some may be acting from a nervous system that learned too well. The task is not to shame it into the future. The task is to make the future believable enough that attention can afford to go there.

14. Conclusion

ADHD-I is likely not one thing. It is a phenotype that may arise through multiple developmental routes. Primary neurodevelopmental ADHD remains well supported, and this paper does not challenge its validity. It argues instead that some high-adversity individuals who meet ADHD-I criteria may present with an overlapping symptom profile partly driven by adversity-calibrated present-focused reward bias.

The Unreliable Future Hypothesis centers a specific mechanism: when childhood environments teach that promised rewards are frequently interrupted before they arrive, the nervous system may rationally prioritize immediacy. Later, in safer but slower environments, that calibration can appear as inattention, procrastination, disorganization, and failure to persist toward delayed goals. The model predicts measurable differences in delay discounting, salience capture, uncertainty reactivity, sensitivity to experimentally manipulated reliability, stimulant phenomenology, and treatment moderation.

The hypothesis should now be forced into contact with data, and the order matters. The cheap experimental test and the existing prospective cohorts should come before the expensive laboratory program, because they can kill the idea faster. If it fails, it should be revised or abandoned. If it survives, it may help psychiatry distinguish between people who share a symptom cluster but do not share the same underlying problem. That is the prize: not a new label, but a more precise map of why attention goes where it goes.

The simplest formulation is this: some people may not be failing to care about the future. Their nervous system may have learned, early and repeatedly, that the future was not a safe place to invest.

References

Acabal, J. P., Jarosik, A. R., Torres, N., Finley, J.-C. A., Fisher, E. P., Ulrich, D. M., Kimchi, E. Z., Phillips, M. S., Cerny, B. M., & Soble, J. R. (2026). Childhood trauma exposure and active posttraumatic stress disorder symptoms differentially influence symptom presentation among young adults undergoing attention-deficit/hyperactivity disorder evaluation: Implications for psychiatric diagnosis and treatment. Annals of Clinical Psychiatry, 37(2), 71-78. https://doi.org/10.1177/10401237261434905

Acheson, A., Vincent, A. S., Cohoon, A. J., & Lovallo, W. R. (2019). Early life adversity and increased delay discounting: Findings from the Family Health Patterns Project. Experimental and Clinical Psychopharmacology, 27(2), 153-159. https://doi.org/10.1037/pha0000241

Agnew-Blais, J. C., Polanczyk, G. V., Danese, A., Wertz, J., Moffitt, T. E., & Arseneault, L. (2016). Evaluation of the persistence, remission, and emergence of attention-deficit/hyperactivity disorder in young adulthood. JAMA Psychiatry, 73(7), 713-720. https://doi.org/10.1001/jamapsychiatry.2016.0465

American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). American Psychiatric Association Publishing.

Amlung, M., Marsden, E., Holshausen, K., Morris, V., Patel, H., Vedelago, L., Naish, K. R., Reed, D. D., & McCabe, R. E. (2019). Delay discounting as a transdiagnostic process in psychiatric disorders: A meta-analysis. JAMA Psychiatry, 76(11), 1176-1186. https://doi.org/10.1001/jamapsychiatry.2019.2102

Bailey, A. J., Romeu, R. J., & Finn, P. R. (2021). The problems with delay discounting: A critical review of current practices and clinical applications. Psychological Medicine, 51(11), 1799-1806. https://doi.org/10.1017/S0033291721002282

Baldwin, J. R., Coleman, O., Francis, E. R., & Danese, A. (2024). Prospective and retrospective measures of child maltreatment and their association with psychopathology: A systematic review and meta-analysis. JAMA Psychiatry, 81(8), 769-781. https://doi.org/10.1001/jamapsychiatry.2024.0818

Baldwin, J. R., Reuben, A., Newbury, J. B., & Danese, A. (2019). Agreement between prospective and retrospective measures of childhood maltreatment: A systematic review and meta-analysis. JAMA Psychiatry, 76(6), 584-593. https://doi.org/10.1001/jamapsychiatry.2019.0097

Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and executive functions: Constructing a unifying theory of ADHD. Psychological Bulletin, 121(1), 65-94. https://doi.org/10.1037/0033-2909.121.1.65

Becker, S. P., Willcutt, E. G., Leopold, D. R., Fredrick, J. W., Smith, Z. R., Jacobson, L. A., Burns, G. L., Mayes, S. D., Waschbusch, D. A., Froehlich, T. E., McBurnett, K., Servera, M., & Barkley, R. A. (2023). Report of a work group on sluggish cognitive tempo: Key research directions and a consensus change in terminology to cognitive disengagement syndrome. Journal of the American Academy of Child & Adolescent Psychiatry, 62(6), 629-645. https://doi.org/10.1016/j.jaac.2022.07.821

Boecker, R., Holz, N. E., Buchmann, A. F., Blomeyer, D., Plichta, M. M., Wolf, I., Baumeister, S., Meyer-Lindenberg, A., Banaschewski, T., & Brandeis, D. (2014). Impact of early life adversity on reward processing in young adults: EEG-fMRI results from a prospective study over 25 years. PLOS ONE, 9(8), e104185. https://doi.org/10.1371/journal.pone.0104185

Brown, N. M., Brown, S. N., Briggs, R. D., German, M., Belamarich, P. F., & Oyeku, S. O. (2017). Associations between adverse childhood experiences and ADHD diagnosis and severity. Academic Pediatrics, 17(4), 349-355. https://doi.org/10.1016/j.acap.2016.08.013

Capusan, A. J., Kuja-Halkola, R., Bendtsen, P., Viding, E., McCrory, E., Marteinsdottir, I., & Larsson, H. (2016). Childhood maltreatment and attention deficit hyperactivity disorder symptoms in adults: A large twin study. Psychological Medicine, 46(12), 2637-2646. https://doi.org/10.1017/S0033291716001021

Caye, A., Rocha, T. B.-M., Anselmi, L., Murray, J., Menezes, A. M. B., Barros, F. C., Gonçalves, H., Wehrmeister, F., Jensen, C. M., Steinhausen, H.-C., Swanson, J. M., Kieling, C., & Rohde, L. A. (2016). Attention-deficit/hyperactivity disorder trajectories from childhood to young adulthood: Evidence from a birth cohort supporting a late-onset syndrome. JAMA Psychiatry, 73(7), 705-712. https://doi.org/10.1001/jamapsychiatry.2016.0383

Coghill, D. R., Seth, S., & Matthews, K. (2014). A comprehensive assessment of memory, delay aversion, timing, inhibition, decision making and variability in attention deficit hyperactivity disorder: Advancing beyond the three-pathway models. Psychological Medicine, 44(9), 1989-2001. https://doi.org/10.1017/S0033291713002547

Danese, A., & Widom, C. S. (2020). Objective and subjective experiences of child maltreatment and their relationships with psychopathology. Nature Human Behaviour, 4(8), 811-818. https://doi.org/10.1038/s41562-020-0880-3

Del Giudice, M., Ellis, B. J., & Shirtcliff, E. A. (2011). The Adaptive Calibration Model of stress responsivity. Neuroscience & Biobehavioral Reviews, 35(7), 1562-1592. https://doi.org/10.1016/j.neubiorev.2010.11.007

Dinkler, L., Lundström, S., Gajwani, R., Lichtenstein, P., Gillberg, C., & Minnis, H. (2017). Maltreatment-associated neurodevelopmental disorders: A co-twin control analysis. Journal of Child Psychology and Psychiatry, 58(6), 691-701. https://doi.org/10.1111/jcpp.12682

Ellis, B. J., Abrams, L. S., Masten, A. S., Sternberg, R. J., Tottenham, N., & Frankenhuis, W. E. (2022). Hidden talents in harsh environments. Development and Psychopathology, 34(1), 95-113. https://doi.org/10.1017/S0954579420000887

Ellis, B. J., Bianchi, J., Griskevicius, V., & Frankenhuis, W. E. (2017). Beyond risk and protective factors: An adaptation-based approach to resilience. Perspectives on Psychological Science, 12(4), 561-587. https://doi.org/10.1177/1745691617693054

Ellis, B. J., Oldehinkel, A. J., & Nederhof, E. (2017). The Adaptive Calibration Model of stress responsivity: An empirical test in the Tracking Adolescents’ Individual Lives Survey study. Development and Psychopathology, 29(3), 1001-1021. https://doi.org/10.1017/S0954579416000985

Faraone, S. V., Banaschewski, T., Coghill, D., Zheng, Y., Biederman, J., Bellgrove, M. A., Newcorn, J. H., Gignac, M., Al Saud, N. M., Manor, I., Rohde, L. A., Yang, L., Cortese, S., Almagor, D., Stein, M. A., Albatti, T. H., Aljoudi, H. F., Alqahtani, M. M. J., Asherson, P., … Wang, Y. (2021). The World Federation of ADHD International Consensus Statement: 208 evidence-based conclusions about the disorder. Neuroscience & Biobehavioral Reviews, 128, 789-818. https://doi.org/10.1016/j.neubiorev.2021.01.022

Fenneman, J., Frankenhuis, W. E., & Todd, P. M. (2022). In which environments is impulsive behavior adaptive? A cross-discipline review and integration of formal models. Psychological Bulletin, 148(7-8), 555-587. https://doi.org/10.1037/bul0000375

Frankenhuis, W. E., & Nettle, D. (2020). The strengths of people in poverty. Current Directions in Psychological Science, 29(1), 16-21. https://doi.org/10.1177/0963721419881154

Frankenhuis, W. E., Young, E. S., & Ellis, B. J. (2020). The hidden talents approach: Theoretical and methodological challenges. Trends in Cognitive Sciences, 24(7), 569-581. https://doi.org/10.1016/j.tics.2020.03.007

Gonzalez, C., Finley, J.-C. A., Khalid, E., Basurto, K. S., VanLandingham, H. B., Frick, L. A., Brooks, J. M., Ellison, R. L., Ulrich, D. M., Soble, J. R., & Resch, Z. J. (2024). The impact of adverse childhood experiences on symptom and performance validity tests among a multiracial sample presenting for ADHD evaluation. Archives of Clinical Neuropsychology, 39(6), 692-701. https://doi.org/10.1093/arclin/acae006

Hanson, J. L., Williams, A. V., Bangasser, D. A., & Peña, C. J. (2021). Impact of early life stress on reward circuit function and regulation. Frontiers in Psychiatry, 12, 744690. https://doi.org/10.3389/fpsyt.2021.744690

Hershfield, H. E. (2011). Future self-continuity: How conceptions of the future self transform intertemporal choice. Annals of the New York Academy of Sciences, 1235(1), 30-43. https://doi.org/10.1111/j.1749-6632.2011.06201.x

Hill, E. M., Jenkins, J., & Farmer, L. (2008). Family unpredictability, future discounting, and risk taking. The Journal of Socio-Economics, 37(4), 1381-1396. https://doi.org/10.1016/j.socec.2006.12.081

Jackson, J. N. S., & MacKillop, J. (2016). Attention-deficit/hyperactivity disorder and monetary delay discounting: A meta-analysis of case-control studies. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 1(4), 316-325. https://doi.org/10.1016/j.bpsc.2016.01.007

Kasparek, S. W., Gastón-Panthaki, A., Hanford, L. C., Lengua, L. J., Sheridan, M. A., & McLaughlin, K. A. (2023). Does reward processing moderate or mediate the link between childhood adversity and psychopathology? A longitudinal study. Development and Psychopathology, 35(5), 2338-2351. https://doi.org/10.1017/S0954579423000962

Kidd, C., Palmeri, H., & Aslin, R. N. (2013). Rational snacking: Young children’s decision-making on the marshmallow task is moderated by beliefs about environmental reliability. Cognition, 126(1), 109-114. https://doi.org/10.1016/j.cognition.2012.08.004

Kollins, S. H., English, J., Robinson, R., Hallyburton, M., & Chrisman, A. K. (2009). Reinforcing and subjective effects of methylphenidate in adults with and without attention deficit hyperactivity disorder (ADHD). Psychopharmacology, 204(1), 73-83. https://doi.org/10.1007/s00213-008-1439-6

Kollins, S. H., MacDonald, E. K., & Rush, C. R. (2001). Assessing the abuse potential of methylphenidate in nonhuman and human subjects: A review. Pharmacology, Biochemistry and Behavior, 68(3), 611-627. https://doi.org/10.1016/S0091-3057(01)00464-6

Kraemer, H. C., Mintz, J., Noda, A., Tinklenberg, J., & Yesavage, J. A. (2006). Caution regarding the use of pilot studies to guide power calculations for study proposals. Archives of General Psychiatry, 63(5), 484-489. https://doi.org/10.1001/archpsyc.63.5.484

Lempert, K. M., Steinglass, J. E., Pinto, A., Kable, J. W., & Simpson, H. B. (2019). Can delay discounting deliver on the promise of RDoC? Psychological Medicine, 49(2), 190-199. https://doi.org/10.1017/S0033291718001770

Lugo-Candelas, C., Corbeil, T., Wall, M., Posner, J., Bird, H., Canino, G., Fisher, P. W., Suglia, S. F., & Duarte, C. S. (2021). ADHD and risk for subsequent adverse childhood experiences: Understanding the cycle of adversity. Journal of Child Psychology and Psychiatry, 62(8), 971-978. https://doi.org/10.1111/jcpp.13352

Luo, Y., Weibman, D., Halperin, J. M., & Li, X. (2019). A review of heterogeneity in attention deficit/hyperactivity disorder (ADHD). Frontiers in Human Neuroscience, 13, 42. https://doi.org/10.3389/fnhum.2019.00042

Martinez, J. L., Hasty, C., Morabito, D., Maranges, H. M., Schmidt, N. B., & Maner, J. K. (2022). Perceptions of childhood unpredictability, delay discounting, risk-taking, and adult externalizing behaviors: A life-history approach. Development and Psychopathology, 34(2), 705-717. https://doi.org/10.1017/S0954579421001607

Marx, I., Hacker, T., Yu, X., Cortese, S., & Sonuga-Barke, E. (2021). ADHD and the choice of small immediate over larger delayed rewards: A comparative meta-analysis of performance on simple choice-delay and temporal discounting paradigms. Journal of Attention Disorders, 25(2), 171-187. https://doi.org/10.1177/1087054718772138

McCrory, E. J., Gerin, M. I., & Viding, E. (2017). Annual research review: Childhood maltreatment, latent vulnerability and the shift to preventative psychiatry - the contribution of functional brain imaging. Journal of Child Psychology and Psychiatry, 58(4), 338-357. https://doi.org/10.1111/jcpp.12713

McCrory, E. J., & Viding, E. (2015). The theory of latent vulnerability: Reconceptualizing the link between childhood maltreatment and psychiatric disorder. Development and Psychopathology, 27(2), 493-505. https://doi.org/10.1017/S0954579415000115

McLaughlin, K. A., Sheridan, M. A., & Lambert, H. K. (2014). Childhood adversity and neural development: Deprivation and threat as distinct dimensions of early experience. Neuroscience & Biobehavioral Reviews, 47, 578-591. https://doi.org/10.1016/j.neubiorev.2014.10.012

Michaelson, L. E., & Munakata, Y. (2016). Trust matters: Seeing how an adult treats another person influences preschoolers’ willingness to delay gratification. Developmental Science, 19(6), 1011-1019. https://doi.org/10.1111/desc.12388

Moffitt, T. E., Houts, R., Asherson, P., Belsky, D. W., Corcoran, D. L., Hammerle, M., Harrington, H., Hogan, S., Meier, M. H., Polanczyk, G. V., Poulton, R., Ramrakha, S., Sugden, K., Williams, B., Rohde, L. A., & Caspi, A. (2015). Is adult ADHD a childhood-onset neurodevelopmental disorder? Evidence from a four-decade longitudinal cohort study. American Journal of Psychiatry, 172(10), 967-977. https://doi.org/10.1176/appi.ajp.2015.14101266

Pepper, G. V., & Nettle, D. (2017). The behavioural constellation of deprivation: Causes and consequences. Behavioral and Brain Sciences, 40, e314. https://doi.org/10.1017/S0140525X1600234X

Peters, J., & Büchel, C. (2010). Episodic future thinking reduces reward delay discounting through an enhancement of prefrontal-mediotemporal interactions. Neuron, 66(1), 138-148. https://doi.org/10.1016/j.neuron.2010.03.026

Rapoport, J. L., Buchsbaum, M. S., Weingartner, H., Zahn, T. P., Ludlow, C., & Mikkelsen, E. J. (1980). Dextroamphetamine: Its cognitive and behavioral effects in normal and hyperactive boys and normal men. Archives of General Psychiatry, 37(8), 933-943. https://doi.org/10.1001/archpsyc.1980.01780210091010

Rapoport, J. L., Buchsbaum, M. S., Zahn, T. P., Weingartner, H., Ludlow, C., & Mikkelsen, E. J. (1978). Dextroamphetamine: Cognitive and behavioral effects in normal prepubertal boys. Science, 199(4328), 560-563. https://doi.org/10.1126/science.341313

Rapoport, J. L., & Inoff-Germain, G. (2002). Responses to methylphenidate in attention-deficit/hyperactivity disorder and normal children: Update 2002. Journal of Attention Disorders, 6(Suppl. 1), S57-S60. https://doi.org/10.1177/070674370200601S07

Rung, J. M., & Madden, G. J. (2018). Experimental reductions of delay discounting and impulsive choice: A systematic review and meta-analysis. Journal of Experimental Psychology: General, 147(9), 1349-1381. https://doi.org/10.1037/xge0000462

Sibley, M. H., Rohde, L. A., Swanson, J. M., Hechtman, L. T., Molina, B. S. G., Mitchell, J. T., Arnold, L. E., Caye, A., Kennedy, T. M., Roy, A., Stehli, A., & MTA Cooperative Group. (2018). Late-onset ADHD reconsidered with comprehensive repeated assessments between ages 10 and 25. American Journal of Psychiatry, 175(2), 140-149. https://doi.org/10.1176/appi.ajp.2017.17030298

Sonuga-Barke, E. J. S. (2002). Psychological heterogeneity in AD/HD - A dual pathway model of behaviour and cognition. Behavioural Brain Research, 130(1-2), 29-36. https://doi.org/10.1016/S0166-4328(01)00432-6

Sonuga-Barke, E. J. S. (2003). The dual pathway model of AD/HD: An elaboration of neuro-developmental characteristics. Neuroscience & Biobehavioral Reviews, 27(7), 593-604. https://doi.org/10.1016/j.neubiorev.2003.08.005

Sonuga-Barke, E. J. S., Bitsakou, P., & Thompson, M. (2010). Beyond the dual pathway model: Evidence for the dissociation of timing, inhibitory, and delay-related impairments in attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 49(4), 345-355. https://doi.org/10.1016/j.jaac.2009.12.018

Sonuga-Barke, E. J. S., Taylor, E., Sembi, S., & Smith, J. (1992). Hyperactivity and delay aversion I: The effect of delay on choice. Journal of Child Psychology and Psychiatry, 33(2), 387-398. https://doi.org/10.1111/j.1469-7610.1992.tb00874.x

Stern, A., Agnew-Blais, J., Danese, A., Fisher, H. L., Jaffee, S. R., Matthews, T., Polanczyk, G. V., & Arseneault, L. (2018). Associations between abuse/neglect and ADHD from childhood to young adulthood: A prospective nationally-representative twin study. Child Abuse & Neglect, 81, 274-285. https://doi.org/10.1016/j.chiabu.2018.04.025

Teicher, M. H., & Samson, J. A. (2016). Annual research review: Enduring neurobiological effects of childhood abuse and neglect. Journal of Child Psychology and Psychiatry, 57(3), 241-266. https://doi.org/10.1111/jcpp.12507

U.S. Department of Veterans Affairs, National Center for PTSD. (n.d.). Overview of psychotherapy for PTSD. https://www.ptsd.va.gov/professional/treat/txessentials/overview_therapy.asp

Woloszyn, M., Hernandez-Vallant, A., Palladino, N. T., Galindo, B., Leese, M. I., Khan, F., Kimchi, E. Z., Cerny, B. M., Phillips, M. S., & Soble, J. R. (2026). Perceived stress is associated with lifetime ADHD symptom reporting over and above anxiety and depression: Implications for psychiatric evaluation and treatment. Journal of Psychiatric Practice, 32(2), 64-69. https://doi.org/10.1097/PRA.0000000000000909

Young, E. S., Frankenhuis, W. E., & Ellis, B. J. (2020). Theory and measurement of environmental unpredictability. Evolution and Human Behavior, 41(6), 550-556. https://doi.org/10.1016/j.evolhumbehav.2020.08.006

Zhang, N., Gao, M., Yu, J., Zhang, Q., Wang, W., Zhou, C., Liu, L., Sun, T., Liao, X., & Wang, J. (2022). Understanding the association between adverse childhood experiences and subsequent attention deficit hyperactivity disorder: A systematic review and meta-analysis of observational studies. Brain and Behavior, 12(10), e2748. https://doi.org/10.1002/brb3.2748