Research Pass for "The Medium Becomes the Ground": Evidence Map for Seven Theoretical Claims

This report assesses the empirical and theoretical literature underpinning seven distinct claims in the perspective paper. For each, the existing evidence is summarized, key authors and works are flagged, and the paper's claim is rated as well-supported, partially supported, or novel/unsupported (which the brief explicitly treats as a potential original contribution rather than a weakness). A cross-cutting observation appears at the end: the paper's strongest original move is integrative — fusing literatures that have rarely if ever been placed in the same frame.

1. Sequential Pathway Modeling: ADHD → Screens → Nicotine → Cannabis → Alcohol as a Cascade

Verdict: Well-supported in fragments, but the integrative four-substance sequential cascade specifically grounded in ADHD-driven prefrontal vulnerability is a novel synthesis.

The literature contains four discrete but overlapping bodies of work that the paper effectively braids together:

(a) The gateway hypothesis as a sequential causal model. Denise Kandel's classical formulation (1975; Kandel et al. 2006) describes a stage-wise progression — alcohol/tobacco → cannabis → other illicit drugs — and was reformulated by Kandel and Kandel (2014, New England Journal of Medicine, "A Molecular Basis for Nicotine as a Gateway Drug") as a neurobiological priming mechanism: nicotine produces FosB-mediated histone hyperacetylation in the striatum that primes responses to later cocaine. Kandel and Hurd's review "Testing the Gateway Hypothesis" (Neuropsychopharmacology 2017) consolidates animal evidence that adolescent nicotine sensitizes to psychostimulants and adolescent cannabinoids sensitize to opioids — i.e., prior exposure mechanistically alters the substrate that subsequent exposures act upon. This is precisely the "successive degradation of the same neurodevelopmental substrate" the paper proposes. The strongest single citation supporting the paper's central mechanism is the Kandel & Kandel (2014) NEJM piece.

(b) Common liability vs. gateway debate. Vanyukov et al. (2012, Drug and Alcohol Dependence, "Common liability to addiction and 'gateway hypothesis': theoretical, empirical and evolutionary perspective") argue these models are complementary rather than competing — shared liability explains general use, while drug-specific neurobiological sensitization explains sequence and specificity. The paper's framing is consistent with this synthesis. Cross-national epidemiological data (e.g., in Japan, 83.2% of illicit-drug users did not use cannabis first; in Nigeria, 77.8% did not) show the cannabis-first pattern is not universal, suggesting availability moderates the cascade — a useful nuance the paper should acknowledge.

(c) Developmental cascade theory. This is the formal psychological framework the paper's title invokes. Masten and Cicchetti (2010, Development and Psychopathology) established the construct; Masten et al. (2005, Developmental Psychology, "Developmental cascades: Linking academic achievement, externalizing and internalizing symptoms over 20 years") provided the foundational empirical model. Dodge et al. (2009, "A dynamic cascade model of the development of substance-use onset") explicitly applied it to substance use, proposing a multistage transactional social learning model. Bradshaw et al. (2013, "Testing a developmental cascade model of adolescent substance use trajectories and young adult adjustment") tested it in an urban African-American cohort. A 2022 systematic review (Hussong et al., Addictive Behaviors Reports) found that while cascade models are conceptually dominant in developmental psychopathology, "there is very little [direct] evidence for developmental cascade effects involving early parental socialization and substance use etiology" — methodological constraints (sample size, follow-up length) more than theoretical failure explain this gap. The paper's claim that successive substances each restructure vulnerability is theoretically aligned with this tradition, but the four-stage substance-specific sequence it proposes (screens → nicotine → cannabis → alcohol) has not been empirically modeled as a single cascade.

(d) ADHD as the upstream node. Lee et al. (2011, Clinical Psychology Review, "Prospective association of childhood attention-deficit/hyperactivity disorder (ADHD) and substance use and abuse/dependence: A meta-analytic review") meta-analyzed 27 prospective longitudinal studies and found childhood ADHD significantly predicted later nicotine, cannabis, cocaine, and "other" substance use disorders (though not lifetime alcohol use specifically — an important caveat). Ilbegi et al. (2018, Journal of Neurodevelopmental Disorders, "Substance use and nicotine dependence in persistent, remittent, and late-onset ADHD: a 10-year longitudinal study") found persistent ADHD carried HR=4.56 for SUD vs. controls, while remittent ADHD did not — pointing to ongoing executive dysfunction rather than trait ADHD per se as the mechanism. Molina and Pelham (2003) and Malone et al. (2010) showed escalating ADHD symptoms in early adolescence predict adolescent substance use even with subclinical childhood ADHD severity. On screen time as a prior cascade link: Zhang et al. (2025, Translational Psychiatry) used ABCD Study data (n≈10,000) to show baseline screen time at 9–10 predicted ADHD symptom increase two years later, mediated by reduced cortical volume (right temporal pole, left superior frontal gyrus, left rostral middle frontal gyrus). Mendelian randomization studies (Yang et al., 2024) complicate this by showing ADHD polygenic risk predicts screen time, suggesting bidirectionality.

Key takeaway for the paper: Every adjacent link in the cascade has empirical support; the integrated five-stage cascade specifically routed through prefrontal–limbic regulation is the paper's original synthetic move. Spear (2016, Neuroscience & Biobehavioral Reviews) provides the cleanest neurodevelopmental framing — adolescent exposure to nicotine, cannabis, and alcohol each produces drug-specific persistent alterations in PFC maturation. The paper should explicitly cite Kandel & Kandel (2014), Kandel & Hurd (2017), Masten & Cicchetti (2010), Dodge et al. (2009), Lee et al. (2011), Ilbegi et al. (2018), and Spear (2016) as the empirical scaffolding for what is otherwise a theoretically original integration.

2. McLuhan and Addiction Neuroscience: An Apparent Theoretical Gap

Verdict: Novel theoretical contribution. No published literature explicitly bridges McLuhan's media theory to the allostatic model of addiction (Koob & Le Moal) or to addiction neuroscience generally.

McLuhan's framework has been applied recently to attention economies and digital cognition. Yoni Van Den Eede's work on "eco-formed attention" (2020, Explorations in Media Ecology, "Marshall McLuhan's Theory of Attention: How to Become a Media Psychonaut") synthesizes figure/ground and cliché/archetype into a constitutive, environmental theory of attention. Robert Logan's Philosophies (2017, "The Alphabet Effect Re-Visited, McLuhan Reversals and Complexity Theory") and Memarovic et al.'s work on figure/ground in public displays extend the framework to mediated environments. McLuhan himself wrote in Understanding Media (1964) that all media are "extensions of man" that produce "autoamputation" — a numbing of the extended sense organ — and this is the closest McLuhan came to an addiction-like phenomenology (the user is numbed, then dependent on the extension).

However, no peer-reviewed work appears to:

Connect McLuhan's "ground" to Koob & Le Moal's allostatic set-point shift, despite the conceptual parallel: both describe a hidden environmental substrate that becomes constitutive of perception and is invisible until it shifts.
Treat substances themselves as McLuhan-style "media" — as extensions that, once integrated, restructure the perceptual field and become "ground."
Use hot/cool media analysis to differentiate substance phenomenology (e.g., nicotine as a "hot" medium of intense, low-participation reinforcement vs. cannabis as a "cool" medium requiring active interpretive participation).
Cite McLuhan in the addiction neuroscience literature on PubMed or in Neuropsychopharmacology, Addiction, or Neuroscience & Biobehavioral Reviews.

The most relevant cognate work is Koob's allostatic model itself: Koob & Le Moal (2001, Neuropsychopharmacology; 2008); Koob & Schulkin (2018, Neuroscience & Biobehavioral Reviews, "Addiction and stress: An allostatic view"); Robbins (2006, Nature Neuroscience, "A walk on the dark side: addiction as allostasis"); and Kwako & Koob (2017, Chronic Stress, "Neuroclinical Framework for the Role of Stress in Addiction"). These describe how repeated drug administration shifts the hedonic set-point downward through within-system and between-system neuroadaptations — the new "normal" is a dysregulated state that the user does not perceive as abnormal because it has become the ground of experience. This is, in McLuhan's vocabulary, exactly what happens when a medium becomes environmental: the user cannot see the figure because the ground has shifted.

This is the paper's most original theoretical contribution and should be flagged as such. The claim is not merely that McLuhan's framework is useful for thinking about addiction — it is that allostasis is the McLuhanian mechanism described in neurobiological terms. The framework offers a vocabulary the allostatic literature lacks: the distinction between figure (the substance, the screen, the explicit "thing") and ground (the regulatory environment it silently restructures); the inversion of attention (we notice the figure precisely because we cannot see the ground); and the tetrad (enhancement / obsolescence / retrieval / reversal) as a tool for analyzing what each substance amplifies, displaces, brings back, and ultimately flips into. Recommendation: the paper should make this explicit in its theoretical contributions section and present it as a generative reframing rather than as an established connection. It is potentially the manuscript's signature original idea.

3. Early-Onset vs. Late-Onset Addiction Phenomenology: "No Prior Self to Return To"

Verdict: Partially supported. The "constitutive vs. modificatory" distinction is intuited but not formally articulated in the existing literature. The paper would be making an original phenomenological refinement.

The closest existing work is in the social identity/recovery literature:

*Dingle, Cruwys, and Frings (2015, "Social Identities as Pathways into and out of Addiction," Frontiers in Psychology) conducted thematic analysis on 21 therapeutic community residents and identified two pathways: an identity loss pathway (the dominant "redemption narrative" — people lose a pre-addiction identity and try to renew it in recovery) and an identity gain pathway (socially isolated individuals who never had a stable pre-addiction identity and instead constructed addiction as their first valued social identity). The identity-gain pathway corresponds precisely to the paper's "no prior self to return to" claim — recovery for these individuals required building an aspirational new identity, not restoring an old one. This is the strongest direct empirical support available.

Female adolescent SUD identity reconstruction work (2021, Behavioral Sciences, "Identity (Re)Construction of Female Adolescents with Substance Use Disorders," PMC8297192) interviewed 20 adolescents in a therapeutic boarding school. The authors explicitly describe how 12-step models, designed for adults, ask adolescents "to let go of the one version of self that they have mastered" — confirming that adolescent-onset recovery is qualitatively different because substance use was constitutive of, rather than overlaid on, identity formation. The authors draw on Erikson and Marcia's identity-status framework to argue adolescent SUD interrupts the very developmental task of identity consolidation.

The Swedish snus focus-group study (Edvardsson et al., 2012, BMC Public Health) found adolescents reported that nicotine use was inseparable from "the creation of an image and identity" — once stuck, cessation was difficult not for pharmacological but for identity reasons.

Cloninger's Type II alcoholism (early-onset, male, externalizing, impulsive) versus Type I (late-onset, environmentally triggered) is a longstanding nosological distinction that maps loosely onto the paper's argument but is framed in terms of personality/genetic typology rather than phenomenology.

Additional cognate literatures: Jellinek's typology of alcoholism (1960); Zucker's developmental subtypes; and the broader emerging adulthood literature (Arnett) on identity moratorium.
Gap and original contribution: No work I located explicitly frames this as the difference between substance as modificatory (altering an existing self that recovery can return to) versus substance as constitutive (the self formed around the substance such that "sobriety" is encountered as identity-construction rather than identity-restoration). The paper's framing extends Dingle's "identity gain" pathway from a recovery-stage observation to a developmental claim about what was never built in the first place. This is a meaningful phenomenological refinement and could be supported by — but is not duplicative of — the cited work.
Useful additional construct: Brené Brown's distinction between guilt ("I did something bad") and shame ("I am bad") maps onto modificatory vs. constitutive addiction phenomenology and may be a productive bridge.

4. Scarcity, Poverty, and ADHD Cognition

Verdict: The Mullainathan/Shafir scarcity framework is well-supported; its application to ADHD specifically is partially supported and represents an underexplored intersection the paper can productively claim.

Foundational scarcity literature: Mani, Mullainathan, Shafir, and Zhao (2013, Science, "Poverty Impedes Cognitive Function") demonstrated experimentally — both with New Jersey mall shoppers contemplating financial scenarios and with Tamil Nadu sugarcane farmers measured pre- and post-harvest — that financial scarcity reduces fluid intelligence and executive function (effect sizes equivalent to losing a full night's sleep, roughly 13 IQ points). Shah, Mullainathan, and Shafir (2012, Science, "Some consequences of having too little") showed scarcity produces attentional tunneling. Mullainathan and Shafir's Scarcity: Why Having Too Little Means So Much (2013) coined "cognitive bandwidth tax." A 2024 meta-analysis (Journal of Behavioral and Experimental Economics) on financial scarcity and cognitive performance found mixed but generally supportive evidence for the cognitive-load mechanism, with some null findings (Carvalho et al. 2016; Wicherts & Scholten 2013). Shah et al.'s 2018 self-replication failed to replicate the cognitive-fatigue finding from 2012 — a caveat the paper should note. Schilbach, Schofield, and Mullainathan (2016, AER Papers and Proceedings, "The Psychological Lives of the Poor") and Gennetian & Shafir (2015) develop the bandwidth framework further.

ADHD × SES literature: Fan et al. (2022, International Journal of Environmental Research and Public Health) modeled SES → executive function → ADHD symptoms in 361 preschoolers and found EF (especially working memory and inhibitory control) mediated SES effects on ADHD symptoms. Russell, Ford, Williams, and Russell's series of papers (e.g., 2016, Journal of Abnormal Child Psychology; PMC5948120) found the association of ADHD with socioeconomic disadvantage holds even after controlling for parent ADHD and child comorbidities, suggesting an independent environmental contribution. A 2022 bi-directional Mendelian randomization study (using ADHD GWAS data with SES outcomes) suggests genetic liability to ADHD has a causal effect on lower SES; evidence for the reverse direction is present but weaker — implicating ADHD as upstream in adulthood while environmental SES remains upstream in early childhood. Population-scale neuroimaging work documents that neighborhood disadvantage is associated with reduced DLPFC activation during working memory tasks and lower prefrontal gray matter volume — the same regions implicated in ADHD.

Specific bidirectional claim:

Does scarcity worsen ADHD? Indirect evidence is substantial. Family stress models (Conger; Evans) document that poverty produces both proximal (parental stress, harsh parenting) and distal (toxin exposure, sleep deprivation, food insecurity) hits on the same prefrontal circuits ADHD already affects. Narrative reviews on poverty-related environmental risks for childhood ADHD make this argument explicitly.

Does ADHD amplify scarcity effects? This is the under-studied direction. Logically, if executive function is the mediator through which scarcity impairs decisions (Mullainathan & Shafir's core thesis), then a population with baseline EF deficits has less bandwidth to lose before functional collapse. No published study I located directly tests this multiplicative interaction — but it is a clean empirical prediction the paper can advance as a contribution.

Recommendation: Cite Mani et al. 2013, Mullainathan & Shafir 2013, Shah et al. 2012, Fan et al. 2022, the SES-ADHD bidirectional MR work, Russell et al. on independent SES effects, and Schilbach et al. 2016. Flag the ADHD × scarcity multiplicative interaction as a tractable empirical claim that follows from existing theory but has not been directly tested.

5. Environmental Affordances and ADHD

Verdict: Well-supported empirically for nature exposure / Attention Restoration Theory; the broader Gibsonian "affordances as regulatory scaffold" framing is theoretically novel. Housing instability evidence is indirect.

Attention Restoration Theory (ART) applied to ADHD is well-established:

Kaplan & Kaplan (1989, The Experience of Nature) and Kaplan (1995, Journal of Environmental Psychology, "The restorative benefits of nature") provide the foundation: directed attention fatigues with use; soft fascination (natural environments) restores it.
Andrea Faber Taylor and Frances Kuo's program of research at the University of Illinois' Landscape and Human Health Laboratory is the dominant ADHD-specific application:

- Taylor, Kuo, & Sullivan (2001, Environment & Behavior, "Coping with ADD: The Surprising Connection to Green Play Settings") — green settings improve attention in children with ADD. - Kuo & Taylor (2004, American Journal of Public Health, "A potential natural treatment for Attention-Deficit/Hyperactivity Disorder: Evidence from a national study"). - Faber Taylor & Kuo (2009, Journal of Attention Disorders) — 20-minute walk in park produced concentration improvements approaching the magnitude of a typical methylphenidate dose.

Population-scale confirmations: Donovan et al. (2019, The Lancet Planetary Health, "Association between exposure to the natural environment, rurality, and attention-deficit hyperactivity disorder in children in New Zealand") — linkage study found rural living and nature exposure independently associated with lower ADHD. Engemann et al. (2020, Environmental Health Perspectives, PMC7755168) — Danish nationwide cohort (n=814,689) using NDVI vegetation measures found residential green space in early childhood associated with lower clinical ADHD diagnosis. Sakhvidi et al. (2022, Environment International) — systematic review confirming nature exposure associated with reduced ADHD prevalence and severity. A 2024 Science of the Total Environment paper ("Nurturing attention through nature") confirms differential benefits of trees vs. grass vs. water exposure.
Mechanisms include not only ART but also reduced stress (Kuo 2015), microbial diversity (Aerts et al. 2018; Prehn-Kristensen 2018 on gut microbiome), and reduced air pollution exposure.

Built environment as regulator: Kuo & Sullivan (2001, Environment and Behavior, "Aggression and violence in the inner city: impacts of environment via mental fatigue") showed barren built environments produce attentional fatigue → aggression — implicating the built environment itself as a regulatory variable on dysregulated behavior. This is a critical citation: environmental impoverishment functions as a chronic depleter of the very regulatory resources ADHD lacks.

Gibsonian affordances: James Gibson's The Ecological Approach to Visual Perception (1979) — that environments offer affordances for action that scaffold or constrain behavior — has been applied to developmental psychology (Adolph; Heft's Ecological Psychology in Context, 2001), to extended cognition (Clark & Chalmers' "The Extended Mind," 1998; Hutchins' Cognition in the Wild, 1995), and to neurodiversity broadly (Davidson & Henderson on autism; some recent work on sensory affordances). It is rarely explicitly linked to ADHD regulatory needs. The specific claim that the built environment functions as a regulatory scaffold for the ADHD nervous system — that affordances can substitute for executive function the person does not have — is theoretically defensible and consistent with extended/embodied cognition but is not a developed empirical literature in ADHD specifically.

Housing instability and ADHD outcomes: Less directly studied. Sandstrom and Huerta's work for the Urban Institute on housing instability and child outcomes; Coley, Leventhal, and colleagues' work on housing quality and child mental health; and the broader ACES (adverse childhood experiences) literature document housing instability as a risk factor for externalizing symptoms and attention problems generally. The specific link between residential instability and ADHD trajectory is plausible and consistent with the affordance-scaffold argument (each move erases the regulatory environment a child has learned to function within), but I did not locate dedicated empirical studies of housing instability × ADHD trajectory. This is another place where the paper can articulate a tractable empirical prediction.

Recommendation: Cite Kaplan & Kaplan 1989; Kuo & Taylor 2004; Faber Taylor & Kuo 2009; Kuo & Sullivan 2001; Donovan et al. 2019; Engemann et al. 2020; Sakhvidi et al. 2022. Frame "environmental affordances as ADHD regulatory scaffold" as a productive theoretical extension combining established Gibsonian ecological psychology, extended-mind philosophy, and the empirically supported ART–ADHD literature. Note that housing-instability-as-affordance-loss is a clean hypothesis awaiting study.

6. Shame, Perfectionism, and Project Abandonment in ADHD

Verdict: Partially supported. The component constructs (RSD, perfectionism-as-shield, schema-based shame, abandonment behaviors) are widely described in clinical and qualitative literature; the specific behavioral pattern of initiation → stall → abandonment-restart (rather than iterating through frustration) is well-documented clinically but under-researched empirically.

Rejection Sensitive Dysphoria (RSD): A construct popularized by William Dodson (psychiatrist, ADDitude magazine; multiple clinical publications) describing intense emotional pain triggered by perceived rejection/criticism/failure in adults with ADHD. Notably, RSD is not a DSM-5 diagnosis and remains a clinical/descriptive construct rather than an empirically validated syndrome. Prevalence estimates (up to 99% in adults with ADHD) come from Dodson's clinical samples and should be treated cautiously. A peer-reviewed qualitative study (Bedrossian and colleagues, 2024–2025, available as PMC12822938, "The lived experience of rejection sensitivity in ADHD") used focus groups with five undergraduates and identified themes of withdrawal, masking, and bodily sensations. This is genuinely new empirical work but small-N and qualitative. Related broader emotional dysregulation literature in ADHD (Shaw et al. 2014, American Journal of Psychiatry, "Emotion Dysregulation in Attention Deficit Hyperactivity Disorder"; Faraone et al. 2019) provides the broader empirical foundation that ADHD involves emotional dysregulation as a core rather than incidental feature.

Perfectionism and ADHD: Multiple studies (Strohmeier et al.; Stoeber et al. on perfectionism dimensions; Hewitt & Flett's tripartite model of self-oriented, other-oriented, and socially prescribed perfectionism) link socially prescribed perfectionism to procrastination and avoidance. Clinical literature consistently describes ADHD perfectionism as "perfectionism as a shield against criticism" — preferring not to start to risking imperfect output. Wheeler, Blankstein and others have shown perfectionism mediates between trait neuroticism and procrastination. Stolte et al. (2022) and others have begun to specifically examine perfectionism in adult ADHD samples.

Young's Schema Therapy and the Defectiveness/Shame domain: Jeffrey Young's Schema Therapy (Young, Klosko, & Weishaar 2003) includes a Disconnection/Rejection domain containing the Defectiveness/Shame schema. Research on schemas in ADHD is limited but growing — Philipsen and colleagues have applied schema-related frameworks to adult ADHD, and Miklósi et al. (2016) have examined early maladaptive schemas in adult ADHD specifically (e.g., Journal of Attention Disorders). The Defectiveness/Shame schema, formed through childhood experiences of being "not enough," is theoretically a strong fit for ADHD adults who have been criticized throughout development for executive function failures attributed to character.

The specific behavioral pattern — initiate → stall → abandon → restart: This is widely described in popular and clinical ADHD literature (Russell Barkley's executive function model of ADHD; Brendan Mahan's "Wall of Awful" framework; Hallowell & Ratey's Driven to Distraction and ADHD 2.0). However, I could not locate a peer-reviewed study empirically measuring this specific cyclic pattern (i.e., demonstrating that ADHD individuals show higher rates of abandonment-and-restart vs. iteration-through-frustration on standardized creative or academic tasks). The closest empirical evidence is the literature on:

Task persistence and frustration tolerance in ADHD (Wilbertz et al.; numerous behavioral studies showing reduced persistence under negative feedback).
Effort-based decision-making (Volkow's work showing ADHD adults discount effort more steeply, mediated by mesolimbic dopamine; Tripp & Wickens on dopamine transfer deficit).
Project completion difficulties documented in occupational and academic functioning studies (Barkley & Murphy; Adler et al.).
Novelty seeking and reward immediacy (Sonuga-Barke's delay-aversion model).

Recommendation: The specific abandonment-restart pattern as the paper describes it is essentially a clinical/phenomenological observation that synthesizes (a) executive function deficits in effort allocation, (b) RSD-driven shame avoidance, (c) perfectionism-as-shield, and (d) the Defectiveness/Shame early maladaptive schema. This synthesis is genuinely novel as a unified theoretical claim, but the paper should be transparent that it is integrating four empirically-supported components rather than citing a single study that has measured the cyclic pattern itself. This is a reasonable place for the paper to advocate for an empirical research program. Cite: Dodson clinical work on RSD; Shaw et al. 2014 and Faraone et al. 2019 on ADHD emotional dysregulation; Barkley on executive function; Young, Klosko & Weishaar on schemas; Miklósi on schemas in ADHD; Sonuga-Barke on delay aversion; the Bedrossian qualitative study (PMC12822938) for lived-experience evidence; and acknowledge that the abandonment-restart pattern remains a clinically observed signature in need of formal measurement.

7. Technological Impermanence and Social Capital

Verdict: Partially supported. There is robust research on (a) memorialization of deceased users on platforms and (b) the transition from social-graph to interest-graph logic, but the specific claim that platform death destroys lateral social capital encoded in early platforms (ICQ, MSN Messenger, AIM, early Friendster/MySpace/Bebo) is genuinely under-researched.

What exists:

Death and mourning on platforms: A large literature on Facebook memorial pages, R.I.P. pages, and digital mourning (Brubaker; Walter et al. 2012, Omega — Journal of Death and Dying, "Does the internet change how we die and mourn?"; Giaxoglou & Döveling 2018; Klastrup; the Journal of Computer-Mediated Communication literature including 2023 work on Facebook memorial pages as social capital resources). This addresses what happens to social capital when users die, not when platforms die.

Social graph vs. interest graph: Eugene Wei's industry analysis (extensively referenced in Ian Bogost's Atlantic essays and in ArtReview coverage) argues that platforms like TikTok have replaced the "social graph" (interpersonal connections that constituted potential social capital) with an "interest graph" derived from surveillance. This is a structural observation about how platform design itself can dissolve social-capital-generating affordances even while users remain present. It is journalistic-theoretical rather than empirical but is widely cited.

Social media use and social capital: A large literature originating in Nicole Ellison, Cliff Lampe, Charles Steinfield, and colleagues' foundational work on Facebook and bridging/bonding social capital (e.g., Ellison, Steinfield, & Lampe 2007, Journal of Computer-Mediated Communication, "The benefits of Facebook 'friends': Social capital and college students' use of online social network sites"); extended by Burke, Marlow, and Lento, and connected to Putnam's Bowling Alone (2000) distinction. This documents that platform engagement creates social capital but says relatively little about what happens when the platform infrastructure disappears.

Platform decay/migration: Some work in HCI and platform studies (e.g., Gehl on alternative social media; danah boyd's work — It's Complicated, 2014 — on MySpace-to-Facebook migration patterns; Anne Helmond's "platformization" framework; José van Dijck's The Culture of Connectivity, 2013) addresses platform change. boyd's MySpace migration analysis addresses class-coded platform abandonment but not the specific destruction of relational graphs.

Cory Doctorow's "enshittification" (2023, Wired and elsewhere) provides a popular framework for platform decay but is not specifically about social capital loss.

Geert Lovink's Sad by Design (2019) and Stuck on the Platform (2022) address networked melancholy and platform-stuckness; Sherry Turkle's Alone Together (2011) addresses the brittleness of mediated intimacy but not specifically platform death.

The gap the paper identifies is real:

When ICQ (effectively defunct by 2024), MSN Messenger (closed 2013), AIM (closed 2017), Friendster (2015), Google+ (2019), Vine (2017), Orkut (2014), Bebo (multiple deaths and revivals), Yahoo Messenger (2018), Path, Peach, Ello, and dozens of smaller platforms ceased operation, the non-portable relationships encoded in those graphs — contact lists, status histories, dyadic conversational threads, identity performances tied to a username that no longer resolves — were simply destroyed. Unlike the death of a person on a persistent platform, the death of a platform leaves relationships in an ambiguous state: the people are alive but the channel through which the relationship was constituted is gone, and there is typically no clean migration path. This question is asked obliquely but is not the subject of a dedicated peer-reviewed empirical literature. The closest are:

Discussions of "digital legacy" (mostly framed around individual death and platform memorialization policies).
HCI work on data ownership and portability (Vertesi et al.; some GDPR-related work).
Tristan Harris / Center for Humane Technology work on engagement design (focused on attention extraction, not relationship destruction).
Some platform-studies work on MySpace's decline (Gillespie; boyd 2014).
Lost in the Cloud / digital preservation literature (Internet Archive scholarship; Kahle's work; Lialina on "vernacular web").

Account instability and identity: There is a small literature on social media account deletion, "Facebook suicide," and platform-quitting (Baumer et al. 2013, "Limiting, leaving, and (re)lapsing: an exploration of Facebook non-use practices and experiences"; Schoenebeck on "media leaving"; Stieger et al. on Facebook quitting and well-being). This documents that quitting is often associated with mental-health benefits but also with felt social isolation. Frequent deletion-and-restart patterns are described clinically (often in the context of social anxiety, RSD, eating disorders, and borderline personality features) but I could not locate a peer-reviewed study specifically measuring whether frequent account deletion-restart predicts social isolation or identity instability cross-sectionally or longitudinally. The phenomenon is described in popular ADHD literature (account-deletion as an RSD-driven impulse — see Dodson's clinical advice literally including "Do not delete your social media account" as a recommended pause during RSD episodes) and in clinical accounts of borderline and avoidant personality features (Linehan; the broader self-disturbance literature in BPD).

Recommendation: The paper has identified an under-theorized phenomenon. Frame the platform-death-as-social-capital-destruction claim as an extension of existing work on (a) Ellison/Steinfield/Lampe-style platform-mediated social capital, (b) Wei/Bogost on social-graph-to-interest-graph migration, (c) boyd on platform migration, (d) van Dijck on the culture of connectivity, and (e) Lovink on networked melancholy and platform-stuckness — while acknowledging that the specific claim awaits empirical investigation. This is a productive original contribution. For the account-instability sub-claim, draw on the RSD literature from Section 6 (impulsive account deletion as an RSD behavior) and on clinical descriptions of identity instability, but be similarly transparent that the cross-sectional empirical work is thin. The Baumer et al. line on non-use practices is the strongest existing footing.

Cross-Cutting Synthesis: Where the Paper Is Most Original

The paper's individual claims rest on substantial but fragmented existing literature. Its most distinctive intellectual contribution is the integration itself — specifically:

The McLuhan–allostasis bridge (Section 2) is the single most novel theoretical move. The figure/ground inversion provides the allostatic literature with a vocabulary it lacks: a way to describe what happens to the perceptual environment when the hedonic set-point shifts, not just to the neural circuitry. This deserves to be the paper's central methodological banner.

The five-stage cascade specification (Section 1) integrates work that is normally siloed: developmental psychopathology (Masten/Cicchetti), epidemiological gateway models (Kandel), neurobiological priming (Kandel & Hurd; Spear), screen-time neuroimaging (ABCD studies), and ADHD substance-use prospective work (Lee meta-analysis; Ilbegi). No single existing study models all five stages.

The constitutive vs. modificatory addiction distinction (Section 3) refines but does not duplicate Dingle et al.'s identity-gain/loss pathway model — extending it from a recovery-stage observation to a developmental claim about what was never built in the first place.

The ADHD × scarcity multiplicative interaction (Section 4) is a clean empirical prediction that follows from existing theory but has not been directly tested.

Affordances as regulatory scaffold (Section 5) extends ART, Gibsonian ecological psychology, and extended-mind frameworks into ADHD-specific territory in a way that has not been developed at theoretical length.

The RSD–perfectionism–schema–abandonment-restart cycle (Section 6) synthesizes four component literatures into a unified behavioral signature.

Platform death as social-capital destruction (Section 7) identifies a genuinely under-theorized phenomenon at the intersection of platform studies, social capital theory, and digital sociology.
The paper should treat the gaps not as weaknesses but as the precise places where it is contributing — and should be explicit about this in framing. The strongest version of the manuscript would explicitly cite the foundational empirical literature in each section (so reviewers see the scholarly grounding) and then clearly demarcate the synthetic/extensional moves as theoretical contributions, especially the McLuhan-allostasis bridge.
Methodological caveats to acknowledge in the manuscript:

The gateway hypothesis is contested cross-nationally and the common-liability/gateway debate remains open; the paper should acknowledge both rather than asserting gateway-as-fact.
RSD lacks DSM validation and most prevalence estimates are clinical rather than epidemiological.
The screen-time → ADHD literature has bidirectional/MR evidence suggesting genetic confounding (ADHD polygenic risk predicts screen time); deterministic framing should be avoided.
The cascade-modeling literature has identified methodological weaknesses (small samples, short follow-ups, parental-socialization focus); empirical confirmation of the full five-stage cascade does not yet exist.
The scarcity literature has had partial replication failure (Camerer's 2018 Many Labs replication vs. Shah et al. 2012 on cognitive fatigue).
The ART/nature-ADHD literature is largely correlational with small experimental supplements; causal direction remains under-determined.

The Lee et al. (2011) meta-analysis specifically did not find childhood ADHD predicted lifetime alcohol use (though it did predict alcohol use disorder*), a nuance worth preserving.

Naming these openly will make the manuscript more defensible and let the genuinely original contributions stand more clearly. The paper's argumentative arc — that distinct conditions are successive expressions of one cascade unified by McLuhanian ground-restructuring — is theoretically coherent, empirically motivated, and currently unwritten in this integrated form. It should be presented confidently as a perspective piece making an original theoretical contribution that is well-grounded in (but not duplicative of) the seven adjacent empirical literatures surveyed here.