pith. sign in

arxiv: 2404.04755 · v1 · submitted 2024-04-06 · 🧬 q-bio.QM · q-bio.NC

A probabilistic model of relapse in drug addiction

Sayun Mao , Tom Chou , Maria D'Orsogna This is my paper

Pith reviewed 2026-05-24 02:23 UTC · model grok-4.3

classification 🧬 q-bio.QM q-bio.NC
keywords relapse probabilitydrug addictionprobabilistic modelpositive activationpeak-end rulecontinuous contentmentlife eventsrecovery interventions
0
0 comments X

The pith

A mathematical model of drug relapse finds that mild continuous contentment protects against return to use more effectively than large but infrequent positive events.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper builds a probabilistic model of relapse by combining the positive activation-negative activation framework with the peak-end rule to create a relapse rate that incorporates the timing, intensity, and ordering of life events along with individual response traits. This rate is then used to compare different sequences of stressors, drug cues, and positive experiences, identifying which patterns most strongly raise or lower the chance of relapse within a recovery period. The central result is that steady, low-level positive experiences outperform sporadic intense ones as a protective factor. Readers would care because relapse rates exceed 60 percent in the first year after treatment and the model supplies concrete, adjustable parameters for designing interventions around daily experience patterns.

Core claim

The authors construct a relapse rate by integrating positive and negative activations with the peak-end rule across sequences of external events, modulated by individual mental-response traits. Systematic comparison of event combinations shows that continuous mild positive experiences produce the lowest relapse probabilities, while large episodic happiness is less protective; the same framework identifies orderings of stressors and cues that maximize relapse risk and supplies intervention designs to minimize it.

What carries the argument

The relapse rate function that combines positive activation-negative activation values with the peak-end rule applied to timed external events and individual trait weights.

If this is right

  • Continuous mild positive events produce lower relapse probabilities than episodic intense ones across the modeled parameter space.
  • Specific orderings and timings of stressors and cues raise relapse probability more than others.
  • Interventions can be designed around the intensity and continuity of positive experiences rather than their peak magnitude.
  • Individual trait differences modulate how strongly any given event sequence affects relapse risk.
  • The model supplies quantitative rankings of protective versus risk-increasing event patterns.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same rate construction could be applied to other repeated behavioral decisions that depend on remembered emotional peaks and averages.
  • Real-world diary or sensor data on event timing could be used to test whether the predicted ordering effects appear in actual relapse statistics.
  • Extending the model with time-varying individual traits might reveal windows when mild continuous support is most effective.
  • The framework could be compared against purely memory-free reinforcement models to isolate the contribution of the peak-end component.

Load-bearing premise

The positive activation-negative activation paradigm and the peak-end rule can be combined with external event factors and individual traits to produce accurate probabilities of relapse.

What would settle it

A longitudinal study that records sequences of daily positive experiences and subsequent relapse outcomes, then finds equal or higher relapse rates among participants with continuous mild contentment than among those with only episodic large positive events.

Figures

Figures reproduced from arXiv: 2404.04755 by Maria D'Orsogna, Sayun Mao, Tom Chou.

Figure 1
Figure 1. Figure 1: Mental state at time t, M(t), and the probability of relapse before time T, P(T), upon exposure to a single stressor {−B1, t b 1 } or to a single positive event {A1, t a 1 } for three processing rates κb or κa = 2, 1, 0.5/day with B1 = A1 = 4, t b 1 , t a 1 = 7 days, R0 = 10−3 /day , M0 = 0. The relapse probability decreases with κb so that the longer a stressor impacts one’s mental state, the larger the l… view at source ↗
Figure 2
Figure 2. Figure 2: Top row: Mental state M(t) and relapse probability P(T) upon exposure to two stressors {−B1, t b 1 } and {−B2, t b 2 } separated by lag times ∆b = t b 2 −t b 1 = 0, 1, 2, 4 days and obtained using Eqs. 1, 2, 4, and 12. Parameters are κb = 1/day, B1 = B2 = 4, R0 = 10−3 /day, M0 = 0. All stressor pairs define the same integrated mental state defined in Eq. 5 and are equivalent to the single event {−Zb, tb} s… view at source ↗
Figure 3
Figure 3. Figure 3: Top row: Mental state M(t) and relapse probability P(T) upon exposure to nb = 1, 2, 4 negative events {−Bi , t b i } and a continuum of stressors. All curves are obtained using Eqs. 1, 2, 4, and 12. Sequences define the same integrated mental state defined in Eq. 5 and are equivalent to the single event {−Zb, tb} shown in the red curve, where Zb = 8 and tb = 7 days. Within each sequence, events carry the s… view at source ↗
Figure 4
Figure 4. Figure 4: Top row: Mental state M(t) and relapse probability P(T) upon exposure to a stressor {−B1, t b 1 } followed by a positive event {A2, t a 2 }. The two events are separated by lag times ∆ba = t a 2 − t b 1 = 0, 0.2, 0.4, 0.6 hours. All curves are obtained using Eqs. 1, 2, 4, and 14. Parameters are κb = κa = κ = 1/day, B1 = 4, A2 = 2, R0 = 10−3 /day, M0 = 0 and by setting H = 2e 7 . All event pairs define the … view at source ↗
Figure 5
Figure 5. Figure 5: Top row: Mental state M(t) and relapse probability P(T) upon exposure to two events of opposite sign {Ai , t a i } and {−Bj , t b j } for κa = 2/day and κb = 1/day where i = 1, j = 2 or j = 1, i = 2. All curves are obtained using Eqs. 1, 2, 4, and 14. Events carry the amplitudes Ai = Bj = 4; the other parameters are set at M0 = 0, R0 = 10−3 . Lag times are evaluated by setting t b j = 7 days where j = 1 fo… view at source ↗
Figure 6
Figure 6. Figure 6: Long-term percentage changes in the relapse probab [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Dynamics of hR(t)i averaged over 100,000 realizations of the Ornstein￾Uhlenbeck described by Eq. 19 and the analytical expression for hR(t)i in Eq. 26a. Parameters are M0 = 0, R0 = 10−3 /day. Values in the legend are in units of /day. For λ = 2Y the t → ∞ relapse rate in Eq. 26b hR(t → ∞)i = R0 matches the neutral case of no exposure to any positive or negative life event. input Y must obey Y > λ/2. Given … view at source ↗
Figure 8
Figure 8. Figure 8: Expected relapse probability hP(T)i as derived from averaging over 5,000 realizations of the Ornstein-Uhlenbeck process in Eq. 19 and the analytical approximation in Eq. 30 for the baseline Y = λ = 0 with R0 = 10−3 , κ = 1/day. Values of λ, Y displayed along the curves are in units of /day. In panel (a) we set Y = 1/day; as λ is increased hP(T)i also increases. For λ = 2Y results from the baseline are reco… view at source ↗
Figure 9
Figure 9. Figure 9: Top row: Mean first passage time Tm(M = 0 → M = Mth) for an initial mental state M = 0 to reach the negative threshold Mth = −2 as a function of the decay rate κ. Results follow from the random process in Eq. 19 and the analytical form in Eq. 32. For large κ, each random fluctuation dissipates quickly so that the effects of multiple inputs to the mental state do not accumulate appreciably, hence Tm(M) incr… view at source ↗
Figure 10
Figure 10. Figure 10: Relapse probability P(T = 100) following exposure to a single, randomly drawn sequence of sensory cues occurring at times generated according to a Poisson process of rate λc. The corresponding relapse rate is evaluated through Eq. 38 and the relapse probability at T = 100 days, P(T = 100), is evaluated using Eqs. 2. The red curves depict the analytical approximation obtained for hP(T = 100)i = 0.5 calcula… view at source ↗
read the original abstract

More than 60% of individuals recovering from substance use disorder relapse within one year. Some will resume drug consumption even after decades of abstinence. The cognitive and psychological mechanisms that lead to relapse are not completely understood, but stressful life experiences and external stimuli that are associated with past drug-taking are known to play a primary role. Stressors and cues elicit memories of drug-induced euphoria and the expectation of relief from current anxiety, igniting an intense craving to use again; positive experiences and supportive environments may mitigate relapse. We present a mathematical model of relapse in drug addiction that draws on known psychiatric concepts such as the "positive activation; negative activation" paradigm and the "peak-end" rule to construct a relapse rate that depends on external factors (intensity and timing of life events) and individual traits (mental responses to these events). We analyze which combinations and ordering of stressors, cues, and positive events lead to the largest relapse probability and propose interventions to minimize the likelihood of relapse. We find that the best protective factor is exposure to a mild, yet continuous, source of contentment, rather than large, episodic jolts of happiness.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript presents a probabilistic model of relapse in drug addiction integrating the positive activation-negative activation (PAN) paradigm and peak-end rule with external event intensities/timings and two individual-trait parameters. It analyzes sequences of stressors, cues, and positive events to compute relapse probabilities and concludes that mild continuous contentment is the optimal protective factor over episodic large positive events.

Significance. If the model's functional form accurately captures psychological dynamics, the result could inform targeted interventions in addiction recovery by prioritizing continuous mild positive experiences. The work provides a quantitative framework combining established psychiatric concepts, which is a strength for generating testable predictions about event ordering, though its applied significance hinges on future empirical grounding.

major comments (2)
  1. [Model construction] Model construction (as described in the methods): The relapse probability is formed by combining PAN, peak-end rule, event factors, and two free parameters for individual traits, but the manuscript reports no calibration to observed relapse trajectories and no comparison to alternative aggregations such as total integral of activation. This is load-bearing for the central claim, as the superiority of mild continuous contentment is a direct output of this specific construction.
  2. [Results] Results section on protective factors: No sensitivity analysis is provided on the choice of peak-end rule versus other functional forms or on the scaling factors for event intensity/timing. Any mismatch between the assumed combination rule and actual dynamics would reverse the reported ordering, undermining the intervention proposal.
minor comments (1)
  1. [Abstract] The abstract states 'more than 60%' relapse within one year; a supporting citation would strengthen the motivation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments. The manuscript develops a theoretical probabilistic framework rather than an empirically fitted model; we address the two major comments below and indicate where revisions will be made.

read point-by-point responses

  1. Referee: [Model construction] Model construction (as described in the methods): The relapse probability is formed by combining PAN, peak-end rule, event factors, and two free parameters for individual traits, but the manuscript reports no calibration to observed relapse trajectories and no comparison to alternative aggregations such as total integral of activation. This is load-bearing for the central claim, as the superiority of mild continuous contentment is a direct output of this specific construction.

    Authors: The model is constructed as a theoretical integration of the PAN paradigm and peak-end rule to generate qualitative predictions about event sequences; it is not intended as a data-driven statistical fit. No calibration to relapse trajectories is reported because the work analyzes the internal consequences of the chosen functional form rather than estimating parameters from observations. We will revise the discussion to explicitly state the theoretical scope, note the absence of empirical calibration, and outline how future data could be used to test or refine the aggregation rule. revision: partial

  2. Referee: [Results] Results section on protective factors: No sensitivity analysis is provided on the choice of peak-end rule versus other functional forms or on the scaling factors for event intensity/timing. Any mismatch between the assumed combination rule and actual dynamics would reverse the reported ordering, undermining the intervention proposal.

    Authors: We agree that robustness checks are valuable. The peak-end rule is adopted from the cited psychological literature, yet alternative aggregations (e.g., time integrals) could alter quantitative outcomes. In the revised manuscript we will add a dedicated sensitivity-analysis subsection that varies the weighting between peak and end, the scaling of event intensities, and the timing decay parameters, reporting how these changes affect the ranking of protective strategies. revision: yes

Circularity Check

0 steps flagged

Theoretical model constructed from external psychiatric concepts; analysis result follows directly without reduction to fitted inputs or self-citations

full rationale

The paper defines a relapse rate by combining the positive activation-negative activation paradigm and peak-end rule (described as known psychiatric concepts) with external event factors and individual traits, then performs mathematical analysis over sequences to identify the ordering that minimizes integrated relapse probability. No parameters are reported as fitted to relapse trajectories, no self-citations are invoked as load-bearing uniqueness theorems, and the headline ordering emerges as a mathematical consequence of the chosen functional form rather than an independent prediction. The derivation chain is therefore self-contained as a theoretical construction and does not exhibit any of the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on the two psychiatric concepts as domain assumptions and likely free parameters for individual differences and event quantification. No new entities are introduced. Assessment limited by abstract-only access.

free parameters (2)
  • mental response parameters for individual traits
    The model incorporates individual traits as mental responses to events, which are likely free parameters fitted or chosen to model different people.
  • scaling factors for event intensity and timing
    External factors' intensity and timing affect the relapse rate, implying parameters to quantify their impact.
axioms (2)
  • domain assumption The positive activation-negative activation paradigm can be used to model mental responses to events
    Explicitly drawn upon to construct the relapse rate.
  • domain assumption The peak-end rule governs how the intensity and timing of events influence craving and relapse decisions
    Used in building the model of relapse probability.

pith-pipeline@v0.9.0 · 5728 in / 1439 out tokens · 35762 ms · 2026-05-24T02:23:22.955572+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

64 extracted references · 64 canonical work pages

  1. [1]

    Cisewski, Lauren M

    Farida Bhuiya Ahmad, Jodi A. Cisewski, Lauren M. Rossen, and Paul Sut- ton. Provisional drug overdose death counts. National Cent er for Health Statistics, 2022

    work page 2022

  2. [2]

    Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Behavio ral Health Statistics and Quality, NSDUH-2021-DS0001

    U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Behavio ral Health Statistics and Quality, NSDUH-2021-DS0001. National Surv ey on Drug Use and Health 2021. https: //datafiles.samhsa.gov, 2021

    work page 2021

  3. [3]

    V olkow and Marisela Morales

    Nora D. V olkow and Marisela Morales. The brain on drugs: F rom reward to addiction. Cell, 162(4):712–725, 2015

    work page 2015

  4. [4]

    Treatment and recover y

    National Institute on Drug Abuse. Treatment and recover y. https://nida.nih.gov/publications/drugs-brains-behavior-science- addiction/treatment-recovery, 2022

    work page 2022

  5. [5]

    Thomas McLellan, David C

    A. Thomas McLellan, David C. Lewis, Charles P . O’ Brien, a nd Herbert D. Kleber. Drug dependence, a chronic medical illness: Implications for treat- ment, insurance, and outcomes evaluation. Journal of the American Med- ical Association, 284:1689–1695, 2000

    work page 2000

  6. [6]

    New findings on biological factors predict ing addiction re- lapse vulnerability

    Rajita Sinha. New findings on biological factors predict ing addiction re- lapse vulnerability. Current Psychiatry Reports, 13:398–405, 2011

    work page 2011

  7. [7]

    Brecht and Diane Herbeck

    Mary L. Brecht and Diane Herbeck. Time to relapse followi ng treatment for methamphetamine use: a long-term perspective on patter ns and predic- tors. Drug and Alcohol Dependence, 139:18–25, 2014

    work page 2014

  8. [8]

    Smyth, Joseph Barry, Eamon Keenan, and Kevin Duc ray

    Bobby P . Smyth, Joseph Barry, Eamon Keenan, and Kevin Duc ray. Lapse and relapse following inpatient treatment of opiate dependence. Irish Med- ical Journal, 103:176–179, 2010

    work page 2010

  9. [9]

    Urman, Elyse M

    Alan David Kaye, Richard D. Urman, Elyse M. Cornett Corne tt, and Am- ber Edinoff. Substance Use and Addiction Research: Methodology, Mech- anisms, and Therapeutics. Academic Press, London, UK, 2023

    work page 2023

  10. [10]

    D’Orsogna, Lucas B¨ ottcher, and Tom Chou

    Maria R. D’Orsogna, Lucas B¨ ottcher, and Tom Chou. Fent anyl-driven acceleration of racial, gender and geographical disparities in drug overdose deaths in the United States. PLOS Global Public Health , 3:e0000769, 2023

    work page 2023

  11. [11]

    Modeling the role of environment in addiction

    Daniele Caprioli, Michele Celentano, Giovanna Paolon e, and Aldo Bal- diani. Modeling the role of environment in addiction. Progress in Neuro- Psychopharmacology and Biological Psychiatry, 31:1639–1653, 2007

    work page 2007

  12. [12]

    V olkow, Joanna S

    Nora D. V olkow, Joanna S. Fowler, and Gene-Jack Wang. Th e addicted human brain: insights from imaging studies. The Journal of Clinical In- vestigation, 111:1444–1451, 2003

    work page 2003

  13. [13]

    Koob and Nora D

    George F. Koob and Nora D. V olkow. Neurocircuitry of add iction. Neu- ropsychopharmacology: o fficial publication of the American College of Neuropsychopharmacology, 35(1):217–238, jan 2010

    work page 2010

  14. [14]

    Goldstein and Nora D

    Rita Z. Goldstein and Nora D. V olkow. Dysfunction of the prefrontal cor- tex in addiction: neuroimaging findings and clinical implic ations. Nature Reviews Neuroscience, 12(11):652–669, 2011

    work page 2011

  15. [15]

    Koob and Nora D

    George F. Koob and Nora D. V olkow. Neurobiology of addic tion: a neu- rocircuitry analysis. The Lancet. Psychiatry, 3:760–773, 2016. 17

    work page 2016

  16. [16]

    Mollick and Hedy Kober

    Jessica A. Mollick and Hedy Kober. Computational model s of drug use and addiction: A review. Journal of Abnormal Psychology, 129:544–555, 2020

    work page 2020

  17. [17]

    D’Orsogna

    Tom Chou and Maria R. D’Orsogna. A mathematical model of reward- mediated learning in drug addiction. Chaos, 32:021102, 2022

    work page 2022

  18. [18]

    Gutkin, Stanislas Dehaene, and Jean P

    Boris S. Gutkin, Stanislas Dehaene, and Jean P . Changeu x. A neurocom- putational hypothesis for nicotine addiction. Proceedings of the National Academy of Sciences, 103:1106–1111, 2006

    work page 2006

  19. [19]

    Intermittent adaptation: A mathematic al model of drug tolerance, dependence and addiction

    Abraham Peper. Intermittent adaptation: A mathematic al model of drug tolerance, dependence and addiction. In Boris S. Gutkin and Serge H. Ahmed, editors, Computational Neuroscience of Drug Addiction, Springer Series in Computational Neuroscience, volume 10, chapter 2, pages 19–56. Springer, New Y ork, NY , 2012

    work page 2012

  20. [20]

    Buchanich, Mark S

    Hawre Jalal, Jeanine M. Buchanich, Mark S. Roberts, Lau ren C. Balmert Balmert, Kun Zhang Zhang, and Donald S. Burke. Changing dyna mics of the drug overdose epidemic in the United States from 1979 thr ough 2016. Science, 361:6408, 2018

    work page 1979

  21. [21]

    D’Orsogna

    Lucas B¨ ottcher, Tom Chou, and Maria R. D’Orsogna. Mode ling and fore- casting age-specific overdose mortality in the United State s. European Physics Journal Special Topics, 232:1743–1752, 2023

    work page 2023

  22. [22]

    D’Orsogna

    Lucas B¨ ottcher, Tom Chou, and Maria R. D’Orsogna. Fore casting drug- overdose mortality by age in the United States at the nationa l and county levels. PNAS Nexus, 3:pgae050, 2024

    work page 2024

  23. [23]

    How does stress increase risk of drug abus e and relapse? Psychopharmacology, 158:343–359, 2001

    Rajita Sinha. How does stress increase risk of drug abus e and relapse? Psychopharmacology, 158:343–359, 2001

    work page 2001

  24. [24]

    Evans and Catherine M

    Christopher J. Evans and Catherine M. Cahill. Neurobio logy of opioid dependence in creating addiction vulnerability. F1000Research, 5:1748, 2016

    work page 2016

  25. [25]

    George F. Koob. Neurobiology of opioid addiction: Oppo nent process, hyperkatifeia, and negative reinforcement. Biological Psychiatry, 87:44– 53, 2020

    work page 2020

  26. [26]

    Ghahremani, Mark A

    Lili Nie, Dara G. Ghahremani, Mark A. Mandelkern, Andy C . Dean, Wei Luo, Anlian Ren, Jing Li, and Edythe D. London. The relations hip be- tween duration of abstinence and gray-matter brain structu re in chronic methamphetamine users. American Journal of Drug and Alcohol Abuse , 47:65–73, 2021

    work page 2021

  27. [27]

    George F. Koob. Anhedonia, hyperkatifeia, and negativ e reinforcement in substance use disorders. In Diego A. Pizzagalli, editor, Anhedonia: Pre- clinical, Translational, and Clinical Integration. Springer, Cham, Switzer- land, 2022

    work page 2022

  28. [28]

    Association of drug cues and craving with drug use and relapse; A systematic review and meta-anal ysis

    Nilofar V afaie and Hedy Kober. Association of drug cues and craving with drug use and relapse; A systematic review and meta-anal ysis. JAMA Psychiatry, 79:641–650, 2022

    work page 2022

  29. [29]

    Kahneman

    Daniel L. Kahneman. Evaluation by moments, past and fut ure. In Choices, V alues and Frames, pages 693–708. Cambridge University Press, 2000

    work page 2000

  30. [30]

    Fredrickson and Daniel L

    Barbara L. Fredrickson and Daniel L. Kahneman. Duratio n neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65:45–55, 1993

    work page 1993

  31. [31]

    Chasing the first high

    Aaron M. Bornstein and Hanna Pickard. “Chasing the first high”: Memory sampling in drug choice. Neuropsychopharmacology, 45:907–915, 2020

    work page 2020

  32. [32]

    McCabe, James A

    Sean E. McCabe, James A. Cranford, and Carol J. Boyd. Str essful events and other predictors of remission from drug dependence in th e United States: Longitudinal results from a national survey. Journal of Substance Abuse Treatment, 71:41–47, 2016

    work page 2016

  33. [33]

    Perry, Isabel Zbukvic, Jee H

    Christina J. Perry, Isabel Zbukvic, Jee H. Kim, and Andr ew J. Lawrence. Role of cues and contexts on drug-seeking behaviour. British Journal of Pharmacology, 171:4636–4672, 2014

    work page 2014

  34. [34]

    Tunstall, Lauren E

    Rajtarun Madangopal, Brendan J. Tunstall, Lauren E. Ko mer, Sophia J. Weber, Jennifer K. Hoots, V eronica A. Lennon, Jennifer M. Bo ssert, David H. Epstein, Y avin Shaham, and Bruce T. Hope. Discriminative stim- uli are su fficient for incubation of cocaine craving. eLife, 8:e44427, 2019

    work page 2019

  35. [35]

    Parson s, Simon Katner, Xiu Liu, Eric P

    Freidbert Weiss, Roberto Ciccocioppo, Loren H. Parson s, Simon Katner, Xiu Liu, Eric P . Zorrilla, Glenn R. V aldez, Osnat Ben-Shahar , Stefania Angeletti, and Regina R. Richter. Compulsive drug-seeking behavior and relapse. Neuroadaptation, stress, and conditioning facto rs. Annals of the New York Academy of Sciences, 937:1–26, 2001

    work page 2001

  36. [36]

    Koob and Michel Le Moal

    George F. Koob and Michel Le Moal. Neurobiological mech anisms for opponent motivational processes in addiction. Philosophical Transactions of the Royal Society B: Biological Sciences , 363(1507):3113–3123, 2008

    work page 2008

  37. [37]

    Mehdi Keramati and Boris S. Gutkin. Homeostatic reinfo rcement learn- ing for integrating reward collection and physiological st ability. eLife, 3:e04811, 2014

    work page 2014

  38. [38]

    Crombag, and David N

    Theodora Duka, Hans S. Crombag, and David N. Stephens. E xperimen- tal medicine in drug addiction: Towards behavioral, cognit ive and neuro- biological biomarkers. Journal of Psychopharmacology , 25:1235–1255, 2011

    work page 2011

  39. [39]

    The two general activation systems of a ffect: Structural findings, evolutionary con- siderations, and psychobiological evidence

    David Watson, David Wiese, Jatin V aidya, and Auke Telle gen. The two general activation systems of a ffect: Structural findings, evolutionary con- siderations, and psychobiological evidence. Journal of Personality and Social Psychology, 76:820–838, 1999

    work page 1999

  40. [40]

    Gross and Robert W

    James J. Gross and Robert W. Levenson. Emotion elicitat ion using films. Cognition and Emotion, 9:87–108, 1995

    work page 1995

  41. [41]

    Ito, John T

    Ti ffany A. Ito, John T. Cacioppo, and Peter J. Lang. Eliciting a ffect using the international a ffective picture system: Trajectories through evaluative space. Personality and Social Psychology Bulletin, 24:855–879, 1998

    work page 1998

  42. [42]

    Cacioppo, W

    John T. Cacioppo, W. L. Gardner, and G. G. Berntson. The a ffect system has parallel and integrative processing components: Form f ollows func- tion. Journal of Personality and Social Psychology, 76:839–855, 1999

    work page 1999

  43. [43]

    Peter J. Lang. The emotion probe. studies of motivation and attention. American Psychologist, 50:372–385, 1995

    work page 1995

  44. [44]

    Ito, Je ff T

    Ti ffany A. Ito, Je ff T. Larsen, N. Kyle Smith, and John T. Cacioppo. Neg- ative information weighs more heavily on the brain: The nega tivity bias in evaluative categorizations. Journal of Personality and Social Psychology , 75:887–900, 1998

    work page 1998

  45. [45]

    Nega- tive and positive life events and their relation to substanc e and behavioral addictions

    Noam Zilberman, Gal Y adid, Y aniv Efrati, and Y uri Rasso vsky. Nega- tive and positive life events and their relation to substanc e and behavioral addictions. Drug and Alcohol Dependence , 204:107562, 2019

    work page 2019

  46. [46]

    Who be- comes addicted and to what? psychosocial predictors of subs tance and behavioral addictive disorders

    Noam Zilberman, Gal Y adid, Y aniv Efrati, and Y uri Rasso vsky. Who be- comes addicted and to what? psychosocial predictors of subs tance and behavioral addictive disorders. Psychiatry Research, 291:113221, 2020

    work page 2020

  47. [47]

    Koob, Patricia Powell, and Aaron White

    George F. Koob, Patricia Powell, and Aaron White. Addic tion as a coping response: Hyperkatifeia, deaths of despair, and COVID-19. American Journal of Psychiatry, 177:1031–1037, 2020

    work page 2020

  48. [48]

    George F. Koob. Drug addiction: Hyperkatifeia /Negative reinforcement as a framework for medications development. Pharmacological Review, 73:163–201, 2021

    work page 2021

  49. [49]

    Imaging stress- an d cue-induced drug and alcohol craving: association with relapse and clin ical implica- tions

    Rajita Sinha and Chaing-Shan Ray Li. Imaging stress- an d cue-induced drug and alcohol craving: association with relapse and clin ical implica- tions. Drug and Alcohol Review, 26:25–31, 2007

    work page 2007

  50. [50]

    Fox, Keri L

    Helen C. Fox, Keri L. Bergquist, Kwang I. Hong, and Rajit a Sinha. Stress- induced and alcohol cue-induced craving in recently abstin ent alcohol dependent individuals. Alcohol Clinincal and Experimental Research , 31:395–403, 2007

    work page 2007

  51. [51]

    National practice gu ideline for the use of medications in the treatment of addiction involving opioid use

    Kyle Kampman and Margaret Jarvis. National practice gu ideline for the use of medications in the treatment of addiction involving opioid use. Jour- nal of Addiction Medicine , 9:358–367, 2015

    work page 2015

  52. [52]

    Uhlenbeck and Leonard S

    George E. Uhlenbeck and Leonard S. Ornstein. On the theo ry of Brownian motion. Physical Review, 36:823–841, 1930

    work page 1930

  53. [53]

    Gardiner

    Crispin W. Gardiner. Handbook of Stochastic Methods . Springer-V erlag, Berlin, 4th edition, 2009

    work page 2009

  54. [54]

    The F okker–Planck Equation: Methods of Solution and Applications

    Hannes Risken. The F okker–Planck Equation: Methods of Solution and Applications. Springer-V erlag, New Y ork, 1989

    work page 1989

  55. [55]

    Mean exit time and escape probability for t he Orn- stein–Uhlenbeck process

    Caibin Zeng. Mean exit time and escape probability for t he Orn- stein–Uhlenbeck process. Chaos, 30(9):093127, 2020

    work page 2020

  56. [56]

    Fox, Makram Talih, Robert Malison, George M

    Helen C. Fox, Makram Talih, Robert Malison, George M. An derson, Mary J Kreek, and Rajita Sinha. Frequency of recent cocaine a nd alcohol use affects drug craving and associated responses to stress and drug-related cues. Psychoneuroendocrinology, 30:880–891, 2005

    work page 2005

  57. [57]

    Dynamica l systems in computational psychiatry: A toy-model to apprehend the dyn amics of psy- chiatric symptoms

    Christophe Gauld and Damien Depannemaecker. Dynamica l systems in computational psychiatry: A toy-model to apprehend the dyn amics of psy- chiatric symptoms. Frontiers in Psychology, 14:1099257, 2023

    work page 2023

  58. [58]

    T he peak-end rule and its dynamic realization through di fferential equations with max- ima

    Olena Trofymchuk, Eduardo Liz, and Sergei Trofimchuk. T he peak-end rule and its dynamic realization through di fferential equations with max- ima. Nonlinearity, 36:507–536, 2023

    work page 2023

  59. [59]

    Mathemat ical mod- eling of depressive disorders: Circadian driving, bistabi lity and dynamical transitions

    Xiaoou Cheng, Maria R D’Orsogna, and Tom Chou. Mathemat ical mod- eling of depressive disorders: Circadian driving, bistabi lity and dynamical transitions. Computational and Structural Biotechnology Journal, 19:664– 690, 2020

    work page 2020

  60. [60]

    Hauser, V asilisa Skvortsova, Munmun De Choudhury, and Niko- laos Koutsouleris

    Tobias U. Hauser, V asilisa Skvortsova, Munmun De Choudhury, and Niko- laos Koutsouleris. The promise of a model-based psychiatry : Building computational models of mental ill health. Lancet Digital Health, 4:e816– e828, 2022

    work page 2022

  61. [61]

    Kim, Maria R

    Lae U. Kim, Maria R. D’Orsogna, and Tom Chou. Onset, timi ng, and exposure therapy of stress disorders: Mechanistic insight from a mathe- 18 matical model of oscillating neuroendocrine dynamics. Biology Direct, 11:13, 2016

    work page 2016

  62. [62]

    Random walkers with extreme value mem ory: Model- ing the peak-end rule

    Rosemary Harris. Random walkers with extreme value mem ory: Model- ing the peak-end rule. New Journal of Physics , 17:053049, 2015

    work page 2015

  63. [63]

    V olkow, Linda Chang, Gene-Jack Wang, Joanna S

    Nora D. V olkow, Linda Chang, Gene-Jack Wang, Joanna S. F owler, Dinko Franceschi, Mark Sedler, Samuel J. Gatley, Eric Miller, Robert Hitzemann, Y u-Shin Ding, et al. Loss of dopamine transporters in metham phetamine abusers recovers with protracted abstinence. Journal of Neuroscience , 21(23):9414–9418, 2001

    work page 2001

  64. [64]

    Allen, Murat Y ucel, and Dan I

    Ali Cheetham, Nicholas B. Allen, Murat Y ucel, and Dan I. Lubman. The role of a ffective dysregulation in drug addiction. Clinical Psychology Re- view, 30:621–634, 2010. 19

    work page 2010