OSCR

Mesocorticostriatal Reinforcement Learning of State Representation and Value with Implications for the Mechanisms of Schizophrenia.

Code ↔ Paper

2 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 2 matches · all tie a paragraph to a whole file, not to given lines: weak matches, whose lines are not tinted
  1. [1] § Results › Simulated inputs to the striatum and possible correspondences to experimental results ↔ OVRNNRB1probrew.m, the whole file · a weak match · score 0.55 · fMRI, BOLD signal, sum, striatal units, RNN units, corticostriatal
  2. [2] § Results › Simulated inputs to the striatum and possible correspondences to experimental results ↔ OVRNNRB1probrew.m, the whole file · a weak match · score 0.54 · fMRI, BOLD signal, proxy, sum, striatum, reward

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

MATLAB · 128 lines · 5 KB · no license · 2 matches

  1. function [os,xs,Vs,VDs,ds,w,wsd,bdc,bds,wbrs,P,meanPs,CDtoS,ef] = OVRNNRB1probrew(brdmode,modeltype,abase,g,drs,rewp,ndim,ns,nd,nt,P,iniPmean,plotyn)
  3. % OVRNNRB1probrew : code for probabilistic reward task
  4. % brdmode: 1:general, 2:two DA units receiving exclusively from two rewards, 3:brdmode 2 + another DA unit receiving evenly from two rewards
  5. % modeltype: 1:oVRNNrf-bio, 2:untrained RNN
  6. % abase: learning rate base, e.g., 0.1
  7. % g: time discount factor, e.g., 0.8
  8. % drs: decay rate for [cortico-striatal, striato-dopamine] weight per time-step, e.g., [0 0], [0.0002, 0.0002]
  9. % rewp: reward probability, e.g., 2/3
  10. % ndim: number of RNN units, e.g., 40
  11. % ns: number of striatal units, e.g., 10
  12. % nd: number of DA units, e.g., 5
  13. % nt: number of trials, e.g., 4000
  15. ITIs = NaN(1,nt);
  16. cue = [];
  17. rew = [];
  18. for k = 1:nt
  19. tmp = randperm(4);
  20. ITIs(k) = 2 + tmp(1);
  21. tmp_rand = rand;
  22. tmp_rand2 = rand;
  23. if tmp_rand <= 0.5
  24. cue = [cue [1;0] zeros(2,3+ITIs(k))];
  25. if tmp_rand2 <= rewp
  26. rew = [rew zeros(2,3) [1;0] zeros(2,ITIs(k))];
  27. else
  28. rew = [rew zeros(2,3) [0;0] zeros(2,ITIs(k))];
  29. end
  30. else
  31. cue = [cue [0;1] zeros(2,3+ITIs(k))];
  32. if tmp_rand2 <= rewp
  33. rew = [rew zeros(2,3) [0;1] zeros(2,ITIs(k))];
  34. else
  35. rew = [rew zeros(2,3) [0;0] zeros(2,ITIs(k))];
  36. end
  37. end
  38. end
  39. os = [cue; rew]; % observations
  41. if isempty(P)
  42. P = randn(ndim,ndim+4) + iniPmean; % initialization of the weights on the RNN units (combined A and B)
  43. end
  44. w = zeros(ns,ndim);
  45. wsd = zeros(nd,ns); % ones(nd,ns)/ns; %rand(nd,ns); % connections from striatum to dopamine neurons
  46. bdc = rand(ndim,nd)/nd; % fixed random connections from dopamine neurons to RNN units
  47. bds = rand(ns,nd)/nd; % fixed random connections from dopamine neurons to striatal neurons
  48. if brdmode == 1
  49. brd = rand(nd,2); % fixed random connections from the two rewards to dopamine neurons
  50. elseif brdmode == 2
  51. if nd ~= 2
  52. error('number of DA units (nd) must be 2 for brdmode=2');
  53. else
  54. brd = [1 0; 0 1];
  55. end
  56. elseif brdmode == 3
  57. if nd ~= 3
  58. error('number of DA units (nd) must be 3 for brdmode=3');
  59. else
  60. brd = [1 0; 0 1; 0.5 0.5];
  61. end
  62. else
  63. error('brdmode must be 1, 2, or 3');
  64. end
  66. ef = [0 0]; % exit flag
  67. tmpc101t = find(sum(cue,1),101,'first'); cue101time = tmpc101t(end); % time-step of cue at 101st trial
  68. tmax = 4*nt + sum(ITIs);
  69. CDtoS = NaN(2,tmax); % [Cortex_to_Striatum; Dopamine_to_Striatum]: sum of these (total inputs to striatum) is a proxy of fMRI BOLD signal at striatum
  70. Vs = NaN(ns,tmax);
  71. VDs = NaN(nd,tmax);
  72. ds = NaN(nd,tmax);
  73. Vs(:,1) = zeros(ns,1);
  74. xs = NaN(ndim,tmax);
  75. xs(:,1) = rand(ndim,1);
  76. wbrs = NaN(2,tmax); % correlation coefficient of [w&bdc; wsd&bds]
  77. meanPs = NaN(1,tmax); % time-development of mean(mean(P))
  78. meanPs(1) = mean(mean(P));
  79. for t = 2:tmax
  80. vec1 = reshape((wsd*w)',nd*ndim,1);
  81. vec2 = reshape(bdc,nd*ndim,1);
  82. [r,p] = corrcoef(vec1,vec2); wbrs(1,t) = r(1,2);
  83. vec1 = reshape(wsd',nd*ns,1);
  84. vec2 = reshape(bds,nd*ns,1);
  85. [r,p] = corrcoef(vec1,vec2); wbrs(2,t) = r(1,2);
  86. xs(:,t) = 1./(1 + exp(-P*[xs(:,t-1);os(:,t-1)]));
  87. Vs(:,t) = w * xs(:,t);
  88. if t >= 3
  89. VDs(:,t) = wsd * Vs(:,t);
  90. ds(:,t-1) = brd*rew(:,t-1) + g*wsd*Vs(:,t) - wsd*Vs(:,t-1);
  91. CDtoS(1,t) = sum(Vs(:,t)); % Cortex_to_Striatum
  92. CDtoS(2,t) = sum(bds*ds(:,t-1)); % Dopamine_to_Striatum
  93. wnew = max(0, w + (abase/(ndim/12))*(bds*ds(:,t-1))*xs(:,t-1)');
  94. wsdnew = max(0, wsd + (abase/(ndim/12))*ds(:,t-1)*Vs(:,t-1)');
  95. if modeltype == 1
  96. %P = P + abase*ds(t-1)*(xs(:,t-1).*(1-xs(:,t-1)).*b')*[xs(:,t-2);os(:,t-2)]';
  97. y = (xs(:,t-1)<0.5).*xs(:,t-1).*(1-xs(:,t-1)) + (xs(:,t-1)>=0.5)*0.5*0.5;
  98. P = P + abase*(y.*(bdc*ds(:,t-1)))*[xs(:,t-2);os(:,t-2)]';
  99. end
  100. w = wnew;
  101. wsd = wsdnew;
  102. end
  103. meanPs(t) = mean(mean(P));
  104. w = w * (1 - drs(1));
  105. wsd = wsd * (1 - drs(2));
  106. if (t>=cue101time) && (max(max(w))==0)
  107. ef(1) = 1;
  108. end
  109. if (t>=cue101time) && (max(max(wsd))==0)
  110. ef(2) = 1;
  111. end
  112. end
  113. if plotyn
  114. F = figure;
  115. A = axes;
  116. hold on;
  117. set(F,'Position',[65 75 600 780]);
  118. subplot(6,1,1); hold on; plot(VDs(1,end-50:end),'r'); plot(VDs(2,end-50:end),'b');
  119. if nd >= 3
  120. plot(VDs(3,end-50:end),'g');
  121. end
  122. subplot(6,1,2); hold on; plot(cue(1,end-50:end),'r:'); plot(rew(1,end-50:end),'r'); plot(cue(2,end-50:end),'b:'); plot(rew(2,end-50:end),'b');
  123. subplot(6,1,3); hold on; plot(CDtoS(1,end-50:end),'m'); plot(CDtoS(2,end-50:end),'c'); plot(sum(CDtoS(:,end-50:end),1),'k');
  124. subplot(6,1,4); hold on; plot([1:tmax],wbrs(1,:));
  125. subplot(6,1,5); hold on; plot([1:tmax],wbrs(2,:));
  126. %subplot(6,1,5); hold on; plot([1:ns],bds(:,1),'r'); plot([1:ns],wsd(1,:),'b'); plot([ns+1:2*ns],bds(:,2),'m'); plot([ns+1:2*ns],wsd(2,:),'c');
  127. subplot(6,1,6); hold on; plot([1:tmax],meanPs);
  128. end

OVRNNRB1probrew.m at commit 81d0d33, no license · at the source

Overview

Authors: Kenji Morita1,2, Arvind Kumar3,4
ORCID iDs: Kenji Morita
  1. Physical and Health Education, Graduate School of Education, The University of Tokyo, Tokyo 113-0033, Japan
  2. International Research Center for Neurointelligence (WPI-IRCN), The University of Tokyo, Tokyo 113-0033, Japan
  3. Division of Computational Science and Technology, School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm SE-100 44, Sweden
  4. Science for Life Laboratory, Solna 17121, Sweden
Institutions: The University of Tokyo (Japan); Science for Life Laboratory (Sweden); KTH Royal Institute of Technology (Sweden)
Journal: The Journal of neuroscience : the official journal of the Society for Neuroscience, volume 46, issue 16, article e1762252026
Dates: received 15 September 2025; accepted 24 February 2026; published online 22 April 2026; in print 22 April 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1523/jneurosci.1762-25.2026 · PMID 41775629 · PMCID PMC13244656 · OpenAlex W7133338368
Open access: hybrid, a free copy (OpenAlex)
Status: code verified
Categories: human (organism), schizophrenia / psychosis (population), computational (subfield)
Methods: Statistics
Keywords: dopamine, excitation/inhibition balance, feedback alignment, recurrent neural networks, reinforcement learning, schizophrenia
MeSH: Cerebral Cortex*, Corpus Striatum*, Models, Neurological*, Reinforcement Machine Learning*, Reinforcement, Psychology*, Schizophrenia*, Animals, Dopamine, Humans, Motivation, Reward (* major topic)
Journal subjects: Behavioral/Cognitive
Topic: Schizophrenia research and treatment (Psychiatry and Mental health, Medicine), according to OpenAlex
Funding: Japan Society for the Promotion of Science (JSPS) (25H02594); Strategic research area StratNeuro
Citations: not cited yet (Europe PMC); 121 references in the paper

Abstract

Mesocorticostriatal dopamine projections are crucial for value learning, motivational control, and cognitive functions. However, while dopamine's role in value learning as reward-prediction-error (RPE) has been much understood, precise roles in motivational control and cognitive functions remain more elusive. Computationally, this corresponds to that while the operation of mesostriatal dopamine could be minimally described by simple reinforcement learning (RL) models with one-dimensional reward/RPE and fixed state representation, (1) how reward-specific motivational control can be achieved through heterogeneous dopamine responses, and (2) how sophisticated cortical state representation can be formed through mesocortical dopamine, cannot be captured by such simple models. To address both of these at once, we combined recent models for each of them: the “Reward Bases (RB),” which achieved reward-specific motivational control through multidimensional RPE (but with fixed cortical representation), and the “online value-recurrent-neutral-network (OVRNN),” which achieved state representation learning through training of RNN by RPE (but of one-dimensional). We show the combined model can achieve both functions simultaneously via double “feedback alignments” of the cortical and striatal downstream connections to the mesocorticostriatal dopamine projections. Crucially, cortical inhibition-dominance is a key for successful learning. Excessive excitation leads to aberrant persistent activity, which disrupts the alignments and impairs reward-specific motivational control and credit assignment. This implies how negative and positive symptoms of schizophrenia could emerge from excitation/inhibition imbalance, and we show how our model could explain altered brain activations in patients. Our model thus provides an integrated computational account for dopamine's functions, with implications on how its dysfunctions link to schizophrenia.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 2 matches between paragraphs and lines of code.

kenjimoritagithub/OVRNN-RB

License: none: the authors keep all their rights
State: the link answers, verified on 29 September 2026
Evidence: files inventoried
Commit: 81d0d336ef92ccd1557322262ec7f6031adcf3e8, 5 March 2026
Languages: MATLAB (21)
Size: 22 files, 21 scripts
Software Heritage: not archived
Found in: “Code accessibility”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Availability: 1 check, the latest on 29 September 2026: the link answers
  • 29 September 2026: the link answers
22 files
At the source: github.com/kenjimoritagithub/OVRNN-RB

Code accessibility

The codes for simulations and analyses are available at GitHub: https://github.com/kenjimoritagithub/OVRNN-RB.

Reproduced under the paper's license (CC BY), from the paper cited above.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 21 scripts, each with its path and the digest of its content;
  • 2 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 29 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 2 authors, 6 keywords, 11 MeSH terms, 2 funders, 121 references.

Cite

This paper

Morita, K., & Kumar, A. (2026). Mesocorticostriatal Reinforcement Learning of State Representation and Value with Implications for the Mechanisms of Schizophrenia. The Journal of neuroscience : the official journal of the Society for Neuroscience, 46(16), e1762252026. https://doi.org/10.1523/jneurosci.1762-25.2026

BibTeX

@article{morita2026mesocorticostriatal,
author = {Morita, Kenji and Kumar, Arvind},
title = {{Mesocorticostriatal Reinforcement Learning of State Representation and Value with Implications for the Mechanisms of Schizophrenia}},
journal = {The Journal of neuroscience : the official journal of the Society for Neuroscience},
year = {2026},
month = apr,
volume = {46},
number = {16},
pages = {e1762252026},
publisher = {Society for Neuroscience},
issn = {0270-6474},
doi = {10.1523/jneurosci.1762-25.2026},
url = {https://doi.org/10.1523/jneurosci.1762-25.2026},
pmid = {41775629},
pmcid = {PMC13244656}
}

RIS

TY - JOUR
AU - Morita, Kenji
AU - Kumar, Arvind
TI - Mesocorticostriatal Reinforcement Learning of State Representation and Value with Implications for the Mechanisms of Schizophrenia
T2 - The Journal of neuroscience : the official journal of the Society for Neuroscience
J2 - J Neurosci
PY - 2026
DA - 2026/04/22
VL - 46
IS - 16
SP - e1762252026
SN - 0270-6474
PB - Society for Neuroscience
DO - 10.1523/jneurosci.1762-25.2026
UR - https://doi.org/10.1523/jneurosci.1762-25.2026
LA - en
ER -

CSL-JSON

{
"id": "10.1523/jneurosci.1762-25.2026",
"type": "article-journal",
"title": "Mesocorticostriatal Reinforcement Learning of State Representation and Value with Implications for the Mechanisms of Schizophrenia",
"container-title": "The Journal of neuroscience : the official journal of the Society for Neuroscience",
"author": [
{
"family": "Morita",
"given": "Kenji"
},
{
"family": "Kumar",
"given": "Arvind"
}
],
"container-title-short": "J Neurosci",
"volume": "46",
"issue": "16",
"page": "e1762252026",
"DOI": "10.1523/jneurosci.1762-25.2026",
"PMID": "41775629",
"PMCID": "PMC13244656",
"ISSN": "0270-6474",
"publisher": "Society for Neuroscience",
"URL": "https://doi.org/10.1523/jneurosci.1762-25.2026",
"language": "en",
"issued": {
"date-parts": [
[
2026,
4,
22
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s41467-026-73994-1 [code]
Prediction error correlates in the striosome-dopamine circuit emerge from information gain.
Journal: Nature communications
In common: 14 references
[2] doi:10.1016/j.isci.2026.117130 [code]
Region-specific weighting of sensory intensity and reward prediction error by dopamine signals.
Journal: iScience
In common: 7 references
[3] doi:10.1038/s42003-026-10485-5 [code]
Dopamine dynamics as a regulatory mechanism for shifting between defensive and reward-seeking behaviors.
Journal: Communications biology
In common: 6 references
[4] doi:10.7554/elife.106506 [code]
Modeling flexible behavior with remapping-based hippocampal sequence learning.
Journal: eLife
In common: schizophrenia / psychosis, 4 references
[5] doi:10.1002/hbm.70545 [code]
Striatal Prediction Error Tracking Moderates the Influence of Social Exposure on Adolescent Substance Use Curiosity.
Journal: Human brain mapping
In common: 3 references
[6] doi:10.1038/s42256-026-01254-4 [code]
Neural sampling from cognitive maps enables goal-directed imagination and planning.
Journal: Nature machine intelligence
In common: 3 references
[7] doi:10.1038/s41586-026-10512-9 [code]
Astrocyte glucocorticoid receptor signalling restricts neuronal plasticity.
Journal: Nature
In common: 3 references
[8] doi:10.1371/journal.pcbi.1014730 [code]
A unified model of short- and long-term plasticity: Effects on network connectivity and information capacity.
Journal: PLoS computational biology
In common: 3 references
[9] doi:10.1038/s41467-026-71664-w [code]
Dorsal prefrontal cortex drives perseverative behavior in mice.
Journal: Nature communications
In common: 3 references
[10] doi:10.1093/schbul/sbaf171
Fetal Head Growth and Head Circumference at Birth in Children of Women with Psychotic Disorders and Population-Based Controls.
Journal: Schizophrenia bulletin
In common: schizophrenia / psychosis, 2 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.