pith. sign in

arxiv: 1907.04191 · v2 · pith:N6OWNOHZnew · submitted 2019-07-08 · 💻 cs.HC · cs.SI

Belief places and spaces: Mapping cognitive environments

Philip Feldman , Aaron Dant , Wayne Lutters This is my paper

Pith reviewed 2026-05-25 01:16 UTC · model grok-4.3

classification 💻 cs.HC cs.SI
keywords belief placesbelief spacescognitive environmentsagent-based modelingfantasy role-playing gamesnarrative environmentsmisinformation
0
0 comments X

The pith

Repeated behaviors in a shared fantasy role-playing game can produce maps of collective belief places and distinct belief spaces.

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

This paper tries to establish a method for mapping cognitive environments by tracking how people behave repeatedly in the same fictional setting. The maps would show what a group agrees on as the shared place while highlighting where individuals or subgroups diverge in their space. A sympathetic reader would care because it offers a way to visualize conflicting beliefs without forcing them into stories, which could help with issues like misinformation. The work uses an agent-based model of fantasy RPG play as the testbed. It claims the techniques are simple enough to apply to real-world data like online discussions.

Core claim

By observing the repeated behaviors of human participants in the same social context, it is possible to build maps that show the shared narrative environment overlaid with traces that show unique, individual or subgroup perspectives. Our contribution is a proof-of-concept system, based on the affordances of fantasy tabletop role-playing games, which support multiple groups interacting with the same dungeon in a controlled, online environment. The techniques used in this process are mathematically straightforward, and should be generalizable to auto-generating larger-scale maps of belief spaces from other corpora, such as discussions on social media.

What carries the argument

An agent-based simulation model that turns repeated behaviors in a fantasy RPG into belief places (agreed salient features) and belief spaces (related but distinct perspectives).

If this is right

  • Belief maps can be generated that overlay unique perspectives on a shared narrative environment.
  • The approach is generalizable to other data sources such as social media discussions.
  • These maps could aid navigation through conflicting factive statements.
  • Controlled environments allow multiple groups to interact with the same setting for comparable results.

Where Pith is reading between the lines

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

  • Applying this to non-fictional contexts like political debates might show how beliefs cluster without relying on game mechanics.
  • Such maps could be used to test interventions aimed at reducing belief divergence in online communities.
  • The straightforward math might enable real-time mapping of belief spaces from live data streams.

Load-bearing premise

That the behaviors in the simulated fantasy RPG setting accurately represent how beliefs are held and expressed by humans in real social contexts.

What would settle it

If independent measures of participants' beliefs, such as direct questionnaires about the same dungeon elements, do not match the positions shown on the generated maps, the claim would be falsified.

Figures

Figures reproduced from arXiv: 1907.04191 by Aaron Dant, Philip Feldman, Wayne Lutters.

Figure 1
Figure 1. Figure 1: Simulations (top) and associated maps (bottom) [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: A play-by-post submission including character actions, speech, dice rolls, and OOC com [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: A real-time chat example with interleaved conversation and dice rolls between multiple [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Data explorer with term embeddings oped for BoW were extended to incorporate TF-IDF “Documents” in this case were the indi￾vidual posts when analyzing single players, players in groups, and groups within time slices 4.5 User evaluation To close the loop on human usability of the maps, we developed a survey for the participants of the dungeon, hosted on Google Forms. This survey was run after all groups had… view at source ↗
Figure 5
Figure 5. Figure 5: Text extraction process lists were created, term centrality [32] was calculated using the terms from all users within a split. This produced the initial term list shown in table 2. Group 1 Group 2 Group 3 Split 1 goblin, arrow stair, spell behind, goblin Split 2 statue, halfling statue, goblin short, goblin Split 3 troll, chest statue, switch piton, around Split 4 grogg, troll troll, sleep grogg, troll Spl… view at source ↗
Figure 7
Figure 7. Figure 7: Convergence of terms were normalized to account for the different online systems that stored the text. Using the iterative text analytic shown in figure 4, we found that the top terms for each sequence across groups stabi￾lized quite quickly [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Initial Belief Places & Spaces map 5 focusing the discussion to those points and no consensus would be reached among the players about what each point means.” (Yanadar: Group 2) Question No Maybe Yes Helpful? 26.7% 13.3% 60% Recognition? 33.3% 66.7% Accuracy? 7.1% 35.7% 57.1% Effective? 13.3% 46.7% 40% [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Final map6 and keep both the old woman and the boy in starving captivity. The problem is carefully worded to imply that the party must choose the boy or the woman, while also allowing that there might be a possibility that one of the party can take their place. The trolley problem has experienced a resurgence in recent years as it provides an ethical frame￾work to understand how we should train/program aut… view at source ↗
read the original abstract

Beliefs are not facts, but they are factive - they feel like facts. This property is what can make misinformation dangerous. Being able to deliberately navigate through a landscape of often conflicting factive statements is difficult when there is no way to show the relationships between them without incorporating the information in linear, narrative forms. In this paper, we present a mechanism to produce maps of belief places, where populations agree on salient features of fictional environments, and belief spaces, where subgroups have related but distinct perspectives. Using a model developed using agent-based simulation, we show that by observing the repeated behaviors of human participants in the same social context, it is possible to build maps that show the shared narrative environment overlaid with traces that show unique, individual or subgroup perspectives. Our contribution is a proof-of-concept system, based on the affordances of fantasy tabletop role-playing games, which support multiple groups interacting with the same dungeon in a controlled, online environment. The techniques used in this process are mathematically straightforward, and should be generalizable to auto-generating larger-scale maps of belief spaces from other corpora, such as discussions on social media.

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 / 0 minor

Summary. The paper claims to present a mechanism, developed via agent-based simulation in a controlled fantasy tabletop RPG setting, for producing maps of 'belief places' (shared narrative environments where populations agree on salient features) and 'belief spaces' (overlaid traces showing unique individual or subgroup perspectives). By observing repeated participant behaviors in the same social context, the approach is said to generate these maps; the techniques are described as mathematically straightforward and generalizable to other corpora such as social media discussions. The work is positioned as a proof-of-concept.

Significance. If the agent-based model were validated against real participant data and shown to produce maps that genuinely reflect cognitive environments rather than game-specific artifacts, the contribution could be significant for HCI and related fields by offering a non-linear visualization tool for navigating conflicting beliefs and misinformation. The use of RPGs as a controlled, multi-group testbed is a creative affordance, but the absence of any empirical grounding or mathematical detail in the presented work limits immediate impact.

major comments (2)
  1. [Abstract] Abstract: the central claim that 'by observing the repeated behaviors of human participants in the same social context, it is possible to build maps that show the shared narrative environment overlaid with traces' rests on an unvalidated assumption that the agent-based simulation accurately captures real human belief expression; no calibration, human-vs-simulation match metrics, sensitivity analysis, or participant data comparisons are supplied to rule out simulation-specific dynamics such as rule-following artifacts.
  2. [Abstract] Abstract: the statement that 'the techniques used in this process are mathematically straightforward' is unsupported, as the abstract (and thus the manuscript) supplies no equations, derivations, formal definitions of the mapping process, or error analysis to demonstrate how behavioral observations are transformed into the described maps.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful review of our proof-of-concept paper. We address each major comment below, clarifying the intended scope of the simulation-based exploration while agreeing to revisions where the presentation can be improved.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that 'by observing the repeated behaviors of human participants in the same social context, it is possible to build maps that show the shared narrative environment overlaid with traces' rests on an unvalidated assumption that the agent-based simulation accurately captures real human belief expression; no calibration, human-vs-simulation match metrics, sensitivity analysis, or participant data comparisons are supplied to rule out simulation-specific dynamics such as rule-following artifacts.

    Authors: The manuscript is framed as a proof-of-concept that develops the mapping mechanism through agent-based simulation in a controlled RPG environment. This allows demonstration of the core idea under controlled conditions before scaling to noisier real-world data. We do not claim empirical validation against human participants in the current work and have added an explicit limitations paragraph in the revised manuscript discussing the simulation's role and the need for future calibration and human-subject studies. The central claim is therefore presented as a demonstration of possibility rather than a validated result. revision: partial

  2. Referee: [Abstract] Abstract: the statement that 'the techniques used in this process are mathematically straightforward' is unsupported, as the abstract (and thus the manuscript) supplies no equations, derivations, formal definitions of the mapping process, or error analysis to demonstrate how behavioral observations are transformed into the described maps.

    Authors: We agree that the manuscript provides only a conceptual description of the mapping process without formal equations or error analysis. We have revised the abstract to remove the unsupported claim of mathematical straightforwardness and now describe the approach in terms of its conceptual accessibility for a proof-of-concept. No formal mathematical treatment is added at this stage, as the contribution centers on the application of the idea rather than its formalization. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper presents a conceptual proof-of-concept for mapping belief environments via agent-based simulation of RPG behaviors, with no equations, derivations, fitted parameters, or predictions that reduce to inputs by construction. No self-citations, uniqueness theorems, or ansatzes are invoked as load-bearing elements, and the techniques are described as mathematically straightforward without any renaming of known results or self-definitional loops. The central claim rests on the simulation producing maps from observed behaviors, but this does not exhibit the enumerated circularity patterns and remains self-contained as a descriptive system without forcing equivalence to its own assumptions.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that behaviors in a controlled fantasy game environment serve as a valid proxy for real cognitive belief structures, and that the authors' agent-based model correctly encodes the mapping process. No free parameters or invented entities are specified in the abstract.

axioms (1)
  • domain assumption Repeated behaviors in a shared social context reliably reveal both common and divergent belief structures.
    Invoked in the description of how maps are built from participant actions in the RPG setting.

pith-pipeline@v0.9.0 · 5721 in / 1257 out tokens · 20792 ms · 2026-05-25T01:16:08.742631+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

49 extracted references · 49 canonical work pages · 5 internal anchors

  1. [1]

    The moral machine experiment

    A WAD, E., D SOUZA , S., K IM, R., S CHULZ , J., H ENRICH , J., S HARIFF , A., B ONNEFON , J.-F., AND RAHWAN , I. The moral machine experiment. Nature 563, 7729 (2018), 59

    work page 2018

  2. [2]

    Role-playing games for opening the black box of multi-agent systems: method and lessons of its application to senegal river valley irrigated systems

    B ARRETEAU , O., B OUSQUET , F., ATTONATY, J.-M., ET AL . Role-playing games for opening the black box of multi-agent systems: method and lessons of its application to senegal river valley irrigated systems. Journal of artificial societies and social simulation 4, 2 (2001), 5

    work page 2001

  3. [3]

    W., AND BOOS , M

    B ELZ , M., P YRITZ , L. W., AND BOOS , M. Spontaneous flocking in human groups. Be- havioural processes 92 (2013), 6–14

    work page 2013

  4. [4]

    A theory of fads, fashion, custom, and cultural change as informational cascades

    B IKHCHANDANI , S., H IRSHLEIFER , D., AND WELCH , I. A theory of fads, fashion, custom, and cultural change as informational cascades. Journal of political Economy 100 , 5 (1992), 992–1026

    work page 1992

  5. [5]

    A revolution in military cartography?: Europe 1650-1815

    B LACK , J. A revolution in military cartography?: Europe 1650-1815. The Journal of Military History 73, 1 (2009), 49–68

    work page 2009

  6. [6]

    Second person: Role-playing and story in games and playable media

    B ORGSTROM , R. Second person: Role-playing and story in games and playable media. In Structure and Meaning in Roleplaying Game Design. The MIT Press, 2010. 7antibubbles.com/tymora-thanks.html 19

    work page 2010

  7. [7]

    Natural language processing (almost) from scratch

    C OLLOBERT , R., W ESTON , J., B OTTOU , L., K ARLEN , M., K AVUKCUOGLU , K., AND KUKSA , P. Natural language processing (almost) from scratch. Journal of machine learn- ing research 12, Aug (2011), 2493–2537

    work page 2011

  8. [8]

    Emergent behavior in flocks

    C UCKER , F., AND SMALE , S. Emergent behavior in flocks. IEEE Transactions on automatic control 52, 5 (2007), 852–862

    work page 2007

  9. [9]

    J., AND WAGNER , R

    D ANCHIN , ´E., G IRALDEAU , L.-A., V ALONE , T. J., AND WAGNER , R. H. Public informa- tion: from nosy neighbors to cultural evolution. Science 305, 5683 (2004), 487–491

    work page 2004

  10. [10]

    Dungeons for Science: Mapping Belief Places and Spaces

    D ANT, A., F ELDMAN , P., AND LUTTERS , W. Dungeons for science: Mapping belief places and spaces. arXiv preprint arXiv:1904.05216 (2019)

    work page internal anchor Pith review Pith/arXiv arXiv 1904

  11. [11]

    Collective patterns and decision-making

    D ENEUBOURG , J.-L., AND GOSS , S. Collective patterns and decision-making. Ethology Ecology & Evolution 1, 4 (1989), 295–311

    work page 1989

  12. [12]

    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

    D EVLIN , J., C HANG , M.-W., L EE, K., AND TOUTANOVA , K. Bert: Pre-training of deep bidi- rectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018)

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  13. [13]

    A., S MYTH , J

    D ILLMAN , D. A., S MYTH , J. D., AND CHRISTIAN , L. M. Internet, phone, mail, and mixed- mode surveys: the tailored design method. John Wiley & Sons, 2014

    work page 2014

  14. [14]

    E PSTEIN , R., AND ROBERTSON , R. E. The search engine manipulation effect (seme) and its possible impact on the outcomes of elections. Proceedings of the National Academy of Sciences 112, 33 (2015), E4512–E4521

    work page 2015

  15. [15]

    E WALT, D. M. Of dice and men: the story of Dungeons & Dragons and the people who play it. Simon and Schuster, 2014

    work page 2014

  16. [16]

    What is a diagram? In 2007 11th International Conference Information Visualization (IV’07) (2007), IEEE, pp

    F ATHULLA , K., AND BASDEN , A. What is a diagram? In 2007 11th International Conference Information Visualization (IV’07) (2007), IEEE, pp. 951–956

    work page 2007

  17. [17]

    With maps and mobs: Searching for trustworthiness using belief spaces

    F ELDMAN , P. With maps and mobs: Searching for trustworthiness using belief spaces. In Proceedings of the 2018 Conference on Human Information Interaction & Retrieval (2018), CHIIR ’18, pp. 351–353

    work page 2018

  18. [18]

    This one simple trick disrupts digital com- munities

    F ELDMAN , P., D ANT, A., AND LUTTERS , W. This one simple trick disrupts digital com- munities. In 2018 IEEE 12th International Conference on Self-Adaptive and Self-Organizing Systems (SASO) (2018), IEEE, pp. 50–59

    work page 2018

  19. [19]

    F INLEY , J. H. Homer’s Odyssey. Harvard University Press Cambridge, 1978

    work page 1978

  20. [20]

    F LAXMAN , S., G OEL , S., AND RAO, J. M. Filter bubbles, echo chambers, and online news consumption. Public Opinion Quarterly 80, S1 (2016), 298–320

    work page 2016

  21. [21]

    Of other spaces

    F OUCAULT , M., AND MISKOWIEC , J. Of other spaces. diacritics 16, 1 (1986), 22–27

    work page 1986

  22. [22]

    G OLDIE , M. B. An early english rutter: The sea and spatial hermeneutics in the fourteenth and fifteenth centuries. Speculum 90, 3 (2015), 701–727

    work page 2015

  23. [23]

    Dungeons and dragons, vol

    G YGAX , G., AND ARNESON , D. Dungeons and dragons, vol. 19. Tactical Studies Rules Lake Geneva, WI, 1974. 20

    work page 1974

  24. [24]

    Second person: Role-playing and story in games and playable media

    H ARRIGAN , P., AND WARDRIP -FRUIN , N. Second person: Role-playing and story in games and playable media. The MIT Press, 2010

    work page 2010

  25. [25]

    Opinion dynamics and bounded confidence models, analysis, and simulation

    H EGSELMANN , R., K RAUSE , U., ET AL . Opinion dynamics and bounded confidence models, analysis, and simulation. Journal of artificial societies and social simulation 5, 3 (2002)

    work page 2002

  26. [26]

    Dungeons & Dragons Player’s Handbook: Arcane, Divine, and Martial Heroes: Roleplaying Game Core Rules

    H EINSOO , R., C OLLINS , A., AND WYATT, J. Dungeons & Dragons Player’s Handbook: Arcane, Divine, and Martial Heroes: Roleplaying Game Core Rules . Wizards of the Coast, 2008

    work page 2008

  27. [27]

    A mechanism for reasoning about time and belief

    I SOZAKI , H., AND SHOHAM , Y. A mechanism for reasoning about time and belief. In Fifth Generation Computer Systems (1992), pp. 694–701

    work page 1992

  28. [28]

    The sovereign map: Theoretical approaches in cartography throughout history

    J ACOB , C., ET AL . The sovereign map: Theoretical approaches in cartography throughout history. University of Chicago Press, 2006

    work page 2006

  29. [29]

    Semi-Supervised Classification with Graph Convolutional Networks

    K IPF, T. N., AND WELLING , M. Semi-supervised classification with graph convolutional networks. Computing Research Repository (CoRR) abs/1609.02907 (2016)

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  30. [30]

    Science and sanity: An introduction to non-Aristotelian systems and general semantics

    K ORZYBSKI , A. Science and sanity: An introduction to non-Aristotelian systems and general semantics. Institute of GS, 1958

    work page 1958

  31. [31]

    Web mining research: A survey

    K OSALA , R., AND BLOCKEEL , H. Web mining research: A survey. SIGKDD Explor. Newsl. 2, 1 (June 2000), 1–15

    work page 2000

  32. [32]

    Wordrank: A method for ranking web pages based on content similarity

    K RITIKOPOULOS , A., S IDERI , M., AND VARLAMIS , I. Wordrank: A method for ranking web pages based on content similarity. In 24th British National Conference on Databases (BNCOD’07) (2007), IEEE, pp. 92–100

    work page 2007

  33. [33]

    The crowd: A study of the popular mind

    L E BON, G. The crowd: A study of the popular mind. Fischer, 1897

    work page

  34. [34]

    The clockwork muse: The predictability of artistic change

    M ARTINDALE , C. The clockwork muse: The predictability of artistic change . Basic Books, 1992

    work page 1992

  35. [35]

    Conflict and consensus: A general theory of collective decisions

    M OSCOVICI , S., AND DOISE , W. Conflict and consensus: A general theory of collective decisions. Sage, 1994

    work page 1994

  36. [36]

    Flocking for multi-agent dynamic systems: Algorithms and theory

    O LFATI-SABER , R. Flocking for multi-agent dynamic systems: Algorithms and theory. IEEE Transactions on automatic control 51, 3 (2006), 401–420

    work page 2006

  37. [37]

    A., AND MURRAY, R

    O LFATI-SABER , R., F AX, J. A., AND MURRAY, R. M. Consensus and cooperation in net- worked multi-agent systems. Proceedings of the IEEE 95, 1 (2007), 215–233

    work page 2007

  38. [38]

    M., AND WHEATLEY , T

    P ARKINSON , C., K LEINBAUM , A. M., AND WHEATLEY , T. Similar neural responses predict friendship. Nature communications 9, 1 (2018), 332

    work page 2018

  39. [39]

    Semi-supervised sequence tagging with bidirectional language models

    P ETERS , M. E., A MMAR , W., B HAGAVATULA , C., AND POWER , R. Semi-supervised se- quence tagging with bidirectional language models. arXiv preprint arXiv:1705.00108 (2017)

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  40. [40]

    Information foraging

    P IROLLI , P., AND CARD , S. Information foraging. Psychological review 106, 4 (1999), 643

    work page 1999

  41. [41]

    R EYNOLDS , C. W. Flocks, herds and schools: A distributed behavioral model. In Proceed- ings of the 14th Annual Conference on Computer Graphics and Interactive Techniques(1987), SIGGRAPH ’87, ACM, pp. 25–34. 21

    work page 1987

  42. [42]

    Mapping the imaginary-maps in fantasy role-playing games

    R ¨OHL , T., AND HERBRIK , R. Mapping the imaginary-maps in fantasy role-playing games. In Forum Qualitative Sozialforschung(2008), vol. 9

    work page 2008

  43. [43]

    J., AND WATTS, D

    S ALGANIK , M. J., AND WATTS, D. J. Leading the herd astray: An experimental study of self-fulfilling prophecies in an artificial cultural market. Social psychology quarterly 71 , 4 (2008), 338–355

    work page 2008

  44. [44]

    Quotes reveal community structure and interaction dynamics

    S AMORY, M., C APPELLERI , V.-M., AND PESERICO , E. Quotes reveal community structure and interaction dynamics. In Proceedings of the 2017 ACM Conference on Computer Sup- ported Cooperative Work and Social Computing (New York, NY , USA, 2017), CSCW ’17, ACM, pp. 322–335

    work page 2017

  45. [45]

    S IMON , H. A. The sciences of the artificial. MIT press, 1968

    work page 1968

  46. [46]

    T HOMSON , J. J. Killing, letting die, and the trolley problem. The Monist 59, 2 (1976), 204– 217

    work page 1976

  47. [47]

    Learning to Speak and Act in a Fantasy Text Adventure Game

    U RBANEK , J., F AN, A., K ARAMCHETI , S., J AIN , S., H UMEAU , S., D INAN , E., ROCKT ¨ASCHEL , T., K IELA , D., S ZLAM , A., AND WESTON , J. Learning to speak and act in a fantasy text adventure game. arXiv preprint arXiv:1903.03094 (2019)

    work page internal anchor Pith review Pith/arXiv arXiv 1903

  48. [48]

    Dynamic word embeddings for evolving semantic discovery

    Y AO, Z., S UN, Y., D ING , W., R AO, N., AND XIONG , H. Dynamic word embeddings for evolving semantic discovery. In Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining(2018), ACM, pp. 673–681

    work page 2018

  49. [49]

    J., AND HASSON , U

    Y ESHURUN , Y., S WANSON , S., S IMONY , E., C HEN , J., L AZARIDI , C., H ONEY, C. J., AND HASSON , U. Same story, different story: the neural representation of interpretive frameworks. Psychological science 28, 3 (2017), 307–319. 22

    work page 2017