arxiv: 2604.16348 · v1 · submitted 2026-03-16 · 💻 cs.HC · cs.CY

Recognition: no theorem link

Beyond the Townhall: Spatial Anchoring and LLM Agents for Scalable Participatory Urban Planning

Carina I Hausladen , Javier Argota S\'anchez-Vaquerizo , Michael Siebenmann , Arthur Capozzi , Sachit Mahajan , Dirk Helbing

Authors on Pith no claims yet

Pith reviewed 2026-05-15 10:15 UTC · model grok-4.3

classification 💻 cs.HC cs.CY

keywords participatory urban planningdigital twinspatial anchoringLLM agentssustainability transitionsinformation recallpublic consultationimmersive interfaces

0 comments

The pith

Embedding sustainability projects in navigable digital twins with spatial anchoring and LLM assistants improves citizen recall and produces more constructive planning feedback.

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

The paper introduces a digital platform that places urban sustainability projects inside a 3D city model where citizens navigate while hearing location-tied audio explanations. This spatial anchoring approach, paired with two LLM agents that supply facts and prompt reflection, was tested against standard static images and text in a randomized online trial with 195 participants. The immersive setup led to stronger retention of project details and redirected attention from personal inconveniences toward shared community benefits. As a result, the feedback submitted to the simulated municipality became more solution-oriented and less complaint-driven. The work positions the system as a scalable alternative to physical town halls for inclusive sustainability planning.

Core claim

Spatially anchored immersive presentation in a digital twin significantly improved information recall, which substantially shifted participants' attention from individual inconveniences to collective, community-oriented sustainability benefits, leading to significantly more constructive, solution-focused feedback compared with conventional static visualizations and text-based consultations.

What carries the argument

A navigable digital twin that uses spatial anchoring and the method of loci to tie project details to specific locations, augmented by one LLM agent for source-grounded clarifications and another for reflective discussion.

If this is right

Cities gain a tool that can reach residents unable to attend physical meetings while still collecting higher-quality input on sustainability projects.
Participant comments move from localized complaints toward proposals that weigh community-wide benefits.
The platform lowers the expertise barrier for understanding technically complex interventions without requiring in-person facilitation.
Feedback volume and focus become more consistent across participants than in unguided text surveys.

Where Pith is reading between the lines

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

If recall gains persist in real settings, planners could reduce the time spent explaining basics and focus meetings on trade-off discussions.
Integrating the system with live sensor data from the city could let residents see current conditions alongside proposed changes.
The approach might generalize to other domains that require public input on spatial decisions, such as infrastructure siting or environmental remediation.

Load-bearing premise

Improvements seen in a short online experiment with a simulated municipality and no real stakes will hold when participants have actual skin in the game and the platform is used by diverse publics in live city processes.

What would settle it

A field study during an active municipal planning process in which real residents use the platform and their recall accuracy plus feedback constructiveness are directly compared against a parallel traditional consultation on the same project.

Figures

Figures reproduced from arXiv: 2604.16348 by Arthur Capozzi, Carina I Hausladen, Dirk Helbing, Javier Argota S\'anchez-Vaquerizo, Michael Siebenmann, Sachit Mahajan.

**Figure 1.** Figure 1: Overview of the behavioral experiment. The study uses a photorealistic digital twin of the city center of Lausanne in which an urban sustainability project is implemented. Participants (N=195) were randomly assigned to treatment or control groups and received identical content across six information blocks. The treatment group experienced immersive firstperson 360-degree video walkthroughs with spatial au… view at source ↗

**Figure 2.** Figure 2: Memory Task. Comparison of tag frequencies between the control and treatment groups, for each information block (panels 1-6), and aggregated high-level characteristics (rightmost panel). Each subplot shows those tags that were mentioned more than ten times in either group and where the difference is larger than 5 counts. Labels in bold font show significant differences according to a Fisher’s exact test (w… view at source ↗

**Figure 3.** Figure 3: Qualitative analysis of participant responses across three interaction points. Left: Conversations with Flo, the fact-based LLM agent providing verified information. Center: Conversations with Gustavo, the deliberative LLM agent facilitating opinion exploration and perspectivetaking. Right: Open-ended feedback submitted to the city after completing both LLM interactions and voting. Codes shown in bold in… view at source ↗

**Figure 4.** Figure 4: Perceptions of LLMs. Left: Quantitative ratings of helpfulness for other citizens, personal trust, and usefulness of each agent (Flo and Gustavo). Center: Thematic analysis of open-ended feedback on whether agents were helpful or unduly influential. Right: Responses regarding whether fact-based and deliberative agent roles should be separated or combined. Dots represent averages across treatment and cont… view at source ↗

read the original abstract

Participatory urban planning is central to sustainable city-making, yet the technically demanding nature of such interventions often limits meaningful involvement by diverse publics. We introduce a scalable digital participation platform that embeds sustainability projects within a navigable digital twin. Citizens experience a guided virtual walkthrough with audio narration employing the method of loci and spatial anchoring to support mnemonic encoding and recall. This immersive interface is augmented by two purpose-built LLM assistants: one delivers source-grounded factual clarifications, while the other facilitates reflective discussion. We evaluated this system in a randomized controlled online experiment (N = 195) against conventional industry practices (static visualizations and text-based consultations). Results show that spatially anchored immersive presentation significantly improved information recall, which substantially shifted participants' attention from individual inconveniences to collective, community-oriented sustainability benefits. Consequently, participants provided significantly more constructive, solution-focused feedback to the (simulated) municipality. These findings establish a practical tool for cities and policymakers to foster inclusive, democratic participation in sustainability transitions.

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.

Desk Editor's Note private letter to a colleague

The RCT finds that spatial anchoring in a digital twin plus two LLM agents boosts recall and shifts feedback toward collective solutions, but the simulated low-stakes setup leaves real applicability open.

read the letter

The paper's main result is that participants using a navigable digital twin with method-of-loci narration and two specialized LLM agents recalled more details about sustainability projects and gave more constructive, community-oriented comments than those seeing static images or text. The randomized online test with 195 people shows this shift in attention from personal inconveniences to broader benefits. That is the concrete claim they test against standard industry practices. The integration itself is new enough in this domain: putting spatial anchoring inside a walkable 3D model and splitting the LLM work into one factual-grounding agent and one reflective-dialogue agent has not been reported in the cited prior work. The controlled comparison to static baselines is a reasonable way to measure the added value. The soft spots are straightforward. The abstract gives no effect sizes, exact statistical tests, exclusion rules, or full outcome measures, so the size and reliability of the gains cannot be judged yet. More critically, the whole study runs in a hypothetical online setting with a simulated municipality and no real consequences for participants. It is unclear whether the same recall and feedback changes would appear when people have actual stakes, ongoing conflicts, or implementation power. This is a common limit in civic-tech experiments, but it directly affects how far the results can be taken for scalable planning tools. The work is aimed at HCI and urban informatics researchers who build digital participation systems. It has an implemented prototype and an empirical test, so it deserves peer review rather than a desk reject. Referees will need to see the full methods and will likely press on generalizability to live municipal processes.

Referee Report

2 major / 2 minor

Summary. The paper introduces a scalable digital participation platform for urban planning that embeds sustainability projects in a navigable digital twin. Citizens engage via guided virtual walkthroughs with audio narration using the method of loci and spatial anchoring for mnemonic support, augmented by two LLM agents (one for source-grounded factual clarifications, one for reflective discussion). In a randomized controlled online experiment (N=195) against static visualizations and text-based consultations, the authors claim that spatially anchored immersive presentation significantly improved information recall, shifted attention from individual inconveniences to collective community-oriented sustainability benefits, and elicited significantly more constructive, solution-focused feedback to the simulated municipality.

Significance. If the experimental results prove robust, the work supplies a practical tool for cities to foster more inclusive democratic participation in sustainability transitions by leveraging immersive interfaces and LLM support to enhance recall and deliberative quality. The randomized controlled design with N=195 participants offers a concrete empirical test of the proposed mechanisms, distinguishing the contribution from purely conceptual proposals in participatory planning.

major comments (2)

[Abstract and Results] Abstract and Results: The abstract asserts that the immersive system 'significantly improved information recall' and that participants 'provided significantly more constructive, solution-focused feedback', yet supplies no statistical tests, p-values, effect sizes, confidence intervals, exact measures, or exclusion criteria. This information is load-bearing for the central empirical claim and must be reported in full to allow evaluation of the evidence strength.
[Discussion] Discussion: The experiment relies on a simulated municipality with no actual implementation consequences or personal costs. The manuscript should include a dedicated limitations analysis testing whether the reported attention shift and feedback-quality gains generalize to live municipal processes where participants have real stakes, as the current low-stakes framing leaves the inference to scalable participatory planning under-supported.

minor comments (2)

[Methods] Methods: Provide additional detail on the precise implementation of the control conditions (static visualizations and text consultations) and the exact prompts or interfaces used for the LLM agents to support replication.
[Figures] Figures: Ensure that any screenshots or interface diagrams clearly label the spatial anchoring elements and LLM interaction panels so readers can directly map them to the described features.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and will revise the manuscript to strengthen the presentation of our empirical results and limitations.

read point-by-point responses

Referee: [Abstract and Results] Abstract and Results: The abstract asserts that the immersive system 'significantly improved information recall' and that participants 'provided significantly more constructive, solution-focused feedback', yet supplies no statistical tests, p-values, effect sizes, confidence intervals, exact measures, or exclusion criteria. This information is load-bearing for the central empirical claim and must be reported in full to allow evaluation of the evidence strength.

Authors: We appreciate the referee drawing attention to the need for transparent reporting of statistical evidence. The Results section (Section 4) already contains the full statistical analyses, including independent-samples t-tests for recall accuracy (with exact p-values, Cohen's d effect sizes, and 95% confidence intervals), chi-square tests for feedback categorization, and participant exclusion criteria (detailed in Section 3.3, e.g., incomplete responses and attention-check failures). To make the abstract self-contained and directly address the concern, we will revise the abstract to include summary statistics such as the key p-values and effect sizes supporting the reported improvements. This revision will ensure the central claims are fully supported without requiring readers to consult the main text. revision: yes
Referee: [Discussion] Discussion: The experiment relies on a simulated municipality with no actual implementation consequences or personal costs. The manuscript should include a dedicated limitations analysis testing whether the reported attention shift and feedback-quality gains generalize to live municipal processes where participants have real stakes, as the current low-stakes framing leaves the inference to scalable participatory planning under-supported.

Authors: We agree that the simulated, low-stakes setting represents an important boundary condition for generalizability. The current Discussion section briefly notes the online experimental context but does not contain a dedicated analysis of this issue. We will add a new subsection (Section 5.3, 'Limitations: Simulation vs. Real-World Stakes') that explicitly discusses how the absence of personal costs or implementation consequences may affect attention shifts and feedback quality, acknowledges potential differences in participant behavior under real municipal stakes, and outlines concrete directions for future field studies with actual city partners to test robustness in high-stakes environments. revision: yes

Circularity Check

0 steps flagged

No significant circularity: empirical RCT report with no derivation chain

full rationale

The paper reports results from a randomized controlled online experiment (N=195) comparing an immersive spatially anchored platform with LLM agents against static visualizations and text consultations. No equations, parameter fittings, or mathematical derivations appear in the manuscript; claims rest on direct empirical measurements of information recall and feedback quality in a simulated low-stakes setting. There are no self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citations that reduce outcomes to inputs by construction. The work is a standard empirical evaluation whose central findings are not forced by any internal definitional loop.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The claim rests on the domain assumption that spatial anchoring reliably aids recall in digital environments and that LLM outputs can be kept source-grounded without introducing new factual errors; no free parameters or invented entities are introduced.

axioms (2)

domain assumption Spatial anchoring via the method of loci improves mnemonic encoding and recall in virtual environments
Invoked to justify the guided virtual walkthrough design.
domain assumption LLM assistants can deliver source-grounded clarifications and facilitate reflective discussion without hallucination in this domain
Stated as the purpose of the two purpose-built assistants.

pith-pipeline@v0.9.0 · 5496 in / 1333 out tokens · 49336 ms · 2026-05-15T10:15:03.716893+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

91 extracted references · 91 canonical work pages

[1]

Polarization and partisan selective exposure.Journal of Communication, 60 (3):556–576, 2010

Natalie Jomini Stroud. Polarization and partisan selective exposure.Journal of Communication, 60 (3):556–576, 2010. doi: 10.1111/j.1460-2466.2010 .01497.x

work page doi:10.1111/j.1460-2466.2010 2010
[2]

Under- standing self-selection bias in online reviews.Infor- mation Systems Research, 31(4):1493–1510, 2020

Dezhi Yin, Samuel Bond, and Han Zhang. Under- standing self-selection bias in online reviews.Infor- mation Systems Research, 31(4):1493–1510, 2020

work page 2020
[3]

The case for pub- lic participation in sustainability transitions

European Environment Agency. The case for pub- lic participation in sustainability transitions. Brief- ing, European Environment Agency, 2023. URL https://www.eea.europa.eu

work page 2023
[4]

Citizen partic- ipation for sustainable cities

Stockholm Environment Institute. Citizen partic- ipation for sustainable cities. Factsheet, Stock- holm Environment Institute, 2021. URLhttps: //www.sei.org

work page 2021
[5]

Seven transformations for more equitable and sustainable cities

World Resources Institute. Seven transformations for more equitable and sustainable cities. Technical report, World Resources Institute, 2023. URLht tps://www.wri.org

work page 2023
[6]

University of Chicago Press, Chicago, IL, 1983

Jane Mansbridge.Beyond Adversary Democracy. University of Chicago Press, Chicago, IL, 1983

work page 1983
[7]

A systemic approach to delibera- tive democracy.Deliberative Systems, pages 1–26,

Jane Mansbridge, James Bohman, Simone Cham- bers, David Estlund, Andreas Follesdal, Ar- chon Fung, Cristina Lafont, Bernard Manin, and José Luis Martí. A systemic approach to delibera- tive democracy.Deliberative Systems, pages 1–26,

work page
[8]

doi: 10.1017/cbo9781139178914.002

work page doi:10.1017/cbo9781139178914.002
[9]

Dryzek.Deliberative Democracy and Be- yond: Liberals, Critics, Contestations

John S. Dryzek.Deliberative Democracy and Be- yond: Liberals, Critics, Contestations. Oxford Uni- versity Press, Oxford, UK, 2000

work page 2000
[10]

Fishkin.Democracy When the People Are Thinking: Revitalizing Our Politics Through Public Deliberation

James S. Fishkin.Democracy When the People Are Thinking: Revitalizing Our Politics Through Public Deliberation. Oxford University Press, Oxford, UK, 2018

work page 2018
[11]

Offshorewind farm siting with extended reality: Effects of immer- sive visualizations on public acceptance.Energy Research & Social Science, 80:102240, 2021

ThomasSchulz, YannickJobin, etal. Offshorewind farm siting with extended reality: Effects of immer- sive visualizations on public acceptance.Energy Research & Social Science, 80:102240, 2021

work page 2021
[12]

Xr choice experiments: Extended reality as a tool for 9 studying public preferences in energy and land- scape planning.Renewable and Sustainable Energy Reviews, 168:112783, 2022

Goda Perlaviciute, Christina Demski, et al. Xr choice experiments: Extended reality as a tool for 9 studying public preferences in energy and land- scape planning.Renewable and Sustainable Energy Reviews, 168:112783, 2022

work page 2022
[13]

Immersion and affect in participatory urban planning: Com- paring analog and immersive digital formats

Quentin Stevens and Paul Alkhoven. Immersion and affect in participatory urban planning: Com- paring analog and immersive digital formats. In Proceedings of the 2020 CHI Conference on Hu- man Factors in Computing Systems, pages 1–12, New York, NY, 2020. ACM

work page 2020
[14]

Haus- laden, Sachit Mahajan, Marc Matter, Michael Siebenmann, Michael A

Javier Argota Sánchez-Vaquerizo, Carina I. Haus- laden, Sachit Mahajan, Marc Matter, Michael Siebenmann, Michael A. B. van Eggermond, and Dirk Helbing. A virtual reality experiment to study pedestrian perception of future street sce- narios.Scientific Reports, 14(1):4571, 2 2024. doi: 10.1038/s41598-024-55073-x

work page doi:10.1038/s41598-024-55073-x 2024
[15]

2d, video, or interactive 3d? exploring how visualization for- mat affects stated preferences in spatial choice ex- periments.Journal of Environmental Psychology, 76:101643, 2021

Martin Latocha and Sebastian Schlögl. 2d, video, or interactive 3d? exploring how visualization for- mat affects stated preferences in spatial choice ex- periments.Journal of Environmental Psychology, 76:101643, 2021

work page 2021
[16]

Par- ticipatory 3d visualization for urban planning: A systematic review.Landscape and Urban Planning, 214:104120, 2021

Oliver Gruebner, Martin Sykora, and et al. Par- ticipatory 3d visualization for urban planning: A systematic review.Landscape and Urban Planning, 214:104120, 2021

work page 2021
[17]

Urbanvr: Using virtual reality to improve public understanding of urban development proposals.Cities, 112:103140, 2021

SimonNewellandThomasFischer. Urbanvr: Using virtual reality to improve public understanding of urban development proposals.Cities, 112:103140, 2021

work page 2021
[18]

Virtual reality and urban planning: A review.Sustainable Cities and Society, 35:621–630, 2017

Ehsan Jamei, Michael Mortimer, Mehdi Seyedmah- moudian, Ben Horan, and Alex Stojcevski. Virtual reality and urban planning: A review.Sustainable Cities and Society, 35:621–630, 2017

work page 2017
[19]

Board games vs

Claire Engelmann and Bastian Köhler. Board games vs. immersive games: How visualization me- dia shape deliberation in participatory planning. Environment and Planning B: Urban Analytics and City Science, 47(8):1462–1478, 2020

work page 2020
[20]

Ian D. Bishop. From static images to immersive environments: Landscape visualization and its ef- fects on perception and preference.Landscape and Urban Planning, 189(2-3):72–83, 2019

work page 2019
[21]

Hardy, Thomas L

Rachit Dubey, Mathew D. Hardy, Thomas L. Grif- fiths, and Rahul Bhui. Ai-generated visuals of car- free us cities help improve support for sustainable policies.Nature Sustainability, 7:399–403, 2024. doi: 10.1038/s41893-024-01299-6

work page doi:10.1038/s41893-024-01299-6 2024
[22]

Immersive virtual reality and the il- lusion of presence in experimental and clinical set- tings.Annual Review of Clinical Psychology, 14(2): 149–169, 2018

Mel Slater. Immersive virtual reality and the il- lusion of presence in experimental and clinical set- tings.Annual Review of Clinical Psychology, 14(2): 149–169, 2018. doi: 10.1146/annurev-clinpsy-050 817-084942

work page doi:10.1146/annurev-clinpsy-050 2018
[23]

Can immersive virtual reality increase pro- environmental behavior?Virtual Reality, 24:567– 581, 2020

Brad Aisbett, Katie Wilson, and David Rowe. Can immersive virtual reality increase pro- environmental behavior?Virtual Reality, 24:567– 581, 2020. doi: 10.1007/s10055-019-00407-y

work page doi:10.1007/s10055-019-00407-y 2020
[24]

Virtual reality as a tool for environmental behavior change: A systematic review.Frontiers in Psychol- ogy, 13:804254, 2022

Jason Bailey, Jeremy Bailenson, and Andrea Won. Virtual reality as a tool for environmental behavior change: A systematic review.Frontiers in Psychol- ogy, 13:804254, 2022. doi: 10.3389/fpsyg.2022.804 254

work page doi:10.3389/fpsyg.2022.804 2022
[25]

Parsons and Albert A

Thomas D. Parsons and Albert A. Rizzo. Cognitive and neural perspectives on spatial anchoring and reference frames in virtual environments.Frontiers in Psychology, 11:783, 2020. doi: 10.3389/fpsyg.20 20.00783

work page doi:10.3389/fpsyg.20 2020
[26]

Deliberation among informed citizens: The value of exploring alternative thinking frames

Ariane Lambert-Mogiliansky and Irénée Frérot. Deliberation among informed citizens: The value of exploring alternative thinking frames. arXiv preprint, 2024. arXiv:2410.05862

work page arXiv 2024
[27]

Oxford University Press, New York, NY, 1990

AllanPaivio.Mental Representations: A Dual Cod- ing Approach. Oxford University Press, New York, NY, 1990

work page 1990
[28]

Levels of processing: A framework for memory research

Fergus IM Craik and Robert S Lockhart. Levels of processing: A framework for memory research. Journal of verbal learning and verbal behavior, 11 (6):671–684, 1972

work page 1972
[29]

En- coding context and the retrieval of events: The role of cognitive operations.Memory & Cognition, 24 (4):511–521, 1996

Kathleen B McDermott and Jason M Watson. En- coding context and the retrieval of events: The role of cognitive operations.Memory & Cognition, 24 (4):511–521, 1996

work page 1996
[30]

Theuseofthemethodof loci in a realistic memory task: Examining memory for people encountered during a simulated social interaction.Advances in Cognitive Psychology, 8 (1):92, 2012

Emily L G Legge, Christopher R Madan, Edith T Ng, andJeremyBCaplan. Theuseofthemethodof loci in a realistic memory task: Examining memory for people encountered during a simulated social interaction.Advances in Cognitive Psychology, 8 (1):92, 2012

work page 2012
[31]

Walking through doorways causes forgetting: Sit- uation models and experienced space.Memory & cognition, 34(5):1150–1156, 2006

Gabriel A Radvansky and David E Copeland. Walking through doorways causes forgetting: Sit- uation models and experienced space.Memory & cognition, 34(5):1150–1156, 2006. doi: 10.3758/bf 03193261

work page doi:10.3758/bf 2006
[32]

Event perception: A mind-brain perspective.Psychologi- cal Bulletin, 133(2):273, 2007

Jeffrey M Zacks, Nicole K Speer, Khena M Swal- low, Todd S Braver, and Jeremy R Reynolds. Event perception: A mind-brain perspective.Psychologi- cal Bulletin, 133(2):273, 2007

work page 2007
[33]

Pettijohn and Gabriel A

Kyle A. Pettijohn and Gabriel A. Radvansky. Nar- rative event boundaries, reading times, and ex- pectation.Memory & Cognition, 44(7):1064–1075,

work page
[34]

doi: 10.3758/s13421-016-0619-6

work page doi:10.3758/s13421-016-0619-6
[35]

Brigham Moll and Ed Sykes. Optimized virtual reality-based method of loci memorization tech- niques through increased immersion and effective memory palace designs: a feasibility study.Virtual Reality, 27(2):941–966, 2023. doi: 10.1007/s10055 -022-00700-z

work page doi:10.1007/s10055 2023
[36]

The method of loci in virtual reality: explicit binding of objects to spatial con- texts enhances subsequent memory recall.Journal of Cognitive Enhancement, 4(1):12–30, 2020

Nicco Reggente, Joey KY Essoe, Hera Younji Baek, and Jesse Rissman. The method of loci in virtual reality: explicit binding of objects to spatial con- texts enhances subsequent memory recall.Journal of Cognitive Enhancement, 4(1):12–30, 2020. doi: 10.1007/s41465-019-00141-8. 10

work page doi:10.1007/s41465-019-00141-8 2020
[37]

Doorways do not always cause forgetting: Studying the effect of lo- comotion technique and doorway visualization in virtual reality

Thomas Van Gemert, Sean Chew, Yiannis Kalait- zoglou, and Joanna Bergström. Doorways do not always cause forgetting: Studying the effect of lo- comotion technique and doorway visualization in virtual reality. InProceedings of the 2024 CHI Conference on Human Factors in Computing Sys- tems, pages 1–13, New York, NY, 2024. ACM. doi: 10.1145/3613904.3642879

work page doi:10.1145/3613904.3642879 2024
[38]

Digital rights to the city: localpracticesandnegotiationsofurbanspace on decidim.Urban Planning, 9, 2024

Aline Suter, Lars Kaiser, Martin Dušek, Florin Hasler, and Simone Tappert. Digital rights to the city: localpracticesandnegotiationsofurbanspace on decidim.Urban Planning, 9, 2024

work page 2024
[39]

A systematic review of e-participation platforms: Features, inclusivity, and democratic impacts

Jiyoon Shin, Thiago Oliveira, and Marina Micheli. A systematic review of e-participation platforms: Features, inclusivity, and democratic impacts. Government Information Quarterly, 41(1):101761, 2024

work page 2024
[40]

E-participation platforms and democratic innovation: A comparative review.Pol- icy & Internet, 16(2):215–236, 2024

Laura Mariani. E-participation platforms and democratic innovation: A comparative review.Pol- icy & Internet, 16(2):215–236, 2024

work page 2024
[41]

Informative architectures and citizen participation: A compar- ative study of decidim and rousseau.International Journal of Communication, 16(2):2200–2221, 2022

Eduardo Magro and Carlos Calderón. Informative architectures and citizen participation: A compar- ative study of decidim and rousseau.International Journal of Communication, 16(2):2200–2221, 2022

work page 2022
[42]

The criti- cal role of retrieval practice in long-term retention

Henry L Roediger and Andrew C Butler. The criti- cal role of retrieval practice in long-term retention. Trends in cognitive sciences, 15(1):20–27, 2011

work page 2011
[43]

ChatGPT sets record for fastest-growing user base.Reuters, February 2023

Reuters. ChatGPT sets record for fastest-growing user base.Reuters, February 2023

work page 2023
[44]

Submitting surveys via a conversational interface: An evalua- tion of user acceptance and approach effectiveness

Irene Celino and Gloria Re Calegari. Submitting surveys via a conversational interface: An evalua- tion of user acceptance and approach effectiveness. International Journal of Human-Computer Stud- ies, 139:102415, 2020. doi: 10.1016/j.ijhcs.2020 .102415

work page doi:10.1016/j.ijhcs.2020 2020
[45]

Public trust in arti- ficial intelligence: A gallup-bentley university sur- vey

Gallup and Bentley University. Public trust in arti- ficial intelligence: A gallup-bentley university sur- vey. Survey Report, 2023. Survey of American adults on AI trust and adoption. Reported in PayIt (2025) and various government technology sources showing 77% distrust of government AI use

work page 2023
[46]

Patel, Raj K

Megh R. Patel, Raj K. Sharma, and Wei Chen. Im- proving citizen-government interactions with gen- erative artificial intelligence.PLOS ONE, 19(12): e0315315, 2024. doi: 10.1371/journal.pone.03153

work page doi:10.1371/journal.pone.03153 2024
[47]

Available at PMC11651614

work page
[48]

Ai for the people: Use cases for gov- ernment

Mark Fagan. Ai for the people: Use cases for gov- ernment. Technical Report M-RCBG Working Pa- per 2024-02, Harvard Kennedy School, Mossavar- Rahmani Center for Business and Government, Cambridge, MA, 2024. URLhttps://hks.harvar d.edu/sites/default/files/centers/mrcbg/wo rking.papers/M-RCBG_Working_Paper_2024-0 2_AI_for_the_People.pdf

work page 2024
[49]

Governing with artificial intelligence

OECD. Governing with artificial intelligence. Tech- nical report, OECD Publishing, Paris, 2025. URL https://www.oecd.org/en/publications/gover ning-with-artificial-intelligence_398fa2 87-en.html

work page 2025
[50]

A survey of hallucinations in large language models: Sources, taxonomy, and mitigation.ACM Computing Surveys, 2025

Nouha Dziri et al. A survey of hallucinations in large language models: Sources, taxonomy, and mitigation.ACM Computing Surveys, 2025

work page 2025
[51]

Beacon Press, Boston, 1984

Jürgen Habermas.The Theory of Communicative Action, Volume 1: Reason and the Rationalization of Society. Beacon Press, Boston, 1984. Originally published in German, 1981

work page 1984
[52]

Pantheon Books, New York, NY, 1980

Michel Foucault.Power/Knowledge: Selected In- terviews and Other Writings, 1972–1977. Pantheon Books, New York, NY, 1980. Edited by Colin Gor- don

work page 1972
[53]

Ian Walker, Alan Tapp, and Adrian Davis. In- ternational Journal of Environment and Health Motonormativity: how social norms hide a major public health hazard Motonormativity: how social norms hide a major public health hazard.Int. J. Environment and Health, 11(1):2023, 2023. ISSN 1743-4955. doi: 10.1504/IJENVH.2023.10060994. URLhttps://www.inderscience.c...

work page doi:10.1504/ijenvh.2023.10060994 2023
[54]

Rethinking the public sphere: A contribution to the critique of actually existing democracy.Social Text, 25/26:56–80, 1990

Nancy Fraser. Rethinking the public sphere: A contribution to the critique of actually existing democracy.Social Text, 25/26:56–80, 1990. doi: 10.4324/9781351202558-6

work page doi:10.4324/9781351202558-6 1990
[55]

D. Kelly. Habermasian discourse theory for edu- cational policy.Policy & Education, 30(1):56–67, 2023

work page 2023
[56]

Deliberative democracy without deliberation.Journal of Deliberative Democracy,

Marit Hammond. Deliberative democracy without deliberation.Journal of Deliberative Democracy,

work page
[57]

doi: 10.16997/jdd.1460

work page doi:10.16997/jdd.1460
[58]

Incentives and prosocial behavior.American Economic Review, 96 (5):1652–1678, 2006

Roland Bénabou and Jean Tirole. Incentives and prosocial behavior.American Economic Review, 96 (5):1652–1678, 2006. doi: 10.3386/w11535

work page doi:10.3386/w11535 2006
[59]

Phillips Davison

W. Phillips Davison. The third-person effect in communication.Public Opinion Quarterly, 47(1): 1–15, 1983

work page 1983
[60]

Two-component models of so- cially desirable responding.Journal of Personality and Social Psychology, 46(3):598–609, 1984

Delroy L Paulhus. Two-component models of so- cially desirable responding.Journal of Personality and Social Psychology, 46(3):598–609, 1984

work page 1984
[61]

Comparing the persuasiveness of role- playing large language models and human experts on polarized us political issues.AI & Society, 2025

Kobi Hackenburg, L Ibrahim, B M Tappin, and M Tsakiris. Comparing the persuasiveness of role- playing large language models and human experts on polarized us political issues.AI & Society, 2025. doi: 10.1007/s00146-025-02464-x

work page doi:10.1007/s00146-025-02464-x 2025
[62]

Llm-generated messages can per- suade humans on policy issues.Preprint at OSF,

H Bai, J G Voelkel, S Muldowney, J C Eichstaedt, and R Willer. Llm-generated messages can per- suade humans on policy issues.Preprint at OSF,

work page
[63]

doi: 10.31219/osf.io/stakv_v8

work page doi:10.31219/osf.io/stakv_v8
[64]

Testing theories of political persuasion using ai.Proceedings of the National Academy of Sciences, 122:e2412815122, 2025

Lisa P Argyle et al. Testing theories of political persuasion using ai.Proceedings of the National Academy of Sciences, 122:e2412815122, 2025

work page 2025
[65]

A framework to assess the persua- sion risks large language model chatbots pose to 11 democratic societies.Preprint at arXiv, 2025

Z Chen et al. A framework to assess the persua- sion risks large language model chatbots pose to 11 democratic societies.Preprint at arXiv, 2025. doi: 10.48550/ARXIV.2505.00036

work page doi:10.48550/arxiv.2505.00036 2025
[66]

Evaluating the persuasive influence of political microtargeting with large language models.Proceedings of the National Academy of Sciences, 121:e2403116121, 2024

Kobi Hackenburg and Helen Margetts. Evaluating the persuasive influence of political microtargeting with large language models.Proceedings of the National Academy of Sciences, 121:e2403116121, 2024

work page 2024
[67]

Durably reducing conspiracy beliefs through dialogues with ai.Science, 385:eadq1814, 2024

Thomas H Costello, Gordon Pennycook, and David G Rand. Durably reducing conspiracy beliefs through dialogues with ai.Science, 385:eadq1814, 2024

work page 2024
[68]

On the conversational persuasiveness of gpt-4.Na- ture Human Behaviour, 2025

F Salvi, M Horta Ribeiro, R Gallotti, and R West. On the conversational persuasiveness of gpt-4.Na- ture Human Behaviour, 2025. doi: 10.1038/s41562 -025-02194-6

work page doi:10.1038/s41562 2025
[69]

Measuring the persuasiveness of language models.Anthropic, 2024

Esin Durmus et al. Measuring the persuasiveness of language models.Anthropic, 2024. URLhttps: //www.anthropic.com/news/measuring-model -persuasiveness

work page 2024
[70]

How persua- sive is ai-generated propaganda?PNAS Nexus, 3:pgae034, 2024

Josh A Goldstein, Jason Chao, Shelby Grossman, Alex Stamos, and Michael Tomz. How persua- sive is ai-generated propaganda?PNAS Nexus, 3:pgae034, 2024

work page 2024
[71]

Robert J. Fisher. Social desirability bias and the validity of indirect questioning.Journal of Con- sumer Research, 20(2):303–315, 1993. doi: 10.108 6/209351

work page 1993
[72]

Alexander and Henry J

Cheryl S. Alexander and Henry J. Becker. The use of vignettes in survey research.Public Opinion Quarterly, 42(1):93–104, 1978. doi: 10.1086/2684 32

work page doi:10.1086/2684 1978
[73]

Herman Aguinis and Kyle J. Bradley. Vignette methodology: Modern and innovative applications in organizational research.Organizational Research Methods, 17(4):351–365, 2014

work page 2014
[74]

A note on presence terminology.Pres- ence Connect, 3(3):1–5, 2003

Mel Slater. A note on presence terminology.Pres- ence Connect, 3(3):1–5, 2003

work page 2003
[75]

Cummings and Jeremy N

James J. Cummings and Jeremy N. Bailenson. How immersive is enough? a meta-analysis of the effect of immersive technology on user presence.Media Psychology, 19(2):272–309, 2016. doi: 10.1080/15 213269.2015.1015740

work page doi:10.1080/15 2016
[76]

B. J. Fogg.Persuasive Technology: Using Comput- ers to Change What We Think and Do. Morgan Kaufmann, San Francisco, 2003

work page 2003
[77]

Place illusion and plausibility can lead to realistic behaviour in immersive virtual environments.Philosophical Transactions of the Royal Society B, 364(1535):3549–3557, 2009

Mel Slater. Place illusion and plausibility can lead to realistic behaviour in immersive virtual environments.Philosophical Transactions of the Royal Society B, 364(1535):3549–3557, 2009. doi: 10.1098/rstb.2009.0138

work page doi:10.1098/rstb.2009.0138 2009
[78]

Arnstein

Sherry R. Arnstein. A ladder of citizen participa- tion.Journal of the American Institute of Planners, 35(4):216–224, 1969

work page 1969
[79]

Ready player me, 2025

Ready Player Me. Ready player me, 2025. URL https://readyplayer.me/

work page 2025
[80]

Pixabay, 2025

Canva. Pixabay, 2025. URLhttps://pixabay.co m/

work page 2025

Showing first 80 references.