pith. sign in

arxiv: 2606.22599 · v2 · pith:P6U6VVUUnew · submitted 2026-06-21 · 💰 econ.EM

Networked risk perception and behavioral bubbles: the case of a pandemic

Sepehr Ilami , Margherita Comola , Silvia Prina , Babak Heydari This is my paper

Pith reviewed 2026-06-26 09:17 UTC · model grok-4.3

classification 💰 econ.EM
keywords risk perceptionbehavioral spilloversmobility networkCOVID-19behavioral bubblespanel dataMassachusetts municipalitiespeer effects
0
0 comments X

The pith

Pandemic behavioral responses spread along mobility networks but remain confined within defined communities.

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

The paper uses weekly mobility data from 313 Massachusetts towns and a pre-pandemic network to estimate how risk perception and behavior respond to local cases, peer-town behavior, and inertia. It separates these effects in two-way fixed-effects regressions that rely on the fixed pre-shock network and lagged peer signals. The central finding is that spillovers are large yet localize strictly inside mobility-defined communities, carry behavioral content beyond case information, and require both connection and demographic similarity. This identifies mobility communities rather than administrative units as the scale at which behavioral responses operate. The setup treats risk perception as shaped by networked observation of peers rather than individual readout of objective hazard.

Core claim

Inter-town behavioral spillovers are substantial and localize almost entirely within mobility-defined communities, with effectively no propagation across community boundaries. When network-exposure-to-cases and network-exposure-to-behavior are raced against each other, the behavioral channel survives while the case-exposure channel becomes null. A joint mobility-by-demographic decomposition shows the spillover requires both routine connection and demographic similarity, concentrates where towns are connected and similar, and vanishes between similar towns that are not connected.

What carries the argument

The pre-pandemic inter-town mobility network, used inside two-way fixed-effects panel regressions to trace behavioral spillovers from peer towns while addressing reflection and endogenous-group problems via lagged signals.

If this is right

  • Risk perception operates as a networked process rather than an individual cognitive readout of objective hazard.
  • Mobility-defined communities, not administrative units, are the operative scale of behavioral response.
  • The spillover is driven by an observational and normative channel, not purely informational transmission of case counts.
  • The pattern rules out a shared-conditions confound because it vanishes between demographically similar towns that lack mobility links.

Where Pith is reading between the lines

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

  • The same mobility-network structure could be used to study behavioral responses in other settings with uncertain evolving risks, such as financial market contagion.
  • Interventions aimed at shifting risk perception might achieve larger effects if targeted at entire mobility communities rather than isolated towns.
  • Demographic similarity alone is insufficient without mobility connection, suggesting future work could test whether adding weak ties across communities reduces bubble formation.

Load-bearing premise

The pre-pandemic mobility network fixes interaction structure before the shock and, together with a lagged peer signal, removes reflection and endogenous-group bias from the regressions.

What would settle it

Observation of substantial behavioral spillovers across mobility-community boundaries after the same controls would falsify the localization result.

read the original abstract

Risk perception is typically modeled as an individual cognitive readout of objective hazard, yet during crises what people judge as risky is shaped by what their peers do. Using weekly mobility data from 313 Massachusetts municipalities over the first year of the COVID-19 pandemic and a pre-pandemic inter-town mobility network that fixes interaction structure before the shock, we estimate two-way fixed-effects panel regressions that separate local case response, inter-town behavioral spillover along the mobility network, and within-town inertia; the pre-shock network and a lagged peer signal address the standard reflection and endogenous-group concerns. Three findings emerge. First, inter-town behavioral spillovers are substantial and localize almost entirely within mobility-defined communities, with effectively no propagation across community boundaries, the empirical referent of behavioral bubbles. Second, the within-community spillover carries behavioral content beyond peer-town case information: when network-exposure-to-cases and network-exposure-to-behavior are raced, the behavioral channel survives and the case-exposure channel goes null. Third, a joint mobility-by-demographic decomposition shows the spillover requires both routine connection and demographic similarity. It concentrates where towns are connected and similar, and vanishes between similar towns that are not connected, ruling out a shared-conditions confound and pointing to an observational and normative channel rather than a purely informational one. These results recast risk perception as a networked phenomenon and identify mobility-defined communities, rather than administrative units, as the operative scale of behavioral response. The pattern should generalize wherever exposure is uncertain, evolving, and socially negotiated, including climate adaptation and financial contagion.

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

0 major / 2 minor

Summary. The paper estimates two-way fixed-effects panel regressions on weekly mobility and case data from 313 Massachusetts municipalities during the first year of COVID-19. It uses a pre-pandemic inter-town mobility network (to fix interaction structure before the shock) plus lagged peer signals to separate local case response, inter-town behavioral spillovers along the network, and within-town inertia. The central claims are that spillovers are substantial yet localize almost entirely within mobility-defined communities (termed behavioral bubbles), that the within-community channel carries behavioral content beyond peer case information (behavioral exposure survives while case exposure goes null when both are included), and that the spillover requires both routine mobility connection and demographic similarity (vanishing between similar but unconnected towns).

Significance. If the results hold after full verification of specifications and robustness, the findings would be significant for the literature on social influences in risk perception and crisis behavior. The pre-determined network plus lagged signal provides a credible approach to reflection and endogenous-group problems in network peer effects; the horse-race between behavioral and case-exposure channels and the mobility-by-demographic decomposition are strengths that help isolate an observational/normative mechanism. The emphasis on mobility communities rather than administrative units as the operative scale could inform modeling in related domains such as financial contagion or climate adaptation.

minor comments (2)
  1. [Abstract] Abstract: the exact community-detection algorithm used to define mobility-defined communities is not stated; because localization within these communities is load-bearing for the behavioral-bubbles claim, a one-sentence description (or reference to the methods section) would improve transparency even if the full text contains it.
  2. [Abstract] Abstract: the construction of the two network-exposure variables (exposure-to-cases and exposure-to-behavior) is summarized but not formalized; adding a brief equation or variable definition would clarify how the horse-race is implemented and why the case channel is reported as null.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive and accurate summary of our manuscript, including its recognition of the pre-determined network design, the horse-race between behavioral and case-exposure channels, and the mobility-by-demographic decomposition. The recommendation for minor revision is noted; we will incorporate any editorial or presentational suggestions in the revised version.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper presents an empirical analysis using two-way fixed-effects panel regressions on predetermined pre-pandemic mobility networks and lagged peer signals. This setup follows standard econometric practice to address reflection and endogenous group concerns without any derivation that reduces to fitted parameters by construction or self-referential definitions. The central claims about behavioral spillovers are data-driven estimates, not tautological outputs. No load-bearing steps match the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

Central claim rests on exogeneity of the pre-pandemic mobility network for identification and validity of the regression decomposition separating behavioral from informational channels.

axioms (2)
  • domain assumption The pre-pandemic inter-town mobility network is exogenous to the pandemic shock and captures the relevant interaction structure without endogenous response to cases.
    Invoked to fix interaction structure before the shock and address reflection/endogenous-group concerns.
  • domain assumption The two-way fixed-effects specification with lagged peer signal isolates inter-town behavioral spillover from local case response and within-town inertia.
    Required for the claim that behavioral channel survives while case-exposure channel goes null.
invented entities (1)
  • behavioral bubbles no independent evidence
    purpose: Label for the observed localization of spillovers with no cross-boundary propagation.
    Empirical pattern identified from the data; no independent falsifiable prediction supplied.

pith-pipeline@v0.9.1-grok · 5811 in / 1481 out tokens · 35252 ms · 2026-06-26T09:17:01.481980+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

75 extracted references · 2 linked inside Pith

  1. [1]

    Kunreuther, H. et al. Risk management and climate change. Nature climate change 3, 447–450 (2013)

    2013

  2. [2]

    & Renn, O

    A ven, T. & Renn, O. Risk management and governance: Concepts, guidelines and applications Vol. 16 (Springer Science & Business Media, 2010)

    2010

  3. [3]

    & Tversky, A

    Kahneman, D. & Tversky, A. Choices, values, and frames. American psychologist 39, 341 (1984)

    1984

  4. [4]

    Perception of risk

    Slovic, P. Perception of risk. The Perception of Risk 220–231 (2016). 16

    2016

  5. [5]

    & Kahneman, D

    Tversky, A. & Kahneman, D. Judgment under uncertainty: Heuristics and biases. Science 185, 1124–1131 (1974)

    1974

  6. [6]

    Perception of risk

    Slovic, P. Perception of risk. Science 236, 280–285 (1987)

    1987

  7. [7]

    Flaxman, S. et al. Estimating the effects of non-pharmaceutical interventions on covid-19 in europe. Nature 584, 257–261 (2020)

    2020

  8. [8]

    Hsiang, S. et al. The effect of large-scale anti-contagion policies on the covid-19 pandemic. Nature 584, 262–267 (2020)

    2020

  9. [9]

    Haug, N. et al. Ranking the effectiveness of worldwide COVID-19 government interventions. Nature Human Behaviour 4, 1303–1312 (2020)

    2020

  10. [10]

    Brauner, J. M. et al. Inferring the effectiveness of government interventions against COVID-19. Science 371, eabd9338 (2021)

    2021

  11. [11]

    & Heydari, B

    Abouk, R. & Heydari, B. The immediate effect of COVID-19 policies on social-distancing behavior in the united states. Public Health Reports 136, 245–252 (2021)

    2021

  12. [12]

    Risk perception

    Slovic, P. Risk perception. Carcinogen Risk Assessment 171–181 (1988)

    1988

  13. [13]

    & Wildavsky, A

    Douglas, M. & Wildavsky, A. How can we know the risks we face? why risk selection is a social process 1 (1982)

    1982

  14. [14]

    L., Alhakami, A., Slovic, P

    Finucane, M. L., Alhakami, A., Slovic, P. & Johnson, S. M. The affect heuristic in judgments of risks and benefits. Journal of Behavioral Decision Making 13, 1–17 (2000)

    2000

  15. [15]

    F., Weber, E

    Loewenstein, G. F., Weber, E. U., Hsee, C. K. & Welch, N. Risk as feelings. Psychological Bulletin 127, 267–286 (2001)

    2001

  16. [16]

    & Mertz, C

    Flynn, J., Slovic, P. & Mertz, C. K. Gender, race, and perception of environmental health risks. Risk analysis 14, 1101–1108 (1994)

    1994

  17. [17]

    Kasperson, R. E. et al. The social amplification of risk: A conceptual framework. Risk analysis 8, 177–187 (1988)

    1988

  18. [18]

    Pidgeon, N., Kasperson, R. E. & Slovic, P. The social amplification of risk (Cambridge University Press, 2003)

    2003

  19. [19]

    Cialdini, R. B. & Goldstein, N. J. Social influence: Compliance and conformity. Annu. Rev. Psychol. 55, 591–621 (2004)

    2004

  20. [20]

    The grammar of society: The nature and dynamics of social norms (Cambridge University Press, 2005)

    Bicchieri, C. The grammar of society: The nature and dynamics of social norms (Cambridge University Press, 2005)

    2005

  21. [21]

    & Fischbacher, U

    Fehr, E. & Fischbacher, U. Social norms and human cooperation. Trends in cognitive sciences 8, 185–190 (2004)

    2004

  22. [22]

    B., Reno, R

    Cialdini, R. B., Reno, R. R. & Kallgren, C. A. A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology 58, 1015–1026 (1990). 17

    1990

  23. [23]

    Brewer, N. T. et al. Meta-analysis of the relationship between risk perception and health behavior: the example of vaccination. Health psychology 26, 136 (2007)

    2007

  24. [24]

    & Lauriola, M

    Savadori, L. & Lauriola, M. Risk perception and protective behaviors during the rise of the covid-19 outbreak in italy. Frontiers in psychology 11, 577331 (2021)

    2021

  25. [25]

    & Ramirez, F

    Lavell, A., Mansilla, E., Maskrey, A. & Ramirez, F. The social construction of the covid-19 pandemic: disaster, risk accumulation and public policy. Red de estudios sociales en Prevención de desastres en América Latina (LA RED) (2020)

    2020

  26. [26]

    Tsoy, D., Tirasawasdichai, T., Kurpayanidi, K. I. et al. Role of social media in shaping pub- lic risk perception during covid-19 pandemic: A theoretical review. International journal of management science and business administration 7, 35–41 (2021)

    2021

  27. [27]

    Bauch, C. T. & Galvani, A. P. Social factors in epidemiology. Science 342, 47–49 (2013)

    2013

  28. [28]

    Reluga, T. C. Game theory of social distancing in response to an epidemic. PLoS computational biology 6, e1000793 (2010)

    2010

  29. [29]

    & Jansen, V

    Funk, S., Salathé, M. & Jansen, V. A. Modelling the influence of human behaviour on the spread of infectious diseases: a review. Journal of the Royal Society Interface 7, 1247–1256 (2010)

    2010

  30. [30]

    Fenichel, E. P. et al. Adaptive human behavior in epidemiological models. Proceedings of the National Academy of Sciences 108, 6306–6311 (2011)

    2011

  31. [31]

    The spread of behavior in an online social network experiment

    Centola, D. The spread of behavior in an online social network experiment. science 329, 1194–1197 (2010)

    2010

  32. [32]

    How behavior spreads: The science of complex contagions Vol

    Centola, D. How behavior spreads: The science of complex contagions Vol. 3 (Princeton University Press Princeton, NJ, 2018)

    2018

  33. [33]

    & Heydari, B

    Cao, Q. & Heydari, B. Micro-level social structures and the success of COVID-19 national policies. Nature Computational Science 2, 595–604 (2022)

    2022

  34. [34]

    Bavel, J. J. V. et al. Using social and behavioural science to support covid-19 pandemic response. Nature human behaviour 4, 460–471 (2020)

    2020

  35. [35]

    How behavioural science data helps mitigate the covid-19 crisis

    Betsch, C. How behavioural science data helps mitigate the covid-19 crisis. Nature human behaviour 4, 438–438 (2020)

    2020

  36. [36]

    Kraemer, M. U. et al. The effect of human mobility and control measures on the covid-19 epidemic in china. Science 368, 493–497 (2020)

    2020

  37. [37]

    Chang, S. et al. Mobility network models of covid-19 explain inequities and inform reopening. Nature 589, 82–87 (2021)

    2021

  38. [38]

    F., Skali, A., Savage, D

    Chan, H. F., Skali, A., Savage, D. A., Stadelmann, D. & Torgler, B. Risk attitudes and human mobility during the covid-19 pandemic. Scientific reports 10, 19931 (2020)

    2020

  39. [39]

    Barbieri, D. M. et al. Impact of covid-19 pandemic on mobility in ten countries and associated perceived risk for all transport modes. PloS one 16, e0245886 (2021). 18

    2021

  40. [40]

    & Nwegbu, C

    Lawal, O. & Nwegbu, C. Movement and risk perception: evidence from spatial analysis of mobile phone-based mobility during the covid-19 lockdown, nigeria. GeoJournal 87, 1543–1558 (2022)

    2022

  41. [41]

    Galasso, V. et al. Gender differences in covid-19 attitudes and behavior: Panel evidence from eight countries. Proceedings of the National Academy of Sciences 117, 27285–27291 (2020)

    2020

  42. [42]

    & Vespignani, A

    Colizza, V., Pastor-Satorras, R. & Vespignani, A. Reaction–diffusion processes and metapopu- lation models in heterogeneous networks. Nature Physics 3, 276–282 (2007)

    2007

  43. [43]

    Balcan, D. et al. Multiscale mobility networks and the spatial spreading of infectious diseases. Proceedings of the national academy of sciences 106, 21484–21489 (2009)

    2009

  44. [44]

    & Helbing, D

    Brockmann, D. & Helbing, D. The hidden geometry of complex, network-driven contagion phenomena. science 342, 1337–1342 (2013)

    2013

  45. [45]

    Meloni, S. et al. Modeling human mobility responses to the large-scale spreading of infectious diseases. Scientific reports 1, 62 (2011)

    2011

  46. [46]

    & Vespignani, A

    Pastor-Satorras, R., Castellano, C., Van Mieghem, P. & Vespignani, A. Epidemic processes in complex networks. Reviews of modern physics 87, 925–979 (2015)

    2015

  47. [47]

    D., Guillaume, J.-L., Lambiotte, R

    Blondel, V. D., Guillaume, J.-L., Lambiotte, R. & Lefebvre, E. Fast unfolding of communities in large networks. Journal of statistical mechanics: theory and experiment 2008, P10008 (2008)

    2008

  48. [48]

    Manski, C. F. Identification of endogenous social effects: The reflection problem. The review of economic studies 60, 531–542 (1993)

    1993

  49. [49]

    & Fortin, B

    Bramoullé, Y., Djebbari, H. & Fortin, B. Identification of peer effects through social networks. Journal of econometrics 150, 41–55 (2009)

    2009

  50. [50]

    Jackson, M. O. et al. Social and economic networks Vol. 3 (Princeton university press Princeton, 2008)

    2008

  51. [51]

    & Macy, M

    Centola, D. & Macy, M. Complex contagions and the weakness of long ties. American journal of Sociology 113, 702–734 (2007)

    2007

  52. [52]

    Christakis, N. A. & Fowler, J. H. The spread of obesity in a large social network over 32 years. New England Journal of Medicine 357, 370–379 (2007)

    2007

  53. [53]

    Threshold models of collective behavior

    Granovetter, M. Threshold models of collective behavior. American Journal of Sociology 83, 1420–1443 (1978)

    1978

  54. [54]

    Watts, D. J. A simple model of global cascades on random networks. Proceedings of the National Academy of Sciences 99, 5766–5771 (2002)

    2002

  55. [55]

    Bond, R. M. et al. A 61-million-person experiment in social influence and political mobilization. Nature 489, 295–298 (2012)

    2012

  56. [56]

    & Walker, D

    Aral, S. & Walker, D. Identifying influential and susceptible members of social networks. Science 337, 337–341 (2012). 19

    2012

  57. [57]

    An experimental study of homophily in the adoption of health behavior

    Centola, D. An experimental study of homophily in the adoption of health behavior. Science 334, 1269–1272 (2011)

    2011

  58. [58]

    Peer effects with random assignment: Results for Dartmouth roommates

    Sacerdote, B. Peer effects with random assignment: Results for Dartmouth roommates. The Quarterly Journal of Economics 116, 681–704 (2001)

    2001

  59. [59]

    & Fortin, B

    Bramoullé, Y., Djebbari, H. & Fortin, B. Peer effects in networks: A survey. Annual Review of Economics 12, 603–629 (2020)

    2020

  60. [60]

    & Prina, S

    Comola, M. & Prina, S. Treatment effect accounting for network changes. Review of Economics and Statistics 103, 597–604 (2021)

    2021

  61. [61]

    & Beutels, P

    Verelst, F., Willem, L. & Beutels, P. Behavioural change models for infectious disease trans- mission: a systematic review (2010–2015). Journal of The Royal Society Interface 13, 20160820 (2016)

    2010

  62. [62]

    Bedford, J. et al. A new twenty-first century science for effective epidemic response. Nature 575, 130–136 (2019)

    2019

  63. [63]

    Rahmandad, H., Lim, T. Y. & Sterman, J. Behavioral dynamics of COVID-19: Estimating underreporting, multiple waves, and adherence fatigue across 92 nations. System Dynamics Review 37, 5–31 (2021)

    2021

  64. [64]

    & Ghaffarzadegan, N

    Rahmandad, H., Xu, R. & Ghaffarzadegan, N. A missing behavioural feedback in COVID-19 models is the key to several puzzles. BMJ Global Health 7, e010463 (2022)

    2022

  65. [65]

    Allcott, H. et al. Polarization and public health: Partisan differences in social distancing during the coronavirus pandemic. Journal of Public Economics 191, 104254 (2020)

    2020

  66. [66]

    & Syverson, C

    Goolsbee, A. & Syverson, C. Fear, lockdown, and diversion: Comparing drivers of pandemic economic decline 2020. Journal of Public Economics 193, 104311 (2021)

    2020

  67. [67]

    & Qiu, T

    Painter, M. & Qiu, T. Political beliefs affect compliance with COVID-19 social distancing orders. Journal of Economic Behavior & Organization 185, 688–701 (2021)

    2021

  68. [68]

    Holtz, D. et al. Interdependence and the cost of uncoordinated responses to COVID-19. Proceedings of the National Academy of Sciences 117, 19837–19843 (2020)

    2020

  69. [69]

    & Srivastava, S

    de Vaan, M., Mumtaz, S., Nagaraj, A. & Srivastava, S. B. Social learning in the COVID-19 pan- demic: Community establishments’ closure decisions follow those of nearby chain establishments. Management Science 67, 4446–4454 (2021)

    2021

  70. [70]

    & Heydari, B

    Ilami, S., Cao, Q. & Heydari, B. Network topology matters, but not always: Mobility networks in epidemic forecasting. arXiv preprint arXiv:2510.20025 (2025)

    Pith/arXiv arXiv 2025

  71. [71]

    & Prina, S

    Comola, M. & Prina, S. Saving accounts’ interplay with network-based financial arrangements: Evidence from a field experiment. The Economic Journal 133, 516–535 (2023)

    2023

  72. [72]

    Clauset, A., Newman, M. E. J. & Moore, C. Finding community structure in very large networks. Physical Review E 70, 066111 (2004). 20

    2004

  73. [73]

    Rousseeuw, P. J. & Van Driessen, K. A fast algorithm for the minimum covariance determinant estimator. Technometrics 41, 212–223 (1999)

    1999

  74. [74]

    The Statistical Analysis of Compositional Data Monographs on Statistics and Applied Probability (Chapman & Hall, London, 1986)

    Aitchison, J. The Statistical Analysis of Compositional Data Monographs on Statistics and Applied Probability (Chapman & Hall, London, 1986)

    1986

  75. [75]

    inside” from “outside

    Garreau, D., Jitkrittum, W. & Kanagawa, M. Large sample analysis of the median heuristic. arXiv preprint arXiv:1707.07269 (2017). 21 Supplementary Information Supplementary Methods This section collects the construction details deferred from the main-text Methods: the continuous demographic-similarity matrix D, the community-detection resolution sweep, an...

    Pith/arXiv arXiv 2017