On the reliability of unreliable information
Gossip as cultural memory
Dominic Mitchell | Department of Computer Science, University of Bath, UK
Joanna J. Bryson | Department of Computer Science, University of Bath, UK | Center for Information Technology Policy, Princeton University, USA
Paul Rauwolf | Department of Computer Science, University of Bath, UK | Institute for New Economic Thinking at the Oxford Martin School, University of Oxford, UK
Gordon P.D. Ingram | Department of Psychology, Universidad de Los Andes, Bogotá, Colombia
When individuals learn from what others tell them, the information is subject to transmission error that does not arise in learning from direct experience. Yet evidence shows that humans consistently prefer this apparently more unreliable source of information. We examine the effect this preference has in cases where the information concerns a judgment on others’ behaviour and is used to establish cooperation in a society. We present a spatial model confirming that cooperation can be sustained by gossip containing a high degree of uncertainty. Accuracy alone does not predict the value of information in evolutionary terms; relevance, the impact of information on behavioural outcomes, must also be considered. We then show that once relevance is incorporated as a criterion, second-hand information can no longer be discounted on the basis of its poor fidelity alone. Finally we show that the relative importance of accuracy and relevance depends on factors of life history and demography.
Keywords: gossip, reputation, cooperation, social norm
Article outline
- 1.Introduction
- 2.Methods
- 2.1Model
- 2.1.1Individual interactions: the donation game
- 2.1.2Reputation assigned by observers according to a social norm
- 2.1.3Reputation impacts on individual choice of action
- 2.1.4Evolution according to payoff
- 2.2Experiment 1. Judgments made without full knowledge
- 2.2.1Method
- 2.2.2Simulation
- 2.2.3Results and discussion
- 2.3Experiment 2. Agents’ knowledge of reputations differs according to the spread of gossip
- 2.3.1Method
- 2.3.2Results and discussion
- 2.3.3Formal analysis of an ALLD agent’s reputation
- 2.3.4Results and discussion
- 2.4Experiment 3. Personal experience versus gossip under limited lifespan
- 2.4.1Method
- 2.4.2Results and discussion
- 2.4.3Formal analysis of gossip versus direct observation
- 2.4.4Results and discussion
- 2.1Model
- 3.General discussion
- 4.Conclusion
- Acknowledgments
- Note
-
References
Published online: 07 October 2016
https://doi.org/10.1075/is.17.1.01mit
https://doi.org/10.1075/is.17.1.01mit
References
Abbink, K., Brandts, J., Herrmann, B., and Orzen, H.
Alexander, R. D.
Alizon, S. and Taylor, P.
Asch, S. E.
Dawkins, R.
Enquist, M. and Leimar, O.
Fahle, M. and Poggio, T.
Gergely, G., Bekkering, H., and Király, I.
Gergely, G. and Csibra, G.
Goldberg, D. and Deb, K.
Gruber, T., Poisot, T., Zuberbühler, K., Hoppitt, W., and Hobaiter, C.
Henrich, J. and McElreath, R.
Hess, N. and Hagen, E.
Jacobs, R. A., Jordan, M. I., Nowlan, S. J., and Hinton, G. E.
Kao, A. B. and Couzin, I. D.
Kokko, H.
McElreath, R. and Boyd, R.
Meltzoff, A. N.
Milinski, M., Semmann, D., and Krambeck, H.-J.
Nagell, K., Olguin, R. S., and Tomasello, M.
Nakamura, M. and Masuda, N.
Nowak, M. A. and Sigmund, K.
Ohtsuki, H. and Iwasa, Y.
Ohtsuki, H., Iwasa, Y., and Nowak, M. A.
Rand, D. G., Nowak, M. A., Fowler, J. H., and Christakis, N. A.
Rauwolf, P., Mitchell, D., and Bryson, J. J.
Richerson, P. J. and Boyd, R.
Roberts, G.
Scott-Phillips, T. C.
Sommerfeld, R. D., Krambeck, H. J., Semmann, D., and Milinski, M.
Taylor, D. J.
Tomasello, M.
Trivers, R. L.
Čače, I. and Bryson, J. J.
Whiten, A., McGuigan, N., Marshall-Pescini, S., and Hopper, L. M.
Cited by
Cited by 2 other publications
Okafuji, Yuki, Jun Baba, Junya Nakanishi, Itaru Kuramoto, Kohei Ogawa, Yuichiro Yoshikawa & Hiroshi Ishiguro
Rudnicki, Konrad, Carolyn Declerck, Charlotte De Backer & Mario Berth
This list is based on CrossRef data as of 15 april 2022. Please note that it may not be complete. Sources presented here have been supplied by the respective publishers. Any errors therein should be reported to them.