Predicting human interactive learning by regret-driven neural networks

Science

Volumn 319, Issue 5866, 2008, Pages 1111-1113

  Marchiori, Davide ^a   Warglien, Massimo ^b  

^a UNIVERSITY OF TRENTO   (Italy)

^b CA' FOSCARI UNIVERSITY OF VENICE   (Italy)

Author keywords

[No Author keywords available]

Indexed keywords

ARTIFICIAL NEURAL NETWORK; DECISION MAKING; LEARNING; MODELING; PREDICTION;

ACCURACY; ARTICLE; BEHAVIORAL SCIENCE; DECISION MAKING; FEEDBACK SYSTEM; GAME; LEARNING; MODEL; NERVE CELL NETWORK; PREDICTION; PRIORITY JOURNAL;

COMPUTER SIMULATION; DECISION MAKING; ECONOMICS; EMOTIONS; GAMES, EXPERIMENTAL; HUMANS; LEARNING; MATHEMATICS; MODELS, PSYCHOLOGICAL; NEURAL NETWORKS (COMPUTER);

EID: 39749189365     PISSN: 00368075     EISSN: 10959203     Source Type: Journal    
DOI: 10.1126/science.1151185     Document Type: Article

References (29)

1. C. F. Camerer, Science 300, 1673 (2003).
2. M. Kosfeld et al., Nature 435, 673 (2005).
3. N. D. Daw, J. P. O'Doherty, P. Dayan, B. Seymour, R. J. Dolan, Nature 441, 876 (2006).
4. D. Sgroi, D. J. Zizzo, Physica A 375, 717 (2007).
5. C. F. Camerer, Behavioral Game Theory (Princeton Univ. Press, Princeton, NJ, 2003).
6. A. Roth, I. Erev, Games Econ. Behav. 8, 164 (1995).
7. I. Erev, A. Roth, Am. Econ. Rev. 88, 848 (1998).
8. I. Erev, A. Roth, L. Slonim, G. Barron, Econ. Theory 33, 29 (2007).
9. C. F. Camerer, T. H. Ho, Econometrica 67, 827 (1999).
10. T. H. Ho, C. F. Camerer, J. K. Chong, J. Econ. Theory 133, 177 (2007).
11. E. Ert, I. Erev, J. Behav. Decision Making 20, 305 (2007).
12. J. Hertz, A. Krogh, R. G. Palmer, Introduction to the Theory of Neural Computation (Addison-Wesley, Redwood City, CA, 1991).
13. J. J. Hopfield, Proc. Natl. Acad. Sci. U.S.A. 84, 8429 (1987).
14. R. Selten, R. Stöcker, J. Econ. Behav. Organ. 7, 47 (1986).
15. R. Selten, K. Abbink, R. Cox, Exp. Econ. 8, 5 (2005).
16. S. Hart, A. Mas-Collel, Econometrica 68, 1127 (2000).
17. N. Camille et al., Science 304, 1167 (2004).
18. G. Coricelli et al., Nat. Neurosci. 8, 1255 (2005).
19. D. Lee, Nature 441, 822 (2006).
20. P. Suppes, R. C. Atkinson, Markov Learning Models for Multiperson Interactions (Stanford Univ. Press, Stanford, CA, 1960).
21. D. Malcolm, B. Lieberman, Psychon. Sci. 12, 373 (1965).
22. B. O'Neill, Proc. Natl. Acad. Sci. U.S.A. 84, 2106 (1987).
23. A. Rapoport, R. B. Boebel, Games Econ. Behav. 4, 261 (1992).
24. J. Ochs, Games Econ. Behav. 10, 202 (1995).
25. J. Avrahami, W. Guth, Y. Kareev, Theory Decision 59, 255 (2005).
26. J. R. Busemeyer, Y. M. Wang, J. Math. Psychol. 44, 171 (2000).
27. R. Selten, Exp. Econ. 1, 43 (1998).
28. In a recent comparison of different learning models (7), NFP was tested on 10 of the games included in our data set, estimating its parameters on a different set of data. In this case, NFP did not perform better than the other models, including REL and reinforcement learning, confirming that (with two free parameters) the choice of the data set on which parameters are estimated may be a source of instability of the predictive performance of the model.
29. I. Erev, J. Ochs, and B. O'Neill kindly provided to us their original data sets. Discussions with R. Selten were a major source of inspiration for the project. P. Pellizzari, C. Gilbert, the anonymous reviewers and the editor helped us to substantially improve the paper. Ministero dell'Università e della Ricerca projects Fondo per gli Investimenti della Ricerca di Base RBNE03A9A7 and Progetti di Ricerca di Interesse Nazionale 2005139342 provided financial support for the research.