Modeling biased information seeking with second order probability distributions

Gernot D. Kleiter

Kybernetika (2015)

  • Volume: 51, Issue: 3, page 469-485
  • ISSN: 0023-5954

Abstract

top
Updating probabilities by information from only one hypothesis and thereby ignoring alternative hypotheses, is not only biased but leads to progressively imprecise conclusions. In psychology this phenomenon was studied in experiments with the “pseudodiagnosticity task”. In probability logic the phenomenon that additional premises increase the imprecision of a conclusion is known as “degradation”. The present contribution investigates degradation in the context of second order probability distributions. It uses beta distributions as marginals and copulae together with C-vines to represent dependence structures. It demonstrates that in Bayes' theorem the posterior distributions of the lower and upper probabilities approach 0 and 1 as more and more likelihoods belonging to only one hypothesis are included in the analysis.

How to cite

top

Kleiter, Gernot D.. "Modeling biased information seeking with second order probability distributions." Kybernetika 51.3 (2015): 469-485. <http://eudml.org/doc/271608>.

@article{Kleiter2015,
abstract = {Updating probabilities by information from only one hypothesis and thereby ignoring alternative hypotheses, is not only biased but leads to progressively imprecise conclusions. In psychology this phenomenon was studied in experiments with the “pseudodiagnosticity task”. In probability logic the phenomenon that additional premises increase the imprecision of a conclusion is known as “degradation”. The present contribution investigates degradation in the context of second order probability distributions. It uses beta distributions as marginals and copulae together with C-vines to represent dependence structures. It demonstrates that in Bayes' theorem the posterior distributions of the lower and upper probabilities approach 0 and 1 as more and more likelihoods belonging to only one hypothesis are included in the analysis.},
author = {Kleiter, Gernot D.},
journal = {Kybernetika},
keywords = {probability logic; Bayes' theorem; degradation; pseudodiagnosticity task; second order probability distributions; probability logic; Bayes' theorem; degradation; pseudodiagnosticity task; second order probability distributions},
language = {eng},
number = {3},
pages = {469-485},
publisher = {Institute of Information Theory and Automation AS CR},
title = {Modeling biased information seeking with second order probability distributions},
url = {http://eudml.org/doc/271608},
volume = {51},
year = {2015},
}

TY - JOUR
AU - Kleiter, Gernot D.
TI - Modeling biased information seeking with second order probability distributions
JO - Kybernetika
PY - 2015
PB - Institute of Information Theory and Automation AS CR
VL - 51
IS - 3
SP - 469
EP - 485
AB - Updating probabilities by information from only one hypothesis and thereby ignoring alternative hypotheses, is not only biased but leads to progressively imprecise conclusions. In psychology this phenomenon was studied in experiments with the “pseudodiagnosticity task”. In probability logic the phenomenon that additional premises increase the imprecision of a conclusion is known as “degradation”. The present contribution investigates degradation in the context of second order probability distributions. It uses beta distributions as marginals and copulae together with C-vines to represent dependence structures. It demonstrates that in Bayes' theorem the posterior distributions of the lower and upper probabilities approach 0 and 1 as more and more likelihoods belonging to only one hypothesis are included in the analysis.
LA - eng
KW - probability logic; Bayes' theorem; degradation; pseudodiagnosticity task; second order probability distributions; probability logic; Bayes' theorem; degradation; pseudodiagnosticity task; second order probability distributions
UR - http://eudml.org/doc/271608
ER -

References

top
  1. Boole, G., An Investigation of the Laws of Thought., Macmillan/Dover Publication, New York 1854/1958. Zbl1205.03003MR1802120
  2. Coletti, G., Petturiti, D., Vantaggi, B., Bayesian inference: the role of coherence to deal with a prior belief function., Statist. Methods Appl., online, 2014. MR3278926
  3. Coletti, G., Scozzafava, R., Probabilistic Logic in a Coherent Setting., Kluwer, Dordrecht 2002. Zbl1040.03017MR2042026
  4. Doherty, M. E., Mynatt, C. R., Tweney, R. D., Schiavo, M. D., 10.1016/0001-6918(79)90017-9, Acta Psychologica 43 (1979), 111-121. MR1872198DOI10.1016/0001-6918(79)90017-9
  5. Gilio, A., 10.1016/j.ijar.2011.08.004, Int. J. Approx. Reasoning 53 (2012), 413-434. MR2902403DOI10.1016/j.ijar.2011.08.004
  6. Hanea, A., 10.1142/9789814299886_0014, In: Dependence Modeling. Vine Copula Handbook (D. Kurowicka and H. Joe, eds.), chapter Non-parametric Bayesian belief nets versus vines, World Scientific, New Jersey 2011, pp. 281-303. MR2856979DOI10.1142/9789814299886_0014
  7. Joe, H., Dependence Modeling with Copulas., Chapman and Hall/CRC, Boca Raton 2015. MR3328438
  8. Kern, L., Doherty, M. E., “Pseudodiagnosticity” in an idealized medical problem-solving environment., J. Medical Education 57 (1982), 100-104. 
  9. Kleiter, G. D., 10.1016/s0004-3702(96)00021-5, Artificial Intelligence 88 (1996), 143-161. Zbl0906.68114DOI10.1016/s0004-3702(96)00021-5
  10. Kleiter, G. D., 10.1007/978-3-642-33042-1_44, In: Synergies of Soft Computing and Statistics for Intelligent Data Analysis (R. Kruse, M.xQ,R. Berthold, C. Moewes, M. A. Gil, P. Grzegorzewski, and O. Hryniewicz, eds.), Advances in Intelligent Systems and Computation 190, Springer, Heidelberg 2012. pp. 409-417. DOI10.1007/978-3-642-33042-1_44
  11. Kurowicka, D., Cooke, R., Distribution-free continuous Bayesian belief nets., In: Proc. Fourth International Conference on Mathematical Methods in Reliability Methodology and Practice, Santa Fe 2004. Zbl1083.62054MR2230715
  12. Kurowicka, D., Cooke, R., Uncertainty Analysis with High Dimension Dependence Modelling., Wiley, Chichester, 2006. MR2216540
  13. Kurowicka, D., Joe, R., Dependence Modeling: Vine Copula Handbook., World Scientific, Singapure 2011. MR2849701
  14. Mai, J.-F., Scherer, M., Simulating Copulas. Stochastic Models, Sampling Algorithms, and Applications., Imperial College Press, London 2012. Zbl1301.65001MR2906392
  15. Nelsen, R. B., An introduction to Copulas., Springer, Berlin 2006. Zbl1152.62030MR2197664
  16. Team, R Development Core, Vienna, Austria, R: A Language and Environment for Statistical Computing, 2014. 
  17. Schepsmeier, U., Stoeber, J., Brechmann, E. C., Graeler, B., Statistical inference of vine copulas., Version 1.2 edition, 2013. 
  18. Seidenfeld, T., Wasserman, L., 10.1214/aos/1176349254, Ann. Statist. 21 (1993), 1139-1154. Zbl0796.62005MR1241261DOI10.1214/aos/1176349254
  19. Tweney, R. D., Doherty, M. E., Kleiter, G. D., 10.1080/13546783.2010.525860, Thinking and Reasoning 16 (2010), 332-345. DOI10.1080/13546783.2010.525860
  20. Wallmann, C., Kleiter, G. D., 10.1007/978-3-642-31724-8_17, In: Communications in Computer and Information Science (S. Greco, B. Bouchon-Meunier, G. Coletti, M. Fedrizzi, B. Matarazzo, and R. R. Yager, eds.), IPMU (4) 300, Springer, Berlin 2012, pp. 157-167. Zbl1252.03043DOI10.1007/978-3-642-31724-8_17
  21. Wallmann, C., Kleiter, G. D., 10.14736/kyb-2014-2-0268, Kybernetika 50 (2014), 268-283. Zbl1296.03018MR3216994DOI10.14736/kyb-2014-2-0268
  22. Wallmann, C., Kleiter, G. D., 10.1007/s11225-013-9513-4, Studia Logica 102 (2014), 913-929. MR3249556DOI10.1007/s11225-013-9513-4
  23. Wasserman, L. A., 10.1214/aos/1176347511, Annals Statist. 18 (1990), 454-464. Zbl0711.62001MR1041404DOI10.1214/aos/1176347511

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

                
Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.