1: \begin{abstract} {\it Background.} Confirmation bias is the tendency

2:   to acquire or evaluate new information in a way that is consistent

3:   with one's preexisting beliefs. It is omnipresent in psychology,

4:   economics, and even scientific practices. Prior theoretical research

5:   of this phenomenon has mainly focused on its economic implications

6:   possibly missing its potential connections with broader notions of

7:   cognitive science.

8: 

9:     {\it Methodology/Principal Findings.}  We formulate a

10:     (non-Bayesian) model for revising subjective probabilistic opinion

11:     of a confirmationally-biased agent in the light of a persuasive

12:     opinion. The revision rule ensures that the agent does not react

13:     to persuasion that is either far from his current opinion or

14:     coincides with it. We demonstrate that the model accounts for the

15:     basic phenomenology of the social judgment theory, and allows to

16:     study various phenomena such as cognitive dissonance and boomerang

17:     effect. The model also displays the order of presentation

18:     effect|when consecutively exposed to two opinions, the preference

19:     is given to the last opinion (recency) or the first opinion

20:     (primacy)|and relates recency to confirmation bias. Finally, we

21:     study the model in the case of repeated persuasion and analyze its

22:     convergence properties.

23: 

24:     {\it Conclusions.} {The standard Bayesian approach to

25:       probabilistic opinion revision is inadequate for describing the

26:       observed phenomenology of persuasion process.} The simple

27:     non-Bayesian model proposed here does agree with this

28:     phenomenology and is capable of reproducing a spectrum of effects

29:     observed in psychology: primacy-recency phenomenon, boomerang

30:     effect and cognitive dissonance. We point out several limitations

31:     of the model that should motivate its future development.

32: 

33: 

34: \end{abstract}