Fig 1.
Four characteristics of the game of Bridge.
Interactions between players are competitive and cooperative. This is an incomplete information game (players know only their hand) with first, a communicational part (bids implying representations about the real meaning), and then, a strategic part (declarer plays using probabilistic reasoning and inferences).
Fig 2.
Hierarchical clustering analysis.
(A) Dendrogram presenting the results of the hierarchical clustering separated in five clusters including n items. (B) Mean adjusted rand index measuring the similarity between the dendrogram on the left and a dendrogram obtained on hundred bootstraps for every number of clusters from 2 to 10. Error bars show the standard error. (C) Significant contribution of items composing each cluster. The items contributing significantly are the items that contribute more than 100 divided by the number of items included in the cluster (expected contribution under the hypothesis that each item is contributing equally).
Fig 3.
Correlation between bridge-related traits and personality traits. Correlation coefficients are Pearson’s r and only significant coefficients are shown. O: Openness, C: Conscientiousness, E: Extraversion, A: Agreeableness, N: Neuroticism.
Fig 4.
Model fit (A) and participants’ mean scores (B) for the Gaussian Mixture model.
(A) Estimated mean z-score of the three gaussian components on each bridge-related trait and standard deviation (error bars). (B) Mean participants’ scores within each type (see S3 Table for descriptive statistics and comparisons with null hypothesis). Error bars represent standard errors; all within traits pairwise comparisons are significant (p < .05 Tukey corrected) except that Type 1 and 3 do not differ on Emotionality (p > .05).
Fig 5.
Mutual relationships between the personality-based description and the bridge-related description.
(A) Proportion of players of each personality type in the three types of bridge players. Personality types (from P.1 to P.5) represent the five personality types we found whereas colors represent the four Gerlach & al.’s types (see S1 Fig for more details). (B) Differences in correct classification rates using personality traits to predict bridge players’ type compared to the use of bridge traits and random variables. Error bars represent 95% confidence intervals (the right one is too small to be seen).