Fig. 5: Effect of information about triangle relationships on the normalized balance (\({\hat{f}}\)) achieved by groups of participants receiving no information (red), irrelevant information (yellow), or relevant information about one of their triangles (blue). | npj Complexity

Fig. 5: Effect of information about triangle relationships on the normalized balance (\({\hat{f}}\)) achieved by groups of participants receiving no information (red), irrelevant information (yellow), or relevant information about one of their triangles (blue).

From: Experimental evidence confirms that triadic social balance can be achieved through dyadic interactions

Fig. 5

Shaded areas represent the 95% confidence intervals. The initial balance (on average acros s groups, \({f}_{1}=0.27\)) is set to \({\hat{f}}_{1}=0\) for easier comparison of tasks and experimental conditions. Note that balance forms very rapidly, reaching \({f}_{4}=0.62\) (\({\hat{f}}_{4}=0.35\) after normalization) in only three study waves. The achieved level of balance corresponds better to the predictions of the dyadic model than of the triadic model (Fig. 3).

Back to article page