Evaluation of the elevated plus-maze and open-field tests for the assessment of anxiety-related behaviour in inbred mice.

The elevated plus-maze test (PM) and open-field test (OF) are routinely used to study anxiety-related behaviour in mouse. However, the data obtained with these tests have often been contradictory, probably because of differences between laboratories in the selection and analysis of behavioural parameters. We have characterised the pattern of mouse anxiety by analysing a number of behavioural parameters with both PM and OF in BALB/c and C57BL/6 mice, two behaviourally distant mouse strains. Twenty-eight variables (15 analysed with PM and 13 with OF) were selected by correlation analysis from those initially recorded with both tests. The scores of the selected variables were first analysed by MANOVAs, and then by principal component analysis (PCA). PCA extracted five factors for PM and four factors for OF. These factors were subjected to a correlation analysis, which showed significant correlation between four of them. The factorial scores of BALB/c and C57BL/6 mice were analysed by MANOVAs, which showed significant effects of both the strain and test used. Our results confirm the multidimensional structure of mouse anxiety-related behaviour as regards both simple components and functional interactions, and comprehensively represent strain- and test-specific features of mouse anxiety-related behaviour. Copyright 2002 Elsevier Science B.V.