To investigate whether there were any differences between the sexes, Mann-Whitney U tests were performed since the data did not conform to normal distributions. For variables where sexes did not differed, data for both sexes was pooled for further analyses.

Because all behaviours were not normally distributed, Spearman rank correlations were performed to investigate whether behaviours were stable or flexible across the different time periods. Spearman rank correlations were also used to investigate the relationship among the different behaviours across contexts.

The correlational structure among the behavioural variables across time was investigated in SPSS, using transformed Pearson correlations.

Kruskal wallis tests were used to test the difference between developmental periods, as the r_s values from the spearman rank correlations suggested that the bird’s behaviour was effected by developmental changes such as weaning, sexual maturity and the period between weaning and sexual maturity. For the vigilance in the novel arena, novel object and startle tests and latency to stand in the tonic immobility test, the mean was taken the 4, 6, 8 and 16 weeks (before sexual maturity) and 34 and 40 weeks (after sexual maturation). For activity, number of vocalisations and latency to walk in the novel arena, the means was taken of 4 and 6 weeks (before weaning), 8 and 16 weeks (the period between weaning and sexual maturity, and 34 and 40 weeks (after sexual maturity).