Abstract
Animal personality can be defined as individual behavioural differences that are repeatable and consistent across time and/or contexts. In the model species great tit Parus major, studies on personality have shown consistent individual behaviour across behavioural axes, which have led to the use of only a single behavioural axis, ranging from shy to bold, to describe personality. However, personality studies have often focused on associations between only a few behavioural traits, and some of them have been restricted to behaviour in captivity. In addition, not all studies have found the same association between traits, questioning the use of only one personality axis. This study conducted four behavioural tests on a wild great tit population. The tests included the measurement of behavioural response towards a human when being handled, towards an intruding human and a (caged) conspecific in the incubation period, and the response towards a predator model (an owl) during the nestling period. In total, 15 behavioural traits were measured, to study whether personality could best be explained by: (H0) independent behavioural traits, (H1) groups of behavioural traits, or (H2) one personality axis ranging from shy to bold. Additional purposes of the study was to investigate which of the measured traits that may explain personality, and if the behavioural responses were associated with body size and/or differed between sex and age categories. The study found some evidence for repeatable behaviour within the study season. However, in general few significant associations were found between the traits, and the principal component analyses failed to reduce dimensionality of traits, supporting hypothesis H1. This is in agreement with personality studies on other taxa. Based on associations with boldness, some of the traits measured may be of help when characterising personality in wild great tits. The study found that males may be bolder than females, but no one-directional differences was found in behaviour between age classes, nor associations with body size. In addition, other factors, such as trial time/date may influence behaviour. Further studies testing repeatability for all traits and following individuals over several seasons and years are necessary to conclude whether the behavioural traits measured and associations found are consistent over time and/or contexts. Comparative studies with exploratory behaviour are also recommended for a better understanding of which of the traits measured could characterise personality.