抄録
Methods for reliably assessing personality in primate species will be extremely useful for researchers interested in the biological basis, evolution, or socio-ecological significance of personality. Subjective assessment has been a popular method for rating primate personality, but the ambiguities in the definitions of adjectives to be rated and the inconsistency in the lists of those adjectives among the different research groups has made comparison and integrated understanding of accumulated results quite difficult.
To avoid the above intrinsic problems of subjective assessment method, personality traits must be scored based on behavioral measures (in either natural or experimental settings) rather than having observers subjectively rating the animals based on their impressions. The former method may be called objective assessment.
In objective assessment of personality, validity of the selected measures must always be considered and the effort to raise it is required. There are four kinds of validity; content validity, predictive validity, concurrent validity, and construct validity. Each type of validity can be measured and raised by following certain procedures.
High validity is necessary to ensure that the selected behavioral measure actually represents what is supposed to be measured. For example, Mehlman et al. (1994) claim that in rhesus monkeys the relative frequency of long leaps is a measure of loss of impulse control. But without checking its validity, universal utilization of this measure as a representation of impulsiveness must be avoided, for the measure may merely be a reflection of individual difference in, for example, motor ability.
Future researches on primate personality should employ objective assessment method rather than subjective assessment method, with careful consideration to its validity.
