Individual Differences in Male Rats in a Behavioral Test Battery: A Multivariate Statistical Approach

Feyissa, Daniel D. and Aher, Yogesh D. and Engidawork, Ephrem and Höger, Harald and Lubec, Gert and Korz, Volker (2017) Individual Differences in Male Rats in a Behavioral Test Battery: A Multivariate Statistical Approach. Frontiers in Behavioral Neuroscience, 11. ISSN 1662-5153

[thumbnail of pubmed-zip/versions/1/package-entries/fnbeh-11-00026/fnbeh-11-00026.pdf] Text
pubmed-zip/versions/1/package-entries/fnbeh-11-00026/fnbeh-11-00026.pdf - Published Version

Download (636kB)

Abstract

Animal models for anxiety, depressive-like and cognitive diseases or aging often involve testing of subjects in behavioral test batteries. The large number of test variables with different mean variations and within and between test correlations often constitute a significant problem in determining essential variables to assess behavioral patterns and their variation in individual animals as well as appropriate statistical treatment. Therefore, we applied a multivariate approach (principal component analysis) to analyse the behavioral data of 162 male adult Sprague-Dawley rats that underwent a behavioral test battery including commonly used tests for spatial learning and memory (holeboard) and different behavioral patterns (open field, elevated plus maze, forced swim test) as well as for motor abilities (Rota rod). The high dimensional behavioral results were reduced to fewer components associated with spatial cognition, general activity, anxiety-, and depression-like behavior and motor ability. The loading scores of individual rats on these different components allow an assessment and the distribution of individual features in a population of animals. The reduced number of components can be used also for statistical calculations like appropriate sample sizes for valid discriminations between experimental groups, which otherwise have to be done on each variable. Because the animals were intact, untreated and experimentally naïve the results reflect trait patterns of behavior and thus individuality. The distribution of animals with high or low levels of anxiety, depressive-like behavior, general activity and cognitive features in a local population provides information of the probability of their appeareance in experimental samples and thus may help to avoid biases. However, such an analysis initially requires a large cohort of animals in order to gain a valid assessment.

Item Type: Article
Subjects: Asian STM > Biological Science
Depositing User: Managing Editor
Date Deposited: 17 Feb 2023 08:07
Last Modified: 06 Jul 2024 06:34
URI: http://journal.send2sub.com/id/eprint/727

Actions (login required)

View Item
View Item