Elsevier

Physiology & Behavior

Volume 87, Issue 1, 30 January 2006, Pages 95-102
Physiology & Behavior

The use of behavioral test batteries, II: Effect of test interval

https://doi.org/10.1016/j.physbeh.2005.09.002Get rights and content

Abstract

Test batteries are commonly used to assess the behavioral phenotype of genetically modified and inbred strains of mice. However, few systematic studies have been employed to address several key issues concerning the use of a test battery. The current study was designed to address whether inter-test interval affects behavioral performance. Male mice of 3 different inbred strains and one F1 hybrid strain were randomly assigned to either a test battery with 1 week inter-test intervals, or a rapid test battery with 1–2 day inter-test intervals. The test battery included a neurological exam, open-field activity, light–dark exploration, rotarod test, prepulse inhibition, and startle habituation. The experiment was repeated with female animals of 2 different strains. As expected, there were strain differences on each of the behavioral assays; however, there was no major difference in performance between mice of the standard test battery and the rapid test battery. Similar results were found with females. These results indicate that the interval between most tests could be as little as 1–2 days, with little significant effect on overall performance. Thus, it is possible with the current test battery to reduce the inter-test interval to facilitate the rate of studying and identifying behavioral phenotypes in mice.

Introduction

Behavioral test batteries are commonly utilized to evaluate the behavioral responses of transgenic and knockout mice (for review, see Refs. [3], [13], [15]). Among the advantages are the use of fewer mice and the ability to assess multiple and overlapping domains of CNS function within the same mouse, which increases the number of phenotypes identified and improves interpretation of observable phenotypic differences. Our laboratory has routinely and successfully used a test battery described by Crawley and Paylor [3] to identify behavioral phenotypes in several lines of mutant mice (e.g., Refs. [1], [4], [7], [8], [9], [10], [14]). Behavioral assays in the test battery were selected to represent several, overlapping domains of CNS function and performed in an order such that the tests thought to be the least invasive were performed prior to those thought to be more invasive [3]. In addition, the inter-test interval varied from one to four weeks to allow the mice adequate recovery time between tests and reduce the likelihood of previous test experience. However, it was not known to what extent these issues influenced test performance, therefore systematic studies (e.g., Refs. [5], [11], [12]) to examine the effects of test history, test order and inter-test interval were initiated.

In a previous study, we addressed how test history and test order affected the performance of inbred mice assessed in our standard test battery [6]. Some of the behavioral tests in the test battery were sensitive to the effects of prior test experience while other tests were immune to these effects. Test order did not have a dramatic impact on behavioral test performance. In the current study, we investigated the effects of inter-test interval on the behavioral responses of inbred mice in our standard test battery compared to a rapid test battery in which tests were administered with a 1–2 day inter-test interval. We were particularly interested in determining whether the test battery could be performed with shorter inter-test intervals to facilitate rapid, high-throughput behavioral phenotyping screens. Except for the open-field test, there were no significant effects of inter-test interval, validating the use of rapid test batteries for high-throughput behavioral phenotyping.

Section snippets

Animals

Male mice of 3 different inbred strains and 1 F1 cross were used for the first experiment: 18 male C57BL/6J, 14 male 129S5/SvEvBrd, 18 male FVB/NJ, and 13 male mice from a 129S6/SvEvTac X C57BL/6J F1 cross (S6B6F1). C57BL/6J (B6), FVB/NJ, and 129S5/SvEvBrd (129S5) were selected as representative strains because they are commonly used in studies with knockout and transgenic mice. In the second experiment, 17 C57BL/6J and 15 129S5/SvEvBrd female mice were used. Parental stock for C57BL/6J and

Results

In the first experiment the effect of inter-test interval was evaluated in 4 strains of male mice. In order to assess whether female mice might be more susceptible to inter-test interval, a second experiment was repeated using female B6 and 129S5 mice. The female strains were chosen because they were likely to represent the behavioral extremes of variation between strains, based on the results of the first experiment with male mice. Although the male and female experiments were performed

Discussion

A summary of our findings is presented in Table 1. As expected, strain-dependent effects were observed for most of the behaviors measured. However, we found very little effect of reducing the inter-test intervals from 1 week to 1–2 days on the overall performance of male or female mice assessed in an abbreviated version of the test battery described by Crawley and Paylor [3]. In the current study the test that appeared to be the most sensitive to the effects of inter-test interval was the

Acknowledgements

This work was supported by the National Institutes of Health, NIA (to RP) and NICHD (Baylor's MRDDRC Neurobehavioral Core).

Cited by (195)

  • Examining the effect of chronic intranasal oxytocin administration on the neuroanatomy and behavior of three autism-related mouse models

    2022, NeuroImage
    Citation Excerpt :

    Pilot testing involving wildtype mice was employed to ensure expected neuroanatomical and behavioral results before the main study commenced. Although it is typical to space behavioral tests apart by several days, the methods employed were developed to increase throughput to maximize the number of subjects in the study, thereby increasing the statistical power to detect a treatment effect (Crawley, 2007; Paylor et al., 2006). Data from this study was compared to the literature as well as previous data acquired from the lab to ensure that the distribution of the data was expected.

View all citing articles on Scopus
View full text