Behavioral evaluation and basil supplementation

Published: 26 September 2023| Version 1 | DOI: 10.17632/53tf32z76p.1
Alberto da Silva Moraes


It contains original data from three behavioral tests used in the study led by Miss Damaris Costa, between 2019 and 2022. The study utilized female C57Bl mice, distributed into four experimental groups, G1, control mature adults, G2 basil leaves supplemented mature adults, G3, old adults, and G4, basil leaves supplemented old adults. Old adults were supplemented long term before getting old. About 1.2 mg of fresh basil leaves were homogenized in distilled water, and mixed with 5 g of shreded rodent chow. After reppeletting, the mixture was dried in the oven at 40 ºC, and given to the animals ad libitum. The animals were tested using elevated plus maze, Morris water maze, and novel object recognition. All sessions were video recorded, and analyzed on ImageJ using the MouBeAT plugin.


Steps to reproduce

1. Separate female C57Bl into 5 or 6 animals per cage 2. Rear under normal conditions until the desired age 3. Give experimental groups either the control chow or the supplemented chow. Measure the amount (g) of chow being put on one day, and the amount still not eaten 24 hours later. 4. The treatment group of mature adults need to be up to 9 months old and receive the supplemented chow for at least 4 months, before reaching desired age 5. The treatment group of old adults need to be more than 18 months old and receive the supplemented chow, for at least 6 to 8 months before reaching the desired age. 6. When reaching the age, each animal must be marked using any available method for animal identification, in compliance with APA guidelines. 7. Test in novel object recognition using the protocol of Lueptow (2017), but using a cylindrical apparatus. 8. Test in the Morris Water maze using the single-day protocol of Nunez (2008), and with a central platform, and no quadrants. 9. Test in the elevated plus maze using the two-day protocol of Kouémou et al. (2017). 10. Record all sessions using a video camera or a mobile. High-quality video files are preferable, though not mandatory. 11. Analise the videos in ImageJ using the MouBeAT plugin. 12. For less than 15 animals in each group, use non-parametric statistics. Always test data for homoscedasticity and for similarities in data distribution before choosing statistical tests. 13. To compare body weights, Mood's medians test was used, along with Grissom's probability of superiority (Ps) as an effect size estimator (Grissom, 1994) 14. For the discrimination indexes (d2) from the NOR task, group medians were compared after testing for homoscedasticity. This comparison was conducted using the Kruskal-Wallis test, followed by the Conover-Iman test for multiple comparisons (Conover & Iman, 1979). Effect sizes were determined using the corrected Hedge's g estimator. 15. Transfer latencies (TL) from elevated plus maze resulted in heteroscedasticity, so the comparison TL on training day x TL on the trial day for each experimental group was done using the Wilcoxon test, and Ps as effect size estimator 16. The comparison total time in closed x total time in open arms for each experimental group resulted in homoscedasticity, and the hypothesis was tested using a Mann-Whitney's median test, and Hedge's g effect size estimator 17. Mood's and Mann-Whitney's median tests, as well as the Kruskal-Wallis test followed by Conover-Iman, were performed using the "RVAideMemoire" and "conover.test" packages in R, respectively. The Hedge's g estimator and Grissom's probability of superiority were calculated using custom implementations in Microsoft Excel worksheets.


Universidade Federal de Uberlandia - Campus Umuarama


Nutrition, Animal Cognition, Learning, Memory, Aging, Basil