Standardisierung, Heterogenisierung und Reproduzierbarkeit von Ergebnissen aus Tierversuchen
Biologie des Verhaltens und der Sinne
Zusammenfassung der Projektergebnisse
The ongoing debate on the “reproducibility crisis” in the life sciences highlights the need for a rethinking of current best practices. Since the trend towards ever more standardized experiments is at risk of causing highly idiosyncratic and experiment-specific results, an alternative approach has been suggested to improve the robustness of findings, in particular from animal experiments. This concept, referred to as “systematic heterogenization postulates increased external validity and hence, improved reproducibility by introducing variation systematically into a single experiment. However, the implementation of this concept in practice requires the identification of suitable heterogenization factors that interact with the treatment under investigation in a way that makes the study samples more representative and the results more robust across the inevitable variation between experiments. As it is still not known which factors are suitable candidates in this respect, the aim of the here proposed project was to evaluate the potential of testing time, batch, and experimenter as feasible and easy-to-handle heterogenization factors for the use in single-laboratory studies. In a series of behavioural experiments with laboratory mice, we could show that the systematic variation of these factors within one experiment can indeed improve the reproducibility of treatment effects (in this case strain differences) between replicate experiments within one laboratory. At the same time, we could show that in a real multi-laboratory situation, the implementation of simple and effective heterogenization strategies is probably much more difficult than originally thought. Nonetheless, we are convinced that our findings make a significant contribution to a fundamental issue of experimental design and conduct in laboratory animal science. They emphasize the need for a paradigm shift and show one possible way out of the reproducibility crisis. Although tested by the example of physiological and behavioural mouse strain differences, benefits of applying such a heterogenized design may not be limited to the field of animal research. A study investigating net-legume effects in a simple microcosm experiment has already shown that the introduction of heterogeneity in the experimental design based on both genetic and environmental variation can also improve the reproducibility of ecological studies. Therefore, using more heterogeneous study populations may also benefit the external validity and hence reproducibility in other research branches throughout the life sciences. Furthermore, public support for animal experiments is granted on the explicit understanding that the research is relevant and produces valid and reproducible results using the smallest possible number of animals. In the present project, we repeatedly demonstrate that current best practice, i.e. conducting highly standardized experiments, creates the risk for obtaining spurious results, thereby compromising the validity and reproducibility of experimental outcomes. At the same time, our findings demonstrate that simple modifications of the design might help to improve the situation without the need for larger sample sizes. Consequently, this kind of research may also help to shape future public opinion in a more positive way.
Projektbezogene Publikationen (Auswahl)
- (2016): Improving external validity of experimental animal data. In: Animal Models for Human Cancer: Discovery and Development of Novel Therapeutics by Martic- Kehl MI, Schubiger PA (Eds), within the series “Methods and Principles in Medical Chemistry” edited by R. Mannhold, H. Kubinyi, G. Folkers, Wiley-VCH, Weinheim, 41-60
Richter SH, Spinello C, Macrì S
(Siehe online unter https://doi.org/10.1002/9783527695881.ch4) - (2017): Systematic heterogenization for better reproducibility in animal experimentation. Lab Animal, 46, 343 – 349
Richter SH
(Siehe online unter https://doi.org/10.1038/laban.1330) - (2019): Have I been here before? Effects of repeated testing and age on the results of behavioural tests. Behavioural Brain Research, 367, 143-148
Von Kortzfleisch VT, Kästner N, Prange L, Kaiser S, Sachser N, Richter SH
(Siehe online unter https://doi.org/10.1016/j.bbr.2019.03.042) - (2019): Heterogenising study samples across testing time improves reproducibility of behavioural data. Scientific Reports, 9, 8247
Bodden C, von Kortzfleisch VT, Karwinkel F, Kaiser S, Sachser N, Richter SH
(Siehe online unter https://doi.org/10.1038/s41598-019-44705-2) - (2020): Automated home-cage testing as a tool to improve reproducibility of behavioral research? Frontiers in Neuroscience, 14, 383
Richter SH
(Siehe online unter https://doi.org/10.3389/fnins.2020.00383) - (2020): Improving reproducibility in animal research by splitting the study population into several ‘mini-experiments’. Scientific Reports, 10, 16579
Von Kortzfleisch VT, Karp NA, Palme R, Kaiser S, Sachser N, Richter SH
(Siehe online unter https://doi.org/10.1038/s41598-020-73503-4) - (2020): It is time for an empirically informed paradigm shift in animal research. Nature Reviews Neuroscience, 21, 660
Richter SH, von Kortzfleisch V
(Siehe online unter https://doi.org/10.1038/s41583-020-0369-0) - (2020): Reproduzierbarkeit von Verhaltensdaten (Kapitel 8.3). In: Methoden der Verhaltensbiologie by Naguib M, Krause ET (Eds), Springer Spektrum, 2nd edition, Berlin, 102-106
Richter SH
