Kelly Spanou

 

Department of Behavioural Biology
Institute of Neuro- and Behavioural Biology
Badestr. 13
D-48149 Münster, Germany
Tel.: +49 251/83-23842
k.spanou@uni-muenster.de

Education

• Since 2024:
  PhD studies in the Department of Behavioural Biology, Institute of Neuro- and Behavioural Biology, University of Münster, Germany
• 2021-2023:
  Msc Neuroscience, Vrije Universiteit Amsterdam
• 2015-2021
  BSc Psychology, University of Crete

Work experience

• 2024:
  Junior Biotechnician, Netherlands Institute For Neuroscience, Amsterdam, Netherlands
• 2023
  Research Intern, Swammerdam Institute for Life sciences, Amsterdam, Netherlands
• 2022
  Research Intern, Center for Neurogenomics & Cognitive Research, Amsterdam, Netherlands
• 2018-2021
  Researcher, Laboratory of Behavioural Neuroscience, Rethymno

Supervisors

• Prof. Dr. Helene Richter, Institute for Neuro- and Behavioural Biology, Department of Behavioural Biology, University of Münster

Research interests

• Laboratory animal welfare,
• Animal personality,
• 3Rs (reduction, replacement, refinement)

PhD project description

The term "reproducibility crisis" describes a phenomenon that has impacted animal research over the past 20 years, characterized by the poor replicability of findings in the field of preclinical research. To address this ongoing issue, various underlying reasons have been identified, such as experimental design and statistical analysis pitfalls. Tools designed to mitigate these problems have become readily accessible to researchers and are integrated into the planning process. Furthermore, quality measures including statistical power calculations, randomization and blindness in the experimental design are now considered standard practice.

Traditionally, laboratory experiments are conducted under highly controlled environments and standardized conditions. This strict standardization process promotes within experiment homogenization, reduces between experiments variation, manages confounding factors and decreases the number of animals used. Consequently, standardization is believed to improve reproducibility and comparability of research findings. However, while this approach succeeds in minimizing confounders and measurement errors, the extent to which it addresses biological variation is questionable. Rigorously standardized experiments may yield in findings that are only applicable to a specific homogenized laboratory setting thereby lacking generalizability and comparability and potentially biological meaning.

Despite efforts at standardization and rigorous control, a notable amount of variability in animal experiments seems to arise from uncontrolled and obscure factors. While systematically integrating variation into the study design has been beneficial for improving consistency and reproducibility, poor intra-experimental control may lead to biased outcomes and hinder the detection of treatment effects. Thus, it is imperative to pinpoint the main contributors to variability and develop tools to effectively manage them both in experimental design and statistical analysis. This foundation prompts the following questions: (1) What are the factors that contribute to the observed residual variation? (2) How can we leverage this understanding to refine our experimental designs and enhance result reproducibility?

Our goal is to incorporate stable individual differences among animals, known as "animal personalities," into study designs to better manage variation within experiments and improve reproducibility. By including animal personalities in experimental group design and data analysis, we anticipate higher-quality results, fewer biases, and enhanced statistical power. Ultimately, our project takes an intradisciplinary approach as we are using insights from evolutionary biology to refine experimental approaches in basic and preclinical animal research. Thus, reproducibility would be enhanced without the need for larger sample sizes.