Causal maps (also called causality maps or explanation maps) are simplified schematic visualizations of [[mechanistic explanations]]. They are a tool to keep an overview of your research topic and can be helpful for [[Interdisciplinary integration]]. Building a causal map starts with an [[explanandum]], a thing-to-be-explained. This takes center stage in the map. Subsequently, those building a causal map can use the literature to explore a topic and adapt the causal map accordingly. For example, they can: 1. Add factors that help explain the explanandum 2. Add factors that follow from the explanandum 3. Decompose factors into multiple constituent parts While doing so, different elements of the map are connected by arrows that indicate causal relations. Be aware that the map reflects understanding of the topic and that it is fine for relations to be uncertain or hypothetical, as long as you keep track of that uncertainty. ## Dealing with correlations Sometimes, the research papers you have read have only shown correlations between factors without identifying underlying causal relations. The causal map, however, forces you to speculate on causal relations anyway. This might feel wrong (as perhaps it should), but it is permitted in this case because the map is not your definitive claim about how the world works, but rather a representation of how you think it *could* work, in order for you to reason about it. If you postulate a causal relation that is especially contested or uncertain, this can lead to several actions. Which one is right for you depends on your particular project 1. You should search for additional literature to further elucidate the correlation you are describing 2. You should design an experiment that can help differentiate between specific proposals for the causal relation 3. You should ensure to devote space to the contested relation when writing your discussion ## Interdisciplinary integration Since causal maps depict theoretical understanding of a topic (even if it's grounded in empirical research) it can be helpful for interdisciplinary integration at the theory level. Here, we can operationalize interdisciplinarity as follows: > Interdisciplinary integration at the theory level is explaining a phenomenon from one discipline in terms of those from other disciplines. This means that if different disciplines create a causal map for a phenomenon, a first step toward integration is to interconnect the different causal maps and see which factors are shared between disciplines. In practice, this can be complicated: it might mean decomposing or merging factors to arrive at a [[common ground]]. But once the common ground is there, a shared understanding of the topic at hand arises. This shared understanding may be located in a specific corner of the interconnected causal map. For interdisciplinary projects, it makes sense to then focus on this specific corner and to: 1. Devise a research question that could further elucidate the common ground portion of the causal map and design an appropriate experiment 2. Provide a problem statement or research question that can be resolved through a review of the literature as it pertains to the common ground portion of the causal map 3. Search for more literature to close remaining gaps in the common ground portion of the causal map Again, it depends on the specific project type which action is most appropriate. For a more detailed description, see: [[Interdisciplinary causal maps]]. ## Keeping it simple The causal map is tool to reason about your research and to guide you in your decision-making. Being *exhaustive* often means overcomplicating the map, so that it no longer provides you with an overview of relevant research and relevant relations at a glance. In our experience, many students are completionists when drafting causal maps and can self-sabotage this way, putting a lot of effort in creating an overview that is then no longer very useful. Like all maps, causal maps should only list those features that guide you in your research process. There are no hard rules for choosing these features, as your choices should adapt to the problem you are trying to solve.