 Hi, my name is Lukas, let me tell you about registered reports. Here is a general outline of a research process. You start with an idea, design your study, get the data, write everything up and finally you submit your manuscript to a journal in order to get it published. Sounds pretty straightforward, but if you went through this a few times, you will notice that something is missing. It becomes peer review. I forget because this is a lightning talk, there should be lightning somewhere. So publishing can be tough. Journals get lots of submissions and they have to choose between them. They need some sort of selection criteria. And how clear the results are has unfortunately become one of these selection criteria. Consequently, researchers are incentivized for tweaking their data, for dropping a condition or a variable and even for not reporting entire experiments. In turn, only half the truth gets published. This has led to many published findings not being replicable. We're talking about up to 50%, but this is another talk. For you, this means that you will have difficulties publishing your results unless they happen to be very unambiguous. If this is not the case, reviewers might ask you to conduct another study or even reject your paper. Registered reports try to solve this problem by eliminating the second point. They eliminate the possibility to evaluate research on the basis of the results. But how? For registered reports, peer review is split into two stages. In stage one, data has not yet been collected, and in stage two it has, but it's not about the results, but about whether everything went according to plan. Specifically, stage one peer review is about theory, hypothesis and methods. For example, is the study well designed to test the hypothesis? If so, the journal will issue an in-principle acceptance. That is, the paper will get published as long as the authors stick to what they have planned so far. In stage two, reviewers get the manuscript again, only that this time the results and discussion sections have been added. Often these are the same reviewers as in stage one. The reviewers in stage two must not criticize the methods as they have already been evaluated. Now their task is to check if the authors stick to their analysis plan. Here you see what a manuscript outline looks like at stage one and stage two. In stage one, it simply stops before the results. Make sure though that everything that you will do to arrive at the results should be specified before you actually have the data, just like in a pre-registration. I listed it here as the analysis plan. It contains details on how you compute variables, what assumptions you test, which participants you exclude, and which statistical tests you run. Registered reports work for all kinds of hypothesis testing studies. Here is an example from MetaScience, where a group of researchers has submitted a meta-analysis as a registered report. They have even already written large parts of the results section with placeholders for the actual values that anybody can take and modify for their own research. If you choose to submit your paper as a registered report, problems in the study design get detected before you spend all the money and time collecting the data. What is published is more trustworthy than research that is published in the traditional way as there are no incentives for questionable research practices here. And early research already indicates that the quality of registered reports is higher. This has been a quick overview over registered reports. You can find everything else that you need to know on the COS website linked here. There are recommendations for preparing your manuscript, lists of journals that offer registered reports, resources for editors and much more. I hope that you will consider this publication format for your next project.