The comparison might be emotional unacceptable at a time like this, but I think its worth using all that studies that exist on human violence in order to unders

The comparison might be emotional unacceptable at a time like this, but I think its worth using all that studies that exist on human violence in order to understand human behavior. Predicts markers in this topic, ie traits of people that allow someone looking at a set of data to make a prediction about who end up killed by the police and who doesn't, would not list racism at the top. If I attempt an simplified meta analyze on the research, I would list at the top gender which the human brain determine at an exceptional speed. Second are social economic status symbols and signals. After that and in third place I would place in-group association with either minority or majority groups, ie the result of racism, homophobia, xenophobia and so on. Prediction models don't have a narratives. You get a set of data and after that you should be able to make risk profiles that match future data sets. If given 1000 people in a given area, a prediction model would pull out risk profiles of who are more likely to end up being killed by the police. Statistically over time the risk profiles would then match the outcomes. Those with lowest risk profile of getting shot by the police would, if we follow the above ordering: female, high status, and part of a majority group. Highest risk would be male, low status, and part of a minority group. What we want to do with such model is a separate issue, including if we want to apply it to the current events. I would note that gender, social economic status and in-group association all effects how the brain perceive threats and get amplified in threatening situations.