I happened to know about Kitty Genovese case while studying the opinion dynamics around 2005, which has been a subject for physicists for decades. I dug into the literature to find out a chain of papers, including psychologist Bibb Latane& #39;s PRA paper. 1/n https://twitter.com/AliceDreger/status/1301331300842864644">https://twitter.com/AliceDreg...

The title is "Statistical mechanics of social impact" (see https://journals.aps.org/pra/abstract/10.1103/PhysRevA.45.763),">https://journals.aps.org/pra/abstr... from which I found and read Latane& #39;s papers as well as documents about Kitty Genovese case. Based on the experimental results by him and his colleagues, I devised a network evolution model. 2/n

I studied our model (originally the name was "rescue model" and later replaced by "coevolving helping network model") to relate the "microscopic" individual behaviors to the "macroscopic" results by field study. See my papers: https://iopscience.iop.org/article/10.1209/epl/i2005-10378-5/meta">https://iopscience.iop.org/article/1... https://journals.aps.org/pre/abstract/10.1103/PhysRevE.74.026120">https://journals.aps.org/pre/abstr... 3/n

One more paper is about the phase transitions in our model: https://journals.aps.org/pre/abstract/10.1103/PhysRevE.81.066108.">https://journals.aps.org/pre/abstr... It took several years to complete the research as I& #39;ve been involved in other projects at the same time. But it was a very fruitful experience to me. 4/n

At that time, the opinion dynamics models at hand were too simplistic in my opinion, which is fine when benefiting from such simplification. However, the simpler models, the farther they seem to be from reality. I tried to bridge the gap between them (kind of). 5/n

For this, several assumptions for the model were based on the empirical findings, yet the model could be solvable to some extent to give a more rigorous insight into the problem. This experience led me to think about how successful the so-called social physics can be. 6/n

Hope to continue this thread in a future when I have more time. 7/7