Critical points of random fields.

The variation of the mean number of critical points as a function of the index is first studied using random matrices tools. In a second part, we study attraction or repulsion between these points again as a function of index. A measure is the correlation function
Our results extend the results of Belyaev, Cammarota and Wigman (2017) to dimension greater than 2, and to general isotropic Gaussian random fields, By specifying the indexes, we shows that the attraction between critical points that occurs when the dimension is greater than 2 is due to attraction between critical points with adjacent indexes.