A new meta-heuristic method, based on Neuronal Communication (NC), is introduced in this article. The neuronal communication illustrates how data is exchanged between neurons in neural system. Actually, this pattern works efficiently in the nature. The present paper shows it is the same to find the global minimum. In addition, since few numbers of neurons participate in each step of the method, the cost of calculation is less than the other comparable meta-heuristic methods. Besides, gradient calculation and a continuous domain are not necessary for the process of the algorithm. In this article, some new weighting functions are introduced to improve the convergence of the algorithm. In the end, various benchmark functions and engineering problems are examined and the results are illustrated to show the capability, efficiency of the method. It is valuable to note that the average number of iterations for fifty independent runs of functions have been decreased by using Neuronal Communication algorithm in comparison to a majority of methods.