“…When system (1.2) reduces to the one-dimensional Cohen-Grossberg neural networks, our result on global exponential stability is different from the existing results. In our result, the hypothesis for differentiability in [7,8,22,25] on the behaved functions is removed and the hypotheses for boundedness in [2, 3, 8-12, 16, 18, 21-25] and monotonicity in [7,15,22,23,25,26] on the activation functions are removed. We just require that the behaved functions satisfy two inequalities and activation functions are globally Lipschitz continuous.…”