Permanence for a class of nonlinear difference systems (Q2491487)

The following system of nonlinear difference equations \[ x_{n+1} = \lambda x_{n} + f(\alpha_{1}y_{n} - \beta_{1}y_{n-1}) \] \[ y_{n+1} = \lambda y_{n} + f(\alpha_{2}x_{n}- \beta_{2}x_{n-1}) \] with \(\lambda \in (0,1)\) and \(\alpha_{i},\beta_{i}\) positive is considered. It is a discrete analog of a system describing a neural network of two neurons with dynamical threshold effects. The main result is that if \(f\) is continuous, nondecreasing (resp. non increasing) and bounded from below (resp. above), if \(f(\alpha_{i}x) \leq\) (resp. \(\geq)\) \(\delta_{i}x\) for large positive (resp. negative) \(x\) and some \(\delta_{i} \in (0,1),\) if \(\alpha_{i}\lambda - \beta_{i} \leq 0,\) and if \(\alpha_{i} \geq \beta_{i}\) \((i = 1,2),\) then the system above is permanent.