The plasticity of Biological Neural Networks (BNNs) is mainly based on the complex interactions of neuromodulators, neurons, and synapses. These properties are thought to be the most important pre-requesites for learning in biological systems. Consequently, in the last years the concept of (artificial) neuromodulators influencing and changing the parameters of an artificial neural network (ANN) has been proposed by a few researchers. The plasticity of an ANN augmented by neuromodulators employed to control a mobile autonomos robot could be the key to "life-time" learning including a robot's adaptation to an unknown environment. We investigate the potential benefits of a neuromodulator controller on the well-known pole balancing problem by evolving an ANNs structure, weights, neuromodulator emission, reception, and reaction.