Prefrontal cortex is known to be involved in many high-level cognitive functions, in particular working memory. Here, we study to what extent a group of randomly connected units can store and maintain (as output) an arbitrary real value from a streamed input, i.e. how such system act as a sustained working memory module without being distracted by the input stream. Furthermore, we explore to what extent such an architecture can take advantage of the stored value in order to produce non-linear computations. Systematic comparison between different architectures (with and without feedback, with and without a working memory unit) shows that explicit memory is required. With Principal Component Analyses (PCA) we show that the reservoir state is encoding time and the memorized value in different ways depending if a supplementary task is required. Moreover, theses memory states are similar to attractors in an input-driven system [3], and in particular, similar to a noisy line attractor [6]. In this study, we did not try to find the optimal number of reservoir units needed for each task. Conversely, we voluntary limited the size of the reservoir to 100 neurons in order to see if such rather small reservoirs were sufficiently competitive.