To investigate distributed synaptic plasticity at the cell assembly level, we used dissociated cortical networks from embryonic rats grown on grids of 60 extracellular substrate-embedded electrodes (micro-electrode arrays). We developed a set of experimental plasticity protocols based on the pairing of tetanic bursts (20 Hz) with low-frequency stimuli (< or = 1 Hz), delivered through two separate channels of the array (i.e. associative tetanic stimulation). We tested our protocols on a large data set of 26 stable cultures, selected on the basis of both their initial level of spontaneous firing and the capability of low-frequency test stimuli to evoke spikes. Our main results are summarized as follows: (i) low-frequency stimuli produce neither short- nor long-term changes in the evoked response of the network; (ii) associative tetanic stimulation is able to induce plasticity in terms of a significant increase or decrease of the evoked activity in the whole network; (iii) the amount of change (i.e. increase or decrease of the evoked firing) strongly depends on the specific features of the applied protocols; and (iv) the potentiation induced by a specific associative protocol can last several hours. The results obtained demonstrate that large in vitro cortical assemblies display long-term network potentiation, a mechanism considered to be involved in the memory formation at cellular level. This pilot study could represent a relevant step towards understanding plastic properties at the neuronal population level.