Figure 2. A Mobile Cellular Automata (MCA) model was used to qualitatively explore the behaviour of groups of automata under interactions. The model has been previously used in the study of ants and here it is used to explore complex dynamics in termites. Two sets of simulations were performed in order to measure, as a function of the group size or lattice density, the behaviour of spontaneous and induced activations and the KS-entropy. In (a) and (d), there is an exponential decay in the number of spontaneous activations measured for two different values of the gain parameter g (0.005 and 0.08 respectively). This behaviour is explained by the fact that, as the density is increased, less individuals are locally isolated and then are more engaged in social contacts. In (b) and (e), it is possible to observe the qualitative differences introduced when the gain parameter is modified. When the value is low, the induced activations grow logarithmically towards a maximum along the increase of the lattice density. In contrast (e) shows the case when g has a relatively large value. The impact of this is noticed by the maximum of induced activations occurring at a certain density value. Finally, in (c) and (d) the KS-entropy was measured and it is observed that a maximum may occur for a combination of values of g. This maximum is important because it signals the presence of a percolation-like phenomena separating a disordered from an ordered phase. In the first region the automata activates in an uncorrelated way, while in the second phase, temporal oscillations develop. No phase transition was detected in our experiment with Cornitermes cumulans. Lattice size = 10². Activation probability = 0.01. each point is the average of 100 replicas.