Collective decision making is a process observed both in natural and artificial swarms, in which agents converge and select a single option without the influence of leading individuals. In this study, we look at a collective decision making process in a group of C. elegans, a model organism in biology, characterised by a fully mapped neural system of 302 neurons, and a pheromone-based system that regulates the interactions among the worms. In particular, we model and optimise swarms of C. elegans through sets of differential equations. The experimental setup compares the performances of two different types of swarms: one in which the agent interactions are modulated by self-released pheromone production, and the other in which this means of interaction is absent. The worms’ task is to choose to selectively swarm towards one of the two target locations containing an attractive odour. We analyse and compare the swarming behaviour and the collective decision making process of the agents in the two models. Our findings suggest that the presence of pheromones is of paramount importance for the capabilities of the swarm to reliably choose one of the two attractive targets.