High shore littorinid snails exhibit complex spatial distribution patterns, forming dense aggregations both in crevices and on crevice-free sections of rocky shores. To understand how these patterns may form, an individual-based computer simulation of littorinid behaviour, similar to those used to model social insects, was used to mimic snail movement on rocky shores. Individual littorinid movement patterns were simulated, along with chance interactions with other littorinids, their trails, crevices and any resultant decisions made by the littorinids when these interactions occurred. The simulation investigated how simple behavioural rules can predict spatial aggregation patterns and the persistence and variation of these aggregation patterns over timescales of several tidal cycles. Morisita's index of dispersal showed good agreement between the simulation and observed patterns of littorinids on-shore. Trail-following was vital in the simulation, since points where trails intersected with crevices, or points where 2 or more trails met, formed the site of aggregations. Both in the simulation and in reality, aggregations often occurred in identical positions over several tidal cycles, both within and outside crevices. This temporal predictability may be due to the persistence of mucus trails on the shore over successive tidal cycles. Removal of the influence of past mucus trails from the simulation resulted in far lower persistence of aggregations over time. The simulation, therefore, is an important tool in examining behavioural mechanisms of intertidal animals. It provides insights into how simple behaviour of grazing animals can explain complex population patterns and subsequently the community dynamics of algal-herbivore interactions.