Vinnayakk Bangarwaa wrote their blog post about certain bots that use smells, particularly pheromones, to carry out their basic functions. With my background in AI this reminded me immediately of Ant Colony Optimization (ACO), which is a pathfinding algorithm used for finding the shortest path from one node to another in a graph. I'm always amazed by these kinds of techniques, where a concept from nature that we barely need to alter can outperform almost any other algorithm! Similar algorithms exist based on genetics, evolution and swarm intelligence. Nature has it figured out really well sometimes. 

Anyway, while pheromone behaviour is incredibly interesting, it isn't really a human trait. It could perhaps be used as an underlying system to show group behaviour in humans? I think the way we follow trends could be portrayed in a very similar way to pheromone behaviour. When a trend emerges, a lot of people will follow it to the letter. This would be the parallel to exploiting a currently strong pheromone trail and reinforcing it. But, there are always some people who deviate from the path and explore new directions. This is exploration, making new trails that will usually amount to nothing; but sometimes be better in which case people will start following it, and thus reinforce that pheromone trail.

I'm not really sure how you would implement this, especially in a way that is achievable for this course, but there might be something there! And even if it's not, look up Ant Colony Optimization. It's fascinating stuff.