Crowd steering force nodes, and the default forces on the solver, each have weights. At any given moment, an agent can potentially be affected by multiple forces (for example, both "follow this path" and "avoid obstacles"). The crowd solver deals with multiple forces by normalizing the weights (scaling them so they add up to 1) before multiplying the individual forces by them.

Imagine an agent has the following potential forces that might apply to it at a given moment, with these weights:

1. The solver normalizes the weights so they add up to 1, making the avoid weight 0.59, the follow weight 0.29, and the align weight 0.12.

2. The solver then multiplies the force amounts by the normalized weights, so the avoid force is 10 * 0.59 = 5.9, the follow force is 1 * 0.29 = 0.29, and the align force is 2 * 0.2 = 0.4.

3. Finally, the solver adds the forces to get the overall force on the agent.

All forces in the agent’s current state (and any forces that apply to all agents regardless of state) are applied to the agent. It’s not possible to have a behavior that completely overrides any other behavior. However, you can use a very high weight (for example, 10), so that when other forces are normalized, they are made unnoticeably small.

You can use regular particle force nodes and other DOPs inside a state to define behavior, not just crowd-specific "steering" nodes. However, non-crowd-specific nodes don’t have Weight parameters, so they will not be subject to the normalization described above. Instead they will simply be added on to any other forces on the agents.