During this four intense days, we discussed all the basic principles of SMS, and one that really got my attention was the practical drift. And that attention came because of late events that, in my honest opinion, show us that sometimes our operational performance is really away from the planned baseline performance.
Psychology can easily explain this because, as humans, our behavior is dynamic and subject to changes, specially when our relations with other humans or the environment is not always the best. In that scenario, we start blaming everything on others, and became "the perfect victim".

This is halfway point between the SHELL model and James Reason's "swiss cheese" model, and that interaction of me as human vs other human, environment, technology or regulation/procedures can trigger a latent condition as result of an error or unsafe action.
SMS states that in order to navigate the drift in a safe manner, the system needs to became predictive, but to reach that goal we have to be compromised with safety and help the system to grow. To archive that you have to be in charge of your emotional and mental status, because any situation that affects your performance and deviates the operation from optimal must be immediately managed, and then needs to be reported, specially when an error is identified as result of that particular situation.
This is my opinion in how we can help the system to became predictive, allowing it to collect data and create a realistic and strong database that will help analysts in the creative process of managing future situations through the implementation of strong defenses.
And is no secret that only through a strong voluntary report system and the help of a efficient data collection process we will be able to avoid drifting too far.
Safe flights!