Observation Oriented Modelling was proposed to overcome some of the problems in the application of statistical inference methods in behavioural sciences. We refine one part of this approach and show connections to methods well known in statistical learning. Precisely, we argue that the Moore-Penrose-Inverse is superior from an statistical point of view compared to the initial solution. We than show that Observation Oriented Modelling can indeed be appropriate for some of the tasks in the analysis of observed data. For a practical example, the revised method is demonstrated by analysing the effect of mindfulness training on attentional processes.