I do not use the PAO, however, I am interested in the following question from a theoretical standpoint. For those who use a PAO, do you still use PAO for the Spoken Numbers discipline, or do you modify to some extent such as only an object per locus, or Person/Action per locus, or some other type of combination? The reason I ask this question is because I have observed that many memory athletes with non-PAO systems tend to change their approach to Spoken Numbers. For example, one who typically puts 3 images per locus uses only 2 images per locus or in the case of Johannes Mallow who uses 2 images per locus puts 1 image per locus for Spoken Numbers. I am curious to know is there some insight into this pattern of a one image reduction? I would like to know the cause of this phenomenon and why people prefer to modify their method for this discipline? Do you all have any thoughts or theories? The reason I ask this question is because I have only made this observation from those who do not use PAO and I am curious to know from another perspective.