My current hypothesis is that spending time on a 1352 image system is not effective unless highly dedicated.
To most effectively use my time, I will focus on a PO. I believe 20 minutes here and there on a PO will outperform the equivalent time on the shadow system.
To build an exit plan if I decide later to try SS, I will not use people or images that would be preferable in a 3 phoneme word.
For instance 00 will be SeeSaw, not Scissor. This is because Scissor is a good image for 004 while SS can only be 2 digits.
If I expand later, my 100 PO images will not make the creation of a SS more difficult if I stick to these rules.
On the topic of not using action:
After speaking to some athletes who have very much stressed the importance of “linking” one image to the next, I will omit actions, instead only using concrete objects and people, while using “actions” to be the glue linking one to the next. While they did not recommend this, I see no downside and I’m fine having less images as I have never trained enough to feel like I have a shortage. 200 images is fine for now instead of 300.
I have also elected to use the major system for every person and object, not semantically related P-A-O groups.
I believe there is potential that being able to “read” the numbers instead of trying to remember a semantic object related to a person will be quicker. The logic behind that is the importance of “reading” the numbers and cards in the SS as it has been stressed to me by some athletes I’ve spoken to.
For example, 05:
Person: Saul (Goodman, Breaking Bad)
Object: Cell (phone)