In a pair (2-item) scheme, how do you decode "Einstein juggling"?

Person → item 1, Action → item 2 — two items, no object slot to read.

With a Person-Action pair system, a composite has only two slots. "Einstein (person) juggling (action)" decodes as: item 1 = whoever's person is Einstein, item 2 = whoever's action is juggling. There's no object to interpret. This is why pair schemes are simpler to decode — one fewer slot per image — at the cost of storing only two items per picture instead of three.

Go deeper:

• Dominic system (Wikipedia) — the Person+Action pair model this decode uses.