I feel like the dual route combined model of reading breaks down when you have no phonological form for the characters

And I want to study this, but I just don't have the tools to recruit participants, nor get my experimental design vetted