“…Following the original IA framework, more recent models of word recognition based on cascaded processing have incorporated the word-letter interactivity assumption (e.g., Coltheart, Curtis, Atkins, & Haller, 1993;Grainger & Jacobs, 1996). However, the above account of the word superiority effect and other related findings has been questioned on both empirical and simulation grounds (Allen, Wallace, & Weber, 1995;Jacobs & Grainger, 1992;Mewhort & Johns, 1988;Paap, Newsome, McDonald, & Schvaneveldt, 1982). Hence, it remains unclear whether reading really involves interactive processes allowing activation at a higher processing level to reverberate and influence activation at a lower processing level.…”