Procedural and declarative knowledge of word recognition and letter decoding in reading an artificial script.

In a previous study [Cogn. Brain Res. 16 (2003) 325], we found that letter knowledge did not evolve from implicit training on whole-word recognition in an artificial Morse-like script, although the participants were adults, experienced in alphabetical reading. Here we show minimal conditions in which letter knowledge may evolve in some individuals from training on whole-word recognition. Participants received multi-session training in reading nonsense words, written in an artificial script, in which each phoneme was represented by two discrete symbols. Three training conditions were compared: alphabetical whole words with letter decoding instruction (Explicit), alphabetical whole words (Implicit), and non-alphabetical whole words (Arbitrary). Subjects were assigned to training either on the explicit and arbitrary or on the implicit and arbitrary conditions. Our results show that: (a) Letter-decoding knowledge evolved implicitly from training on alphabetical whole-word recognition, in some individuals. However, (b) a clear double dissociation was found between effectively applied implicit letter knowledge and declarative letter knowledge. (c) There was no advantage of the implicitly derived over the explicitly instructed letter knowledge. (d) Long-term retention was more effective in the explicit compared to the arbitrary condition. (e) Word-specific recognition contributed significantly to performance in all three training conditions, i.e. even under conditions that presumably afford advantage for word segmentation. Altogether, our results suggest that both declarative and procedural knowledge contributed to letter decoding as well as to word-specific recognition performance. Moreover, a greater dependency on declarative knowledge may not be an inherent characteristic of word-specific recognition, but rather that both letter decoding and word-recognition routines can become proceduralized given sufficient practice.