Cascaded processing in written naming: Evidence from the picture–picture interference paradigm

Abstract

The issue of how information flows within the lexical system in written naming was investigated in five experiments. In Experiment 1, participants named target pictures that were accompanied by context pictures having phonologically and orthographically related or unrelated names (e.g., a picture of a “ball” superimposed on a picture of a “bed”). In Experiment 2, the related condition consisted of target and context pictures that shared the initial letter but not the initial sound. In both experiments, a facilitatory effect of relatedness was observed on the latencies. Experiment 3 tested whether phonology contributes to the facilitation effect found in the written latencies in Experiment 2 by using target and context pictures that shared the initial phoneme but not the initial grapheme. This experiment did not reveal any reliable difference between the related and unrelated conditions. In Experiments 4 and 5, control tasks were used to rule out a perceptual and conceptual account of the orthographic facilitation effect found in Experiment 2. The findings suggest that the recognition of an object leads to the activation of its name, and thus, that the activation within the lexical system in written-naming flows in a cascaded manner.

Keywords:

• Cascading
• Written naming
• Picture–picture interference

Acknowledgements

The authors thank Professor Michael Spivey and two anonymous reviewers for helpful comments on earlier versions of the article.

Notes

¹There has been some debate as to whether speech production entails the activation of a neutral level, called the lemma, which mediates between concepts and the lexical form of the words (e.g., lexemes), or whether concepts map directly onto lexemes (e.g., Caramazza 1997; Miozzo & Caramazza, 1997; Roelofs, Meyer, & Levelt, 1996). As far as written production is concerned, there has been no such debate but the lemma-lexeme distinction is not critical for the issue we address here.

²Since there were only about one third of the items in each condition which were tested in each block, a by-item analysis was not feasible.

³Stadthagen-Gonzalez et al. (2009) argued that no-match latencies might also reflects the use of a deadline response criterion for negative responses.