Abstract
The present paper evaluated the nature of the organization of 22 structural measures of object pictures from the Snodgrass and Vanderwart (1980) picture set (Study 1), and their contribution to object decision and to object naming latencies (Studies 2 and 3). Study 1 employed a principal components analysis and provided evidence of four underlying components: “Object parts”, “internal details”, “object contours”, and “variability of the representation”. Study 2 examined the contribution of these components to object decision and object naming and highlighted variability of the representation and internal details as the most relevant indexes of structural similarity. Study 3 investigated the interactions between these structural components and lexical frequency. Main results showed an interaction effect between variability of the representation and lexical frequency and other effects associated to internal details. Implications for the concept of structural similarity and for object recognition are discussed from a continuous and cascade processing perspective.
Acknowledgements
This research was supported by the Centro de Investigação em Psicologia da Universidade de Lisboa. We thank Glyn Humphreys, Keith Laws, and Toby Lloyd-Jones for generously providing us with the full data of their structural measures. We are also grateful to Keith Laws, Jorge Almeida, and Christine Cannard for their comments to a previous version of the manuscript.
Notes
1References in this section refer to the study that first proposed the measure.