Morpho-orthographic segmentation without semantics

Elisabeth Beyersmann (1) , Johannes Ziegler (1) , Anne Castles (2) , Max Coltheart (2) , Yvette Kezilas (2) , Jonathan Grainger (1)

1 LPC - Laboratoire de psychologie cognitive
2 Macquarie University

Masked priming studies have repeatedly provided evidence for a form-based morpho-orthographic segmentation mechanism that blindly decomposes any word with the mere appearance of morphological complexity (e.g., corn + er). This account has been called into question by Baayen et al. Psychological Review, 118, 438–482 (2011), who pointed out that the prime words previously tested in the morpho-orthographic condition vary in the extent to which the suffix conveys regular meaning. In the present study, we investigated whether evidence for morpho-orthographic segmentation can be obtained with a set of tightly controlled prime words that are entirely semantically opaque. Using a visual lexical decision task, we compared priming from truly suffixed primes (hunter-HUNT), completely opaque pseudo-suffixed primes (corner-CORN), and non-suffixed primes (cashew-CASH). The results show comparable magnitudes of priming for the truly suffixed and pseudo-suffixed primes, and no priming from non-suffixed primes, and therefore provide further important evidence in support of morpho-orthographic segmentation processes operating in the absence of any possible role for semantics.

Mots clés: Morphological processing; Morpho-orthographic priming; Pseudo-affixes; Masked priming

Domaines: Sciences cognitives; Psychologie; Neurosciences; Linguistique

Elisabeth Beyersmann, Johannes Ziegler, Anne Castles, Max Coltheart, Yvette Kezilas, et al.. Morpho-orthographic segmentation without semantics. Psychonomic Bulletin and Review, 2016, 23 (2), pp.533-539. ⟨10.3758/s13423-015-0927-z⟩. ⟨hal-02087747⟩