Search

In one’s native language, visual word identification is based on early morphological analysis and is sensitive to the statistical structure of the mapping between form and meaning (Orthography–to–Semantic Consistency, OSC). How these mechanisms apply to a second language is much less clear. We recruited L1 Italian–L2 English speakers for a masked priming task where the relationship between prime and target was morphologically transparent, e.g., employer–EMPLOY, morphologically opaque, e.g., corner-CORN, or merely orthographic, e.g., brothel–BROTH. Critically, participants underwent thorough testing of their lexical, morphological, phonological, spelling, and semantic proficiency in their second language. By exploring a wide spectrum of L2 proficiency, we showed that this factor critically qualifies L2 priming. Genuine morphological facilitation only arises as proficiency grows, while orthographic priming shrinks as L2 competence increases. OSC was also found to modulate priming and interact with proficiency, providing an alternative way of describing the transparency continuum in derivational morphology. Overall, these data illustrate the trajectory towards a fully consolidated L2 lexicon and show that masked priming and sensitivity to OSC are key trackers of this process.

Research strongly suggests that printed words are recognized in terms of their constituent morphemes, but researchers have tended to consider the recognition of derivations and inflections in separate theoretical debates. Recently, Crepaldi et al. (2010) proposed a theory that claims to account for the recognition of both derivations and inflections. We investigated brain potentials in the context of masked priming to test 2 key predictions of this theory: (a) that regular inflections should prime their stems to a greater degree than irregular inflections should prime their stems and (b) that priming for regular inflections should arise earlier in the recognition process than priming for irregular inflections. Significant masked priming effects were observed for both regular and irregular inflections, though these effects were greater for regular inflections. ERP data further suggested that masked priming effects for regular and irregular inflections had different time courses. Priming for regular but not irregular inflections emerged in a time window reflecting processing up to 250 ms post target onset, and although priming for regular and irregular inflections was observed in a time window reflecting processing 400 to 600 ms post target onset, these effects arose earlier and were of greater magnitude for the regular inflections. These findings support a form-then-meaning characterization of the visual word processing system such as that proposed by Crepaldi et al. (2010) and raise challenges for alternative approaches.

It is not clear from previous research (e.g., Caramazza et al., 1988; Deutsch et al., 1998; Frost et al., 1997) whether morphological stems that sub-serve the formation of both nouns and verbs (e.g., 'deal-') have a unique, grammatical class-independent representation in the visual word identification system, or rather feature two separate, grammatical-class specific representations. In a first experiment, participants were asked to read aloud nouns and verbs that were anticipated by morphologically-related primes belonging to the opposite grammatical class (e.g., partenza-PARTIRE, departure-TO LEAVE). In order to disambiguate genuine morphological priming from semantic facilitation, the same target words were also paired in a second condition with semantically related, but morphologically unrelated primes (e.g., viaggio-PARTIRE, trip-TO LEAVE). Morphological and semantic primes were contrasted with separate sets of control primes. The results showed reliable cross-class morphological priming. This effect was also shown to be independent from whether nouns primed verbs or vice versa, and from SOA (100 ms vs. 300 ms). In a second experiment, cross-class morphological priming was shown to emerge even when the related primes were compared with control words that shared their orthographic and phonological onset (e.g., abbraccio-ABBRACCIARE, (the) hug-to hug vs. abbazia-ABBRACCIARE, abbey-to hug), thus proving to hold independently of orthography and phonology.

New evidence has emerged in recent years suggesting that the recognition of morphologically‑complex words starts with a rapid morphemic segmentation that is orthographically based (Rastle & Davis, 2008). This evidence appears inconsistent with evidence showing that masked priming of irregular inflections (e.g., stole‑STEAL) does not differ from masked identity priming (Forster et al., 1987) or masked priming of regular inflections (e.g., walked-WALK; Meunier et al., 2004). We carried out a masked priming experiment (SOA=42 ms) with a lexical decision task comparing the facilitation triggered by (i) irregular inflected primes (e.g., dug‑DIG), (ii) orthographically‑matched, but morphologically unrelated primes (dog‑DIG) and (iii) completely unrelated primes (pop‑DIG). Results showed that irregular inflections facilitate recognition of their stems significantly more than orthographically matched primes. These data are at odds with the existence of a pre‑lexical, orthographically‑based moprhological stage, unless very early feedback from central, semantically- and syntactically‑based stages of morphological processing is hypothesized.

New evidence has emerged in recent years suggesting that the recognition of morphologically‑complex words starts with a rapid morphemic segmentation that is orthographically based (Rastle & Davis, 2008). This evidence appears inconsistent with data showing masked priming of irregular inflections over their base forms (e.g., fell‑FALL) as compared to orthographically matched control words (fill‑FALL; Pastizzo & Feldman, 2002), suggesting the existence of morphological bounds independent of orthography at the earliest stage of printed word recognition. However, these data are not completely consistent and have also been challenged by Voga & Grainger (2004) and Kielar et al. (2008). We carried out a masked priming experiment (SOA=42 ms) with a lexical decision task comparing the facilitation triggered by (i) irregular inflected primes (e.g., dug‑DIG), (ii) orthographically‑matched, but morphologically unrelated primes (dog‑DIG) and (iii) completely unrelated primes (pop‑DIG). Results showed that irregular inflections facilitate recognition of their stems significantly more than orthographically matched (t36=2.80, p < .01) and unrelated primes (t36=3.33, p < .005; see Table 1). As several irregular present‑past pairs form isles of sub‑regularity (e.g., bend‑bent, lend‑lent, send‑sent), this effect might generalize to unrelated words showing the same orthographic pattern (tend‑tent). This has been tested in Experiment 2, which was completely identical to Experiment 1 except for the fact that pseudo‑present‑past pairs were used (e.g., bake‑book, ray‑raid and peak‑poke). Results showed no effect whatsoever in this condition (F2,72=.24, p=.78; see Table 1). Across‑experiment ANOVAs were also carried out to check for interaction, which was significant by subjects (F2,152=4.02, p=.02) and approached significance by items (F2,144=2.59, p=.08), showing that irregular masked priming effect is truly morphological in nature. These data are at odds with the existence of a pre‑lexical, orthographically‑based morphological stage, unless very early feedback from central, semantically- and syntactically‑based stages of morphological processing is hypothesized.

Forster et al. (1987) and Meunier et al. (2004) investigated irregular morphological priming with short SOA; they both report equal facilitation when a stem is primed by itself (or a regular inflected word) and by an irregular related word. However, both these studies used only one set of control words, making impossible a perfect prime-control matching in both repetition/regular and irregular conditions. We replicated these lexical decision studies using similar SOA (40ms) and two different control word sets, each of which was matched for length, frequency, number of orthographic neighbours and orthographic overlap with either the repetition or the irregular prime list. The results show that there is a repetition priming effect, there is a borderline irregular priming effect, and this latter is smaller than the repetition effect, contrary to the previous findings. The consequence of these results for the models of complex word recognition will be discussed.