Phantom word activation in L2

Abstract

L2 listening can involve the phantom activation of words which are not actually in the input. All spoken-word recognition involves multiple concurrent activation of word candidates, with selection of the correct words achieved by a process of competition between them. L2 listening involves more such activation than L1 listening, and we report two studies illustrating this. First, in a lexical decision study, L2 listeners accepted (but L1 listeners did not accept) spoken non-words such as groof or flide as real English words. Second, a priming study demonstrated that the same spoken non-words made recognition of the real words groove, flight easier for L2 (but not L1) listeners, suggesting that, for the L2 listeners only, these real words had been activated by the spoken non-word input. We propose that further understanding of the activation and competition process in L2 lexical processing could lead to new understanding of L2 listening difficulty.

Spoken-word recognition in foreign-accented speech by L2 listeners

“…For Dutch listeners in Experiment 1a, Dutchaccented /ekt/ and /mu:f/ were perceptually closer to the correct English pronunciation than Japanese-accented primes were. Note however, that Dutch listeners are less likely to confuse English /e/ with /ae/ than they are to confuse /ae/ with /e/ , and they can categorize English final voicing contrasts as accurately as native English listeners can (Broersma, 2005(Broersma, , 2008(Broersma, , 2010; they just seem to be less good at using this information for online word recognition (e.g., Broersma & Cutler, 2008;Cutler et al, 2006;). While we do not want to claim that perceptual difficulties did not modulate the results in Experiment 1a, it is simply not possible to tease apart the influence of perceptual confusions from experiential effects for Dutch-accented primes and Dutch listeners.…”

The influence of native-language tones on lexical access in the second language

“…For bilingual speakers, experience distinguishing words based on acoustic features in one language can result in changes to how that feature is perceived or learned in another language (Dupoux et al, 2010;Mora and Nadeu, 2012). For instance, L1 speakers of Catalan who have extensive L2 Spanish experience are less accurate at distinguishing /e/ and /E/, a phonetic contrast that is important for Catalan, but which does not exist in Spanish (Broersma and Cutler, 2008). Furthermore, differences in the ability to perceive phonetic distinctions that are driven by experience with multiple languages, as in the Catalan-Spanish example, can impact listeners' ability to recognize or discriminate lexical items (Broersma and Cutler, 2008;Cutler et al, 2006;Mora and Nadeu, 2012;Pallier et al, 2001).…”

Resolving ambiguity in familiar and unfamiliar casual speech

“…Thus Japanese listeners' goodness ratings on tokens of English /r/ and /l/ pattern differently for the two categories (Iverson et al, 2003), but at the lexical level this difference is not exploited to enable, for instance, correct early choice between rocket versus locker in an eyetracking task (Cutler et al, 2006). The same pattern appears with Dutch listeners' processing both of [ae] versus [e] (vowels that divide a single Dutch category) and of word-final voicing distinctions (which are neutralized in Dutch); in both cases Dutch listeners perform quite well in a low-level choice task (Broersma, 2005), but fail to distinguish lexical minimal pairs such as cattle-kettle or roberope in cross-modal priming (Broersma & Cutler, 2008, 2011 or in eyetracking (Escudero, Hayes-Harb, & Mitterer, 2008;Weber & Cutler, 2004). In fact one of the most wellknown L2 effects, the disproportionate difficulty of listening to an L2 in a noisy environment, shows the same pattern; a review of four decades of literature on this topic (Lecumberri, Cooke, & Cutler, 2010) motivated the conclusion that noise impinges upon the initial uptake of speech by L1 and L2 listeners to an equivalent degree, but L1 listeners recover better from its effects.…”

“…Shoemaker (2010) used two-alternative forced choice, and Darcy et al (2007) asked listeners to decide whether a word in a sentence was the same as a prior pronunciation of the word in isolation (probe detection that was effectively same-different judgement). As Broersma (2005;Broersma & Cutler, 2008, 2011 showed with phoneme contrasts, low-level tasks may reveal discrimination that does not carry through to lexical processing. We do not in fact know whether Darcy et al's participants would continue to demonstrate native-level performance in cross-modal priming or eyetracking.…”

“…For any listener, spoken-word recognition involves multiple activation of candidate words, with the simultaneously activated forms competing for recognition (for a review see, e.g., McQueen, 2007). Although L2 listeners typically have smaller vocabularies than native listeners, listening to L2 actually involves more such activation and competition between word forms than listening to L1, as studies with the classic word recognition methods of cross-modal priming and eye-tracking have demonstrated (Broersma & Cutler, 2008, 2011Weber & Cutler, 2004). The increased competition can even persist despite decades of experience with a non-dominant L2 from childhood onwards (Pallier, Bosch, & Sebastian-Gallés, 1997;Sebastián-Gallés, Rodríguez-Fornells, Diego-Balaguer, & Diaz, 2006).…”