Understanding Bilingual Word Learning: The Role of Phonotactic Probability and Phonological Neighborhood Density Purpose Previous research has shown that the language-learning mechanism is affected by bilingualism resulting in a novel word learning advantage for bilingual speakers. However, less is known about the factors that might influence this advantage. This article reports an investigation of 2 factors: phonotactic probability and phonological neighborhood density. ... Research Article
Newly Published Free
Research Article  |   November 07, 2017
Understanding Bilingual Word Learning: The Role of Phonotactic Probability and Phonological Neighborhood Density
 
Author Affiliations & Notes
  • Vishnu KK Nair
    Discipline of Speech Pathology and Audiology, Flinders University, Adelaide, Australia
    ARC Centre of Excellence in Cognition and its Disorders, Department of Cognitive Science, Macquarie University, Sydney, Australia
  • Britta Biedermann
    ARC Centre of Excellence in Cognition and its Disorders, Department of Cognitive Science, Macquarie University, Sydney, Australia
    School of Psychology and Speech Pathology, Curtin University, Perth, Australia
  • Lyndsey Nickels
    ARC Centre of Excellence in Cognition and its Disorders, Department of Cognitive Science, Macquarie University, Sydney, Australia
  • Disclosure: The authors have declared that no competing interests existed at the time of publication.
    Disclosure: The authors have declared that no competing interests existed at the time of publication.×
  • Correspondence to Vishnu KK Nair, who is now only affiliated with Macquarie University, Sydney, Australia: vishnu.nair@mq.edu.au
  • Editor: Rhea Paul
    Editor: Rhea Paul×
  • Associate Editor: Ann Tyler
    Associate Editor: Ann Tyler×
Article Information
Development / Cultural & Linguistic Diversity / Attention, Memory & Executive Functions / Speech, Voice & Prosody / Newly Published / Research Article
Research Article   |   November 07, 2017
Understanding Bilingual Word Learning: The Role of Phonotactic Probability and Phonological Neighborhood Density
Journal of Speech, Language, and Hearing Research, Newly Published. doi:10.1044/2017_JSLHR-L-15-0376
History: Received October 30, 2015 , Revised May 14, 2016 , Accepted August 14, 2017
 
Journal of Speech, Language, and Hearing Research, Newly Published. doi:10.1044/2017_JSLHR-L-15-0376
History: Received October 30, 2015; Revised May 14, 2016; Accepted August 14, 2017

Purpose Previous research has shown that the language-learning mechanism is affected by bilingualism resulting in a novel word learning advantage for bilingual speakers. However, less is known about the factors that might influence this advantage. This article reports an investigation of 2 factors: phonotactic probability and phonological neighborhood density.

Method Acquisition of 15 novel words varying in phonotactic probability and phonological neighborhood density was examined in high-proficiency, early onset, Mandarin–English bilinguals and English monolinguals.

Results Both bilinguals and monolinguals demonstrated a significant effect of phonotactic probability and phonological neighborhood density. Novel word learning improved when the phonological neighborhood density was higher; in contrast, higher phonotactic probability resulted in worse learning. Although the bilingual speakers showed significantly better novel word learning than monolingual speakers, this did not interact with phonotactic probability and phonological neighborhood density manipulations.

Conclusion Both bilingual and monolingual word learning abilities are constrained by the same learning mechanisms. However, bilingual advantages may be underpinned by more effective allocation of cognitive resources due to their dual language experience.

A large body of research has found converging evidence for a positive relationship between bilingualism and nonlinguistic skills (e.g., Abutalebi et al., 2011; Bialystok, Craik, Klein, & Viswanathan, 2004; Bialystok, Craik, & Ryan, 2006; Costa, Hernandez, & Sebastián-Gallés, 2008). In contrast, studies examining the impact of bilingualism on linguistic skills have yielded mixed findings. For example, whereas Sheng, McGregor, and Marian (2006)  found a bilingual advantage in children for a lexical production task, Rogers, Lister, Febo, Besing, and Abrams (2006)  showed that bilingual adults have more difficulty in recognizing words in challenging (noisy) environments compared with monolinguals. It has also been reported that bilingual adults take longer to retrieve words from the mental lexicon (i.e., slower lexical access; e.g., Bialystok, 2009; Gollan, Montoya, Fennema-Notestine, & Morris, 2005; Ivanova & Costa, 2008) and may demonstrate a disadvantage in picture-naming tasks (Gollan et al., 2005). Despite these mixed reports, superior bilingual performance has been consistently found for linguistic tasks associated with language learning (e.g., Antoniou, Liang, Ettlinger, & Wong, 2015; Kaushanskaya, 2012; Kaushanskaya & Marian, 2008, 2009a, 2009b; Nair, Biedermann, & Nickels, 2016; Papagno & Vallar, 1995; Van Hell & Mahn, 1997). For instance, Kaushanskaya and Marian (2009b)  showed that early bilinguals were significantly better than monolinguals in learning novel words. Nair, Biedermann, and Nickels (2016)  found that this bilingual advantage in novel word learning remained present even in a population of late bilinguals with delayed onset of learning their second language (L2 speaking acquired after 12.45 years of age). Although evidence suggests a generally positive influence of bilingualism on word learning, the factors underpinning this bilingual effect remain unclear. Further investigation of the constraints on the effects of bilingualism on the language-learning mechanism will enable us not only to further specify how bilinguals differ from monolinguals but also to contribute to the development of theories explaining why bilinguals differ from monolinguals. A variety of participant-related factors have been found to influence the novel word learning advantage in bilinguals, such as age of L2 acquisition (Kaushanskaya & Marian, 2008; Nair et al., 2016), length of L2 exposure (Hernandez & Li, 2007), cognitive control (Bradley, King, & Hernandez, 2013) and, for developing bilinguals (L2 learners), the linguistic structures of their first language (L1; for theoretical accounts, see, e.g., Flege, 1987; Lado, 1957; Zobl, 1980). However, more recent investigations with monolingual speakers have shown that phonological and lexical properties of novel words, such as phonotactic probability (PP) and phonological neighborhood density (PND), may also influence word-learning outcomes (Storkel, Armbrüster, & Hogan, 2006). However, it is not known how these factors affect word learning in bilinguals. An investigation of whether (and how) these factors affect word learning differently in bilinguals and monolinguals would provide further insights into the changes bilingualism brings to cognitive processing.
PP refers to the likelihood of sounds and sound combinations occurring in a given language (Vitevitch & Luce, 2005). PND refers to the total number of words that sound similar to a given word (Luce & Pisoni, 1998). Although these two variables are different, they are related: Words with common phoneme sequences (high PP) tend to have dense neighborhoods, whereas words with rare sequences generally have sparse neighborhoods (e.g., Vitevitch, Luce, Pisoni, & Auer, 1999). Nevertheless, Storkel et al. (2006)  argued that PP and PND have distinct influences on word learning. They examined the independent effects of these two variables on novel word learning in the context of a story in monolingual adults. During the initial stages of word learning (i.e., analyzing partially correct responses), they found a disadvantage for words with high PP compared with words with lower PP. These effects of PP found during initial stages of word learning differed from developmental studies where an advantage is usually found with increasing PP (e.g., Storkel, 2001; but see also Storkel & Lee, 2011; Hoover, Storkel, & Hogan, 2010  for contrasting evidence). In contrast, the effect of PND was not significant: Initial learning performance did not vary between words of high and low density.
For the later stages of word learning (i.e., analyzing completely correct responses), Storkel et al. (2006)  found a contrasting pattern: no significant effect of PP but an advantage for words of high neighborhood density. Storkel et al. suggest that the advantage for low PP words may be due to these stimuli being less wordlike. For instance, these words could be more easily detected, with detection triggering differing processing for novel words (learning) and known words (lexical access).
In contrast, the advantage for words with high neighborhood density during the later stages of word learning was suggested to be due to these words activating more neighbors from a long-term memory. These neighbors may facilitate acquisition of novel words by strengthening representations through feedback from shared phonemes.
In sum, Storkel et al. (2006)  found that monolingual adults demonstrate a disadvantage for learning words with high PP but an advantage for words with high PND (albeit over different phases of learning). Nevertheless, it is clear that the effects of PP and PND on language processing in general are far from straightforward.
It seems that PND effects on speech production vary depending not only on the task but also across languages and populations studied (e.g., healthy speakers vs. individuals with aphasia; see Sadat, Martin, Costa, & Alario, 2014  for a detailed discussion). By extension, it is possible that monolingual and bilingual adults may differ in the effects of PND and PP on their word learning. Moreover, given that bilinguals show advantages for word learning compared with monolinguals, perhaps this advantage is influenced or modulated by the effects of PP and PND. It is also possible that the PP and phonological neighborhood effects are different in bilinguals compared with those of monolinguals given that these variables are known to be sensitive to different methodologies (e.g., word recognition, word learning, nonword repetition) and participant characteristics (e.g., language experience, language performance, age). For example, in monolingual adults, high PP is generally associated with facilitatory effects for word recognition (e.g., Vitevitch & Luce, 1999) and nonword repetition (e.g., Vitevitch & Luce, 2005). A high PND may produce an advantage for word production (e.g., Baus, Costa, & Carreiras, 2008; Vitevitch, 2002) and word learning (e.g., Storkel et al., 2006) but not for word recognition (e.g., Luce & Pisoni, 1998).
Bilinguals have extensive experience in learning unfamiliar words in an L2; it is possible that this experience generalizes to better learning of novel words regardless of their similarity to other words in their lexicon (i.e., even those words with lower PP and lower PND). In the context of L2 learning, two contrasting theories have also been developed specifically to account for learning familiar or unfamiliar phonemes/phoneme strings. For example, the contrastive analysis hypothesis suggests that L2 learning may be susceptible to the structural properties of L1 (Lado, 1957). L2 phonemic structures that closely resemble L1 structures may be easier to learn than L2 structures that are dissimilar to L1. These prior L1 knowledge effects on L2 learning have been proposed to be highly selective depending on the linguistic structure of the languages (Zobl, 1980). However, the speech learning model (Flege, 1987; Flege & Hammond, 1982) predicts that phonemes that are less familiar and unique in an L2 will be better learned than commonly occurring phonemes (irrespective of L1).
In a broadly related study, Kaushanskaya, Yoo, and van Hecke (2013)  found that for English speakers with varying L2 (Spanish) exposure, increased L2 experience was associated with enhanced novel word acquisition but only for novel words that were phonologically unfamiliar (non-English/Spanish sounds) paired with familiar semantic referents (animals). This supports the speech-learning model (Flege, 1987; Flege & Hammond, 1982) that argues that bilinguals are adept at learning unfamiliar phonological combinations. Kaushanskaya et al. note that, while these are the items that most closely simulate the L2 language-learning experience, broader benefits for other items (e.g., familiar phonological words with unfamiliar semantic referents) may be found with more L2 exposure. Kaushanskaya et al.'s (2013)  findings seem to suggest, therefore, that bilingual experience might be expected to specifically facilitate learning of less frequent sound combinations (words with low PP) and/or words that are less similar to words in the lexicon (words with low phonological neighborhood). If a bilingual advantage is demonstrated only for words with low PP/low neighborhood density, then this would indicate that the cognitive mechanisms that underpin the bilingual advantage are sensitive to these psycholinguistic variables. However, to the best of our knowledge, there have been no attempts to investigate the impact of bilingualism on the effects of PND or PP in word learning or vice versa.
In sum, the critical question from the previous literature is whether the bilingual advantage for novel word learning is specific to only certain phonotactic patterns (e.g., words with low PP) and of certain PND (low neighborhood density) or whether bilinguals exhibit an overall word-learning advantage regardless of the phonotactic and neighborhood patterns of the novel word. This question has theoretical implications for the understanding of bilingual word learning and will inform theories of L2 learning and for the specification of the cognitive mechanisms of word learning more generally and, hence, is the focus of the research presented here.
Method
Participants
The participants were 20 monolingual native speakers of English (13 women and 7 men) and 20 Mandarin–English early onset, highly proficient, bilingual speakers (11 women and 9 men). The participants were all university undergraduate students, and the groups were matched for age, bilinguals: M = 21.55 years, SD = 1.00; monolinguals = 21.47 years, SD = 0.94, t(38) = −0.245, p = .807.
All bilingual participants rated their L2 proficiency across four language categories ranging from 0 (no proficiency) to 4 (native like). The self-reported proficiency questionnaire captured both language proficiency as well as the linguistic history of the participants and was similar to other L2 proficiency measures, such as the language experience and proficiency questionnaire (Marian, Blumenfeld, & Kaushanskaya, 2007), the international L2 proficiency rating scale (Ingram & Wylie, 1999), and language proficiency categories (Collier, 2007).
The indicator of language proficiency used in this study was the age of active bilingualism (indexed through speaking). The mean start age of active bilingualism was 6.15 years (age range = 5–7 years, SD = 0.88 years) with an average of almost 15 years of exposure to English (see Table 1). The bilingual participants were native speakers of Mandarin (L1) and had acquired English (L2) in both a classroom context and by immersion in an English-language environment. The language history revealed that the bilingual participants were born to immigrant parents of Chinese background. The participants' mean start age of active bilingualism often coincided with their mean age when their parents migrated to Australia. This indicates that, although the bilingual participants acquired English in classrooms, the L2 learning happened in a classroom context where English is spoken as the native language. This is a critical indicator for our bilingual participants' increased proficiency in L2 in contrast to other bilinguals who learn English as an L2 in an impoverished classroom (nonnative) context (a scenario that is commonplace for most nonnative speakers of English).
Table 1. Demographic and background data of participants.
Demographic and background data of participants.×
Demographic variables Monolinguals Bilinguals p Value a
Age (years) 21.47 (0.938) 21.55 (0.998) .807
Nonword repetition b 70.65 (6.70) 69.55 (7.39) .625
Digit span c 68.10 (9.91) 69.15 (10.17) .746
L2 acquisition age (speaking) 6.15 (0.812)
Proficiency ratings d
Speaking 3.05 (0.394)
Listening 3.40 (0.502)
Reading 3.35 (0.489)
Writing 3.28 (0.487)
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.×
a Significance of a two-sample t test (two-tailed).
Significance of a two-sample t test (two-tailed).×
b Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
c Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
d Proficiency ratings from 0 = not proficient to 4 = highly proficient.
Proficiency ratings from 0 = not proficient to 4 = highly proficient.×
Table 1. Demographic and background data of participants.
Demographic and background data of participants.×
Demographic variables Monolinguals Bilinguals p Value a
Age (years) 21.47 (0.938) 21.55 (0.998) .807
Nonword repetition b 70.65 (6.70) 69.55 (7.39) .625
Digit span c 68.10 (9.91) 69.15 (10.17) .746
L2 acquisition age (speaking) 6.15 (0.812)
Proficiency ratings d
Speaking 3.05 (0.394)
Listening 3.40 (0.502)
Reading 3.35 (0.489)
Writing 3.28 (0.487)
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.×
a Significance of a two-sample t test (two-tailed).
Significance of a two-sample t test (two-tailed).×
b Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
c Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
d Proficiency ratings from 0 = not proficient to 4 = highly proficient.
Proficiency ratings from 0 = not proficient to 4 = highly proficient.×
×
These participants reported that they spoke Mandarin at home and in social situations, especially while communicating with other family members and friends from similar cultural and linguistic backgrounds. English was used to communicate with friends in both formal (e.g., university) and informal (e.g., social) settings. All bilingual participants currently lived and attended university in Australia and met the (stringent) English language requirements for admission. Their proficiency was also reflected in the fact that their English was spoken with a near-native or native accent. The participants did not report any significant language or cognitive impairments.
Before the learning phase, subtests from the Comprehensive Test of Phonological Processing (Wagner, Torgesen, & Rashotte, 1999) were used to test participants' nonword repetition and digit span abilities. Participants' demographic characteristics and self-ratings of bilingual language proficiency are reported in Table 1. The participant groups did not differ in their nonword repetition, t(38) = 0.493, p = .625, or digit span scores, t(38) = −0.220, p = .827.
Stimuli
To examine the effects of PP and PND, we first created 35 bisyllabic nonwords with varying English phonotactic probability and PND as calculated using the English vocabulary of the CELEX database (Baayen, Piepenbrock, & Gulikers, 1995) using an algorithm on the basis of the conventions used by Storkel et al. (2006) . The nonwords were created by first selecting real words that contained either high or low bigram and trigram frequencies (bi-/trigram frequencies refer to the number of occurrences of a particular two- or three-letter string across all the words in a language), or high or low PND. 1   We changed these real words into a nonword by deleting or changing the position of a single phoneme. This generated a list of bisyllabic nonwords with varying bigram and trigram frequencies, whose phonological neighborhood size, biphone frequency, positional segment frequency, and summed biphone frequency were once again extracted from CELEX.
We then constructed matched sets that were high or low in each variable (i.e., higher or lower than the median value of the set). We originally had hoped to have orthogonal contrasts (i.e., four sets that were high in PP and either high or low in PND, or low in PP and either high or low in PND). However, it was not possible to create a set of nonwords that were both low in PP and high in PND. Therefore, the final stimuli consisted of three sets of five words that were (a) low in PP and PND, (b) high in PP and low in PND, and (c) high in PP and high in PND. Although these three categories enabled us to fulfill our aims, they did not allow for examination of the full range of one variable while manipulating the other. Therefore, we examined PP and PND effects in three ways: (a) effects of PP were examined when PND was low; (b) effects of PND were examined when PP was high; and (c) correlated effects of PP and PND were examined by comparing the low PP/low PND set to the high PP/high PND set. This final comparison allowed us to look at what might be considered the “norm” in language learning: when stimuli are high in both variables or low in both variables.
The similarity of the target words to Mandarin was rated by five native speakers of Mandarin on the basis of a 1–4-point rating scale (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin). The ratings indicated that none of the novel words resembled Mandarin words. While carrying the familiarity ratings for Mandarin, we asked the judges to identify whether the stimuli contained any sounds that were unusual for Mandarin. The judges were not able to identify any such sounds. This was not completed for English because the nonwords were constructed using English phonemes (see Appendix).
Referents
Each novel word was paired with a novel picture as a referent. The pictures consisted of color images of 15 novel alien creatures differing in physical appearance and characteristics selected from the Gupta et al. (2004)  stimulus set in such a way that all 15 were visually distinct. Each alien was also assigned with attributes (definition) relating to physical or mental characteristics of the alien (e.g., “/tæbɛk/ likes flowers and owns a beautiful garden”) and unrelated to the physical appearance. The novel words and their definitions are given in the Appendix.
Procedure
All 15 novel words were presented for learning in one session. The word-learning session followed background testing and completion of the language proficiency questionnaire. Each learning phase consisted of a presentation of the referent picture on an Apple Mac OS X 10.7 laptop monitor together with simultaneous presentation of an audio recording of the novel word and its definition (to provide additional semantic/associative elaboration). For example, “this is fɒni:s, fɒni:s can sing beautifully and is known as the heavenly singer.” Pictures remained on the screen for 30 s and were followed by the next stimulus.
Following presentation of all stimuli, they were presented again in a different random order four more times. In the first four presentations, participants were told to look at the picture and listen and memorize the name of the picture and the sentence about its characteristics. After the final (fifth) presentation of each item, while the picture of the item remained on the screen, the participants were asked to repeat the word aloud three times to maximize the learning.
We carried out two phases of testing: one test immediately following the five presentations for word learning and a second test one week later. Each participant was assessed on the acquisition of the novel words using a picture-naming task. The picture-naming task aimed to evaluate learning of the word form and its association with the picture referent and also allowed for examination of the effects of PP and PND on learning. In this task, the target picture was presented, and the participant was asked to name it as quickly as possible. The responses were audio-recorded for later analysis.
Analyses
Analyses were performed on response accuracy (raw number of correct responses out of five) for each task. Object naming was considered correct if all phonemes were produced correctly. This was scored on the basis of review of the audio recordings by the first author, on two separate occasions to ensure reliability. On three instances when the response was unclear, an independent researcher reviewed the recording. The total correct responses for all stimuli subsets were calculated.
We analyzed the accuracy for naming using a series of mixed analyses of variance. These analyses of variance comprised one between-subjects factor, language group (bilingual/monolingual), and two within-subject factors: testing time and condition (high/low in the variable under consideration). The first analysis examined PP (high vs. low); the second, PND (high vs. low); and the final analysis, the correlated comparisons (high in both PP and PND vs. low in both PP and PND).
Results
Figure 1 provides the results of the naming task, and Table 2 gives the results of the statistical analyses. For clarity, we first summarize the effects of group and time across all analyses and then report the results of the analyses manipulating PP and PND.
Figure 1.

Effects of phonotactic probability (high = stimuli with high phonotactic probability, low = stimuli with low phonotactic probability).

 Effects of phonotactic probability (high = stimuli with high phonotactic probability, low = stimuli with low phonotactic probability).
Figure 1.

Effects of phonotactic probability (high = stimuli with high phonotactic probability, low = stimuli with low phonotactic probability).

×
Table 2. Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.
Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.×
Effect Degrees of freedom F p ηp 2
Analysis 1: phonotactic probability
 Group (monolingual vs. bilingual) 1, 38 15.70 <.001* .292
 Time (immediate vs. delayed) 1, 38 35.10 <.001* .480
 PP (high vs. low) 1, 38 18.41 <.001* .326
 Time × Group 1, 38 1.40 .243
 Time × PP 1, 38 1.96 .170
 Group × PP 1, 38 0.38 .543
 Time × Group × PP 1, 38 0.04 .843
Analysis 2: phonological neighborhood density
 Group (monolingual vs. bilingual) 1, 38 12.51 .001* .248
 Time (immediate vs. delayed) 1, 38 43.41 <.001* .533
 PND (high vs. low) 1, 38 10.72 .002* .220
 Time × Group 1, 38 0.89 .353
 Time × PND 1, 38 3.22 .080
 Group × PND 1, 38 2.55 .118
 Time × Group × PND 1, 38 0.01 .932
Analysis 3: correlated comparisons
 Group (monolingual vs. bilingual) 1, 38 11.31 .002* .229
 Time (immediate vs. delayed) 1, 38 85.74 <.001* .693
 Condition (HiPP/PND vs. LowPP/PND) 1, 38 0.005 .942
 Time × Group 1, 38 1.75 .194
 Time × Condition 1, 38 0.498 .485
 Group × Condition 1, 38 1.20 .279
 Time × Group × Condition 1, 38 0.091 .764
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.×
* p < .01, ηp 2 (effect size) = partial eta square.
p < .01, ηp 2 (effect size) = partial eta square.×
Table 2. Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.
Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.×
Effect Degrees of freedom F p ηp 2
Analysis 1: phonotactic probability
 Group (monolingual vs. bilingual) 1, 38 15.70 <.001* .292
 Time (immediate vs. delayed) 1, 38 35.10 <.001* .480
 PP (high vs. low) 1, 38 18.41 <.001* .326
 Time × Group 1, 38 1.40 .243
 Time × PP 1, 38 1.96 .170
 Group × PP 1, 38 0.38 .543
 Time × Group × PP 1, 38 0.04 .843
Analysis 2: phonological neighborhood density
 Group (monolingual vs. bilingual) 1, 38 12.51 .001* .248
 Time (immediate vs. delayed) 1, 38 43.41 <.001* .533
 PND (high vs. low) 1, 38 10.72 .002* .220
 Time × Group 1, 38 0.89 .353
 Time × PND 1, 38 3.22 .080
 Group × PND 1, 38 2.55 .118
 Time × Group × PND 1, 38 0.01 .932
Analysis 3: correlated comparisons
 Group (monolingual vs. bilingual) 1, 38 11.31 .002* .229
 Time (immediate vs. delayed) 1, 38 85.74 <.001* .693
 Condition (HiPP/PND vs. LowPP/PND) 1, 38 0.005 .942
 Time × Group 1, 38 1.75 .194
 Time × Condition 1, 38 0.498 .485
 Group × Condition 1, 38 1.20 .279
 Time × Group × Condition 1, 38 0.091 .764
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.×
* p < .01, ηp 2 (effect size) = partial eta square.
p < .01, ηp 2 (effect size) = partial eta square.×
×
Effects of Group and Time
All three analyses showed a significant main effect of group indicating better naming for bilinguals compared with monolinguals. There was also, as might be expected, a significant main effect of time: Participants performed better immediately compared with at one-week delay. The interactions between group and time were not significant.
Analysis 1: Manipulating PP
This analysis examined the effects of PP when the sets were matched on PND (both low in PND). There was a significant main effect of PP with lower probability words better named than higher probability words (see Figure 1, Table 2). There were no significant two- or three-way interactions with group and time. This showed that the bilingual advantage in learning, as indexed by the naming task, did not vary according to PP.
Analysis 2: Manipulating PND
This analysis examined the effects of PND when sets were matched for PP (both high in PP) and found a main effect of PND with higher accuracy for high-density words than low-density words but no interactions with group or time (see Figure 2, Table 2). Once again, this indicated that the bilingual advantage did not differ across high and low PND novel word learning.
Figure 2.

Effects of phonological neighborhood density (high = stimuli with high phonological neighborhood density, low = stimuli with low phonological neighborhood density).

 Effects of phonological neighborhood density (high = stimuli with high phonological neighborhood density, low = stimuli with low phonological neighborhood density).
Figure 2.

Effects of phonological neighborhood density (high = stimuli with high phonological neighborhood density, low = stimuli with low phonological neighborhood density).

×
Analysis 3: Correlated Comparisons
This analysis examined the effects of phonological properties of stimuli on naming comparing sets that were high in both PP and PND with sets that were low in both variables. This analysis found no effect of this manipulation on naming and no significant interactions (see Figure 3, Table 2).
Figure 3.

Correlated comparisons (high = words with high phonotactic probability and phonological neighborhood density, low = words with low phonotactic probability and phonological neighborhood density).

 Correlated comparisons (high = words with high phonotactic probability and phonological neighborhood density, low = words with low phonotactic probability and phonological neighborhood density).
Figure 3.

Correlated comparisons (high = words with high phonotactic probability and phonological neighborhood density, low = words with low phonotactic probability and phonological neighborhood density).

×
Discussion
This study aimed not only to replicate the bilingual advantage that has been found for novel word learning (Kaushanskaya & Marian, 2009b) but also, more importantly, to examine whether this advantage was influenced by the PP and PND of novel words. In order to investigate this, we compared the performance of high-proficiency early bilinguals and matched monolinguals on a word-learning task that manipulated the PP and PND of the novel words. We found that bilinguals outperformed monolinguals in learning as measured by a picture-naming task. There were also clear effects of PP and PND on word learning. However, there was no difference between bilinguals and monolinguals in the extent to which these variables influenced learning.
Interestingly, there was no difference in learning between stimuli that were high in both PP and PND and low in both. This result is consistent with recent studies with monolinguals that have reported nonsignificant effects for correlated comparisons with PP and PND (e.g., Storkel & Lee, 2011). Our results provide initial evidence for nonsignificant effects for correlated comparisons in bilinguals too. However, although it was not possible to develop stimuli for a full factorial design, when we examined the effects of these two variables independently, we found that the two factors had opposite effects on learning.
When manipulating PP (in stimuli that were low in PND), both bilinguals and monolinguals demonstrated better accuracy for learning words of lower PP. This effect is consistent with that found by Storkel et al. (2006), although they found an effect of PP for only the initial stages of learning, arguing that learning was triggered better when the stimuli were more novel. However, this explanation seems less appropriate to a context, such as that used here, where the words were explicitly flagged as novel (and do not need to be detected in a story context). Perhaps, in a direct learning task, such as that used here, learning abilities are at a peak when the novelty associated with the task increases. In the context of our task, this would occur specifically for low PP items because of their unfamiliarity (and, therefore, novelty) compared with high PP items. Therefore, this may result in a low probability advantage in low neighborhood density words for both bilinguals and monolinguals.
For the PND manipulation (within words of high PP), words of higher neighborhood density were better learned. Our results therefore replicate Storkel et al.'s (2006)  findings on monolingual adult word learning in a story context. Storkel et al. suggested that the advantage for high neighborhood density words could be due to better consolidation of representations through the links with many neighbors. They suggest that the activated neighbors of the novel word would activate their phonemes and that these phonemes may provide feedback that facilitates the acquisition of the novel word. This interactive process results in the strengthening of the representation for the novel word. This could also hold for our task. Hence, while low PP may enhance initial learning, higher phonological neighborhood could enhance consolidation of that learning.
An advantage for words with low PP also may speak to the speech-learning model (Flege, 1987). This model suggests that the structural similarities between L1 and L2 may have little effect on L2 learning. Flege (1987)  argues that L2 learners are highly capable of learning new phonemes in L2 without accessing prior knowledge from L1. When they learn a new phoneme in L2, they are also capable of independently modifying the previously learned similar phonemic patterns in L2 without relating it to L1 structures. Although the speech-learning model (Flege, 1987) was developed in the context of L2 learning, we suggest that the model, in fact, may logically suggest that any speaker should be more adept at learning words containing less familiar phonemes. Critically, however, all of our phonemes were of high familiarity to all our speakers, but what varied was the familiarity of the combinations of these phonemes. It is possible, therefore, that not only less familiar phonemes but also less familiar combinations of phonemes are better learned.
Overall, bilinguals performed more accurately in learning, as indexed by naming, irrespective of the PP and PND of the stimuli. This finding replicates the previous research demonstrating facilitatory effects of bilingualism for novel word learning (e.g., Kaushanskaya & Marian 2009a, 2009b; Nair et al., 2016). It is also possible that early experience with an L2 could generally facilitate the language-learning mechanism (e.g., Bartolotti & Marian, 2012; Bartolotti, Marian, Schroeder & Shook, 2011; Grey, 2013; Van Hell & Mahn, 1997, Wang & Saffran, 2014; Yoshida, Tran, Benitez, & Kuwabara, 2011). For example, the phonological system of bilingual participants may have influenced their word-learning skills: Kaushanskaya and Marian (2009b)  have argued that an experience with more than one language makes the bilingual's phonological system comparatively more open. However, this might be thought to imply that the bilingual's phonological system may therefore be more open to accepting any phonological combination (even unusual combinations), in contrast to the specific phonological tuning that occurs for monolinguals in their native language (e.g., Kuhl, Williams, Lacerda, Stevens, & Lindblom, 1992).
How might the effects of PP and PND on bilingual word learning relate to the language production mechanism of a bilingual speaker? This is particularly critical given that we did not find an interaction between either PP or PND and the learning of the different participant groups. This suggests that the bilingual advantage in novel word learning is at least partially rooted in factors other than phonological or lexical properties of the novel words. In other words, it is likely that the bilingual advantage transcends the potential effects of PP and PND.
In the “inhibitory control” model for language production (Green, 1998), the language production mechanism of bilinguals is mediated by the lexico-semantic system, the language task schema, and the supervisory attentional system. It is the language task schema that helps in selecting the appropriate language and inhibiting the nonrelevant language. In the context of word learning, the most crucial component of the inhibitory control model is the supervisory attentional system. Green (1998)  suggested that the supervisory attentional system is a goal-oriented mechanism, which is especially skilled at facilitating tasks that have been not previously performed. Therefore, when an individual performs a novel task associated with language production, such as novel word learning, the supervisory attentional system is employed to ensure its successful completion.
While it is likely that components of the inhibitory control model are present in both bilinguals and monolinguals, the supervisory attentional system of bilinguals is argued to be more efficient and more active than in monolinguals (see Bialystok, Craik, & Luk, 2012  for a detailed discussion), and our word-learning data appear consistent with this account. We propose that the supervisory attentional system recognizes learning of any novel word as a novel task and allocates all available attentional resources to execute the successful completion of the task. This leads to more attentional resources being available in the bilingual speaker than the monolingual, resulting in enhanced learning. Therefore, we hypothesize that the bilingual advantage in novel word learning could be due to the enhancement of this mechanism that underpins word learning. Moreover, our data show that this mechanism is not sensitive to the effects of PP or PND.
There are, of course, limitations related to the current study. First, we did not explicitly manipulate or control PP and PND in Mandarin as well as in English. Therefore, we cannot be sure the extent to which the bilingual's L1 (Mandarin) affected these variables. For example, it is possible (if unlikely) that some stimuli may have been of higher PP and/or PND for the bilinguals. Future research should examine this potential confound from L1. Second, it would have been preferable to completely orthogonally manipulate PND and PP; however, at least within our stimuli, this was not possible. Therefore, we could only manipulate PP within words with low PND and PND within words with high PP. The lack of a full orthogonal manipulation restricted our ability to examine, for example, whether there was an interaction between the two effects.
Conclusions
This study replicates and extends the findings regarding the effects of PP and PND on learning. Like Storkel et al. (2006), we demonstrate that, despite the high correlation between these two variables, their effects not only can be dissociated but are also in different directions—an inhibitory effect of PP and a facilitatory effect of phonological neighborhood on word learning. Moreover, we also replicate previous findings of a bilingual advantage in novel word learning and provide two important contributions to the literature on the linguistic effects of bilingualism. First, we demonstrate that the facilitatory effects of bilingualism on novel word learning are stable even when the PP and neighborhood density of the novel words varies; and second, we propose that the loci of these advantages may be an efficient supervisory attentional system. These results have theoretical implications for understanding the effect of cognitive mechanisms on bilingual novel word learning as well as potential future clinical implications.
Acknowledgments
During the preparation of this article, Vishnu KK Nair was supported by an International Macquarie University Research Excellence Scholarship (iMQRES 2011105), ARC Centre of Excellence in Cognition and its Disorders, Department of Cognitive Science, Macquarie University. Lyndsey Nickels was supported by an Australian Research Council Future Fellowship (FT 120100102), and Britta Biedermann was supported by an Australian Research Council Postdoctoral Fellowship and Discovery Project (DP 11010079). We would like to thank Katherine Demuth for her suggestions in the planning stages of this research and on an earlier version of the paper, and Steven Saunders and Lois MacCullagh for their assistance in stimulus preparation and recording.
References
Abutalebi, J., Della Rosa, P. A., Green, D. W., Hernandez, M., Scifo, P., Keim, R., … Costa, A. (2011). Bilingualism tunes the anterior cingulate cortex for conflict monitoring. Cerebral Cortex, 22, 2076–2086. https://doi.org/10.1093/cercor/bhr287 [Article] [PubMed]
Abutalebi, J., Della Rosa, P. A., Green, D. W., Hernandez, M., Scifo, P., Keim, R., … Costa, A. (2011). Bilingualism tunes the anterior cingulate cortex for conflict monitoring. Cerebral Cortex, 22, 2076–2086. https://doi.org/10.1093/cercor/bhr287 [Article] [PubMed]×
Antoniou, M., Liang, E., Ettlinger, M., & Wong, P. C. (2015). The bilingual advantage in phonetic learning. Bilingualism: Language and Cognition, 18(04), 683–695. https://doi.org/10.1017/S1366728914000777 [Article]
Antoniou, M., Liang, E., Ettlinger, M., & Wong, P. C. (2015). The bilingual advantage in phonetic learning. Bilingualism: Language and Cognition, 18(04), 683–695. https://doi.org/10.1017/S1366728914000777 [Article] ×
Baayen, R. H., Piepenbrock, R., & Gulikers, L. (1995). The CELEX lexical database (release 2). Philadelphia, PA: Linguistic Data Consortium.
Baayen, R. H., Piepenbrock, R., & Gulikers, L. (1995). The CELEX lexical database (release 2). Philadelphia, PA: Linguistic Data Consortium.×
Bartolotti, J., & Marian, V. (2012). Language learning and control in monolinguals and bilinguals. Cognitive Science, 36(6), 1129–1147. https://doi.org/10.1111/j.1551-6709.2012.01243.x [Article] [PubMed]
Bartolotti, J., & Marian, V. (2012). Language learning and control in monolinguals and bilinguals. Cognitive Science, 36(6), 1129–1147. https://doi.org/10.1111/j.1551-6709.2012.01243.x [Article] [PubMed]×
Bartolotti, J., Marian, V., Schroeder, S. R., & Shook, A. (2011). Bilingualism and inhibitory control influence statistical learning of novel word forms. Frontiers in Psychology, 2(324), 1–9. https://doi.org/10.3389/fpsyg.2011.00324 [PubMed]
Bartolotti, J., Marian, V., Schroeder, S. R., & Shook, A. (2011). Bilingualism and inhibitory control influence statistical learning of novel word forms. Frontiers in Psychology, 2(324), 1–9. https://doi.org/10.3389/fpsyg.2011.00324 [PubMed]×
Baus, C., Costa, A., & Carreiras, M. (2008). Neighbourhood density and frequency effects in speech production: A case for interactivity. Language and Cognitive Processes, 23(6), 866–888. https://doi.org/10.1080/01690960801962372 [Article]
Baus, C., Costa, A., & Carreiras, M. (2008). Neighbourhood density and frequency effects in speech production: A case for interactivity. Language and Cognitive Processes, 23(6), 866–888. https://doi.org/10.1080/01690960801962372 [Article] ×
Bialystok, E. (2009). Bilingualism: The good, the bad, and the indifferent. Bilingualism: Language and Cognition, 12(01), 3–11. https://doi.org/10.1017/S1366728908003477 [Article]
Bialystok, E. (2009). Bilingualism: The good, the bad, and the indifferent. Bilingualism: Language and Cognition, 12(01), 3–11. https://doi.org/10.1017/S1366728908003477 [Article] ×
Bialystok, E., Craik, F. I., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: Evidence from the Simon task. Psychology and Aging, 19(2), 290–303. https://doi.org/10.1037/0882-7974.19.2.290 [Article] [PubMed]
Bialystok, E., Craik, F. I., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: Evidence from the Simon task. Psychology and Aging, 19(2), 290–303. https://doi.org/10.1037/0882-7974.19.2.290 [Article] [PubMed]×
Bialystok, E., Craik, F. I., & Luk, G. (2012). Bilingualism: Consequences for mind and brain. Trends in Cognitive Sciences, 16(4), 240–250. https://doi.org/10.1016/j.tics.2012.03.001 [Article] [PubMed]
Bialystok, E., Craik, F. I., & Luk, G. (2012). Bilingualism: Consequences for mind and brain. Trends in Cognitive Sciences, 16(4), 240–250. https://doi.org/10.1016/j.tics.2012.03.001 [Article] [PubMed]×
Bialystok, E., Craik, F. I., & Ryan, J. (2006). Executive control in a modified antisaccade task: Effects of aging and bilingualism. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32(6), 1341–1354. https://doi.org/10.1037/0278-7393.32.6.1341 [Article] [PubMed]
Bialystok, E., Craik, F. I., & Ryan, J. (2006). Executive control in a modified antisaccade task: Effects of aging and bilingualism. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32(6), 1341–1354. https://doi.org/10.1037/0278-7393.32.6.1341 [Article] [PubMed]×
Bradley, K. A., King, K. E., & Hernandez, A. E. (2013). Language experience differentiates prefrontal and subcortical activation of the cognitive control network in novel word learning. NeuroImage, 67, 101–110. https://doi.org/10.1016/j.neuroimage.2012.11.018 [Article] [PubMed]
Bradley, K. A., King, K. E., & Hernandez, A. E. (2013). Language experience differentiates prefrontal and subcortical activation of the cognitive control network in novel word learning. NeuroImage, 67, 101–110. https://doi.org/10.1016/j.neuroimage.2012.11.018 [Article] [PubMed]×
Collier, C. (2007). Classroom language interaction checklist. Retrieved from http://www.k12.wa.us/MigrantBilingual/pubdocs/CLIC2016.pdf
Collier, C. (2007). Classroom language interaction checklist. Retrieved from http://www.k12.wa.us/MigrantBilingual/pubdocs/CLIC2016.pdf ×
Costa, A., Hernández, M., & Sebastián-Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task. Cognition, 106(1), 59–86. https://doi.org/10.1016/j.cognition.2006.12.013 [Article] [PubMed]
Costa, A., Hernández, M., & Sebastián-Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task. Cognition, 106(1), 59–86. https://doi.org/10.1016/j.cognition.2006.12.013 [Article] [PubMed]×
Flege, J. E. (1987). The production of “new” and “similar” phones in a foreign language: Evidence for the effect of equivalence classification. Journal of Phonetics, 15(1), 47–65.
Flege, J. E. (1987). The production of “new” and “similar” phones in a foreign language: Evidence for the effect of equivalence classification. Journal of Phonetics, 15(1), 47–65.×
Flege, J. E., & Hammond, R. M. (1982). Mimicry of non-distinctive phonetic differences between language varieties. Studies in Second Language Acquisition, 5(01), 1–17. https://doi.org/10.1017/S0272263100004563 [Article]
Flege, J. E., & Hammond, R. M. (1982). Mimicry of non-distinctive phonetic differences between language varieties. Studies in Second Language Acquisition, 5(01), 1–17. https://doi.org/10.1017/S0272263100004563 [Article] ×
Gollan, T. H., Montoya, R. I., Fennema-Notestine, C., & Morris, S. K. (2005). Bilingualism affects picture naming but not picture classification. Memory & Cognition, 33(7), 1220–1234. https://doi.org/10.3758/BF03193224 [Article] [PubMed]
Gollan, T. H., Montoya, R. I., Fennema-Notestine, C., & Morris, S. K. (2005). Bilingualism affects picture naming but not picture classification. Memory & Cognition, 33(7), 1220–1234. https://doi.org/10.3758/BF03193224 [Article] [PubMed]×
Green, D. W. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1(02), 67–81. https://doi.org/10.1017/S1366728998000133 [Article]
Green, D. W. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1(02), 67–81. https://doi.org/10.1017/S1366728998000133 [Article] ×
Grey, S. E. (2013). A neurocognitive investigation of bilingual advantages at additional language learning. Washington, DC: Georgetown University.
Grey, S. E. (2013). A neurocognitive investigation of bilingual advantages at additional language learning. Washington, DC: Georgetown University.×
Gupta, P., Lipinski, J., Abbs, B., Lin, P. H., Aktunc, E., Ludden, D., … Newman, R. (2004). Space aliens and nonwords: Stimuli for investigating the learning of novel word-meaning pairs. Behavior Research Methods, Instruments, & Computers, 36(4), 599–603. https://doi.org/10.3758/BF03206540 [Article]
Gupta, P., Lipinski, J., Abbs, B., Lin, P. H., Aktunc, E., Ludden, D., … Newman, R. (2004). Space aliens and nonwords: Stimuli for investigating the learning of novel word-meaning pairs. Behavior Research Methods, Instruments, & Computers, 36(4), 599–603. https://doi.org/10.3758/BF03206540 [Article] ×
Hernandez, A. E., & Li, P. (2007). Age of acquisition: Its neural and computational mechanisms. Psychological Bulletin, 133(4), 638–650. https://doi.org/10.1037/0033-2909.133.4.638 [Article] [PubMed]
Hernandez, A. E., & Li, P. (2007). Age of acquisition: Its neural and computational mechanisms. Psychological Bulletin, 133(4), 638–650. https://doi.org/10.1037/0033-2909.133.4.638 [Article] [PubMed]×
Hoover, J. R., Storkel, H. L., & Hogan, T. P. (2010). A cross-sectional comparison of the effects of phonotactic probability and neighborhood density on word learning by preschool children. Journal of Memory and Language, 63(1), 100–116. https://doi.org/10.1016/j.jml.2010.02.003 [Article] [PubMed]
Hoover, J. R., Storkel, H. L., & Hogan, T. P. (2010). A cross-sectional comparison of the effects of phonotactic probability and neighborhood density on word learning by preschool children. Journal of Memory and Language, 63(1), 100–116. https://doi.org/10.1016/j.jml.2010.02.003 [Article] [PubMed]×
Ingram, D., & Wylie, E. (1999). International second language proficiency ratings. Brisbane, Australia: Centre for Applied Linguistics and Languages, Griffith University.
Ingram, D., & Wylie, E. (1999). International second language proficiency ratings. Brisbane, Australia: Centre for Applied Linguistics and Languages, Griffith University.×
Ivanova, I., & Costa, A. (2008). Does bilingualism hamper lexical access in speech production? Acta Psychologica, 127(2), 277–288. https://doi.org/10.1016/j.actpsy.2007.06.003 [Article] [PubMed]
Ivanova, I., & Costa, A. (2008). Does bilingualism hamper lexical access in speech production? Acta Psychologica, 127(2), 277–288. https://doi.org/10.1016/j.actpsy.2007.06.003 [Article] [PubMed]×
Kaushanskaya, M. (2012). Cognitive mechanisms of word learning in bilingual and monolingual adults: The role of phonological memory. Bilingualism: Language and Cognition, 15(03), 470–489. https://doi.org/10.1017/S1366728911000472 [Article]
Kaushanskaya, M. (2012). Cognitive mechanisms of word learning in bilingual and monolingual adults: The role of phonological memory. Bilingualism: Language and Cognition, 15(03), 470–489. https://doi.org/10.1017/S1366728911000472 [Article] ×
Kaushanskaya, M., & Marian, V. (2008). Age-of-acquisition effects in the development of a bilingual advantage for word learning. In Proceedings of the 32nd Annual Boston University Conference on Language Development (pp. 213–224). Evanston, IL: Northwestern University.
Kaushanskaya, M., & Marian, V. (2008). Age-of-acquisition effects in the development of a bilingual advantage for word learning. In Proceedings of the 32nd Annual Boston University Conference on Language Development (pp. 213–224). Evanston, IL: Northwestern University.×
Kaushanskaya, M., & Marian, V. (2009a). Bilingualism reduces native-language interference during novel word learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(3), 829–835. https://doi.org/10.1037/a0015275 [Article]
Kaushanskaya, M., & Marian, V. (2009a). Bilingualism reduces native-language interference during novel word learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(3), 829–835. https://doi.org/10.1037/a0015275 [Article] ×
Kaushanskaya, M., & Marian, V. (2009b). The bilingual advantage in novel word learning. Psychonomic Bulletin & Review, 16(4), 705–710. https://doi.org/10.3758/PBR.16.4.705 [Article]
Kaushanskaya, M., & Marian, V. (2009b). The bilingual advantage in novel word learning. Psychonomic Bulletin & Review, 16(4), 705–710. https://doi.org/10.3758/PBR.16.4.705 [Article] ×
Kaushanskaya, M., Yoo, J., & Van Hecke, S. (2013). Word learning in adults with second-language experience: Effects of phonological and referent familiarity. Journal of Speech, Language, and Hearing Research, 56(2), 667–678. https://doi.org/10.1044/1092-4388(2012/11-0084) [Article]
Kaushanskaya, M., Yoo, J., & Van Hecke, S. (2013). Word learning in adults with second-language experience: Effects of phonological and referent familiarity. Journal of Speech, Language, and Hearing Research, 56(2), 667–678. https://doi.org/10.1044/1092-4388(2012/11-0084) [Article] ×
Kuhl, P. K., Williams, K. A., Lacerda, F., Stevens, K. N., & Lindblom, B. (1992). Linguistic experience alters phonetic perception in infants by 6 months of age. Science, 255(5044), 606–608. https://doi.org/10.1126/science.1736364 [Article] [PubMed]
Kuhl, P. K., Williams, K. A., Lacerda, F., Stevens, K. N., & Lindblom, B. (1992). Linguistic experience alters phonetic perception in infants by 6 months of age. Science, 255(5044), 606–608. https://doi.org/10.1126/science.1736364 [Article] [PubMed]×
Lado, R. (1957). Linguistics across cultures: Applied linguistics for language teachers. Ann Arbor: University of Michigan Press.
Lado, R. (1957). Linguistics across cultures: Applied linguistics for language teachers. Ann Arbor: University of Michigan Press.×
Luce, P. A., & Pisoni, D. B. (1998). Recognizing spoken words: The neighborhood activation model. Ear and Hearing, 19(1), 1. [Article] [PubMed]
Luce, P. A., & Pisoni, D. B. (1998). Recognizing spoken words: The neighborhood activation model. Ear and Hearing, 19(1), 1. [Article] [PubMed]×
Marian, V., Blumenfeld, H. K., & Kaushanskaya, M. (2007). The Language Experience and Proficiency Questionnaire (LEAP-Q): Assessing language profiles in bilinguals and multilinguals. Journal of Speech, Language, and Hearing Research, 50(4), 940–967. https://doi.org/10.1044/1092-4388(2007/067) [Article]
Marian, V., Blumenfeld, H. K., & Kaushanskaya, M. (2007). The Language Experience and Proficiency Questionnaire (LEAP-Q): Assessing language profiles in bilinguals and multilinguals. Journal of Speech, Language, and Hearing Research, 50(4), 940–967. https://doi.org/10.1044/1092-4388(2007/067) [Article] ×
Nair, V. K., Biedermann, B., & Nickels, L. (2016). Consequences of late bilingualism for novel word learning: Evidence from Tamil–English bilingual speakers. International Journal of Bilingualism, 20(4), 473–487. https://doi.org/10.1177/1367006914567005 [Article]
Nair, V. K., Biedermann, B., & Nickels, L. (2016). Consequences of late bilingualism for novel word learning: Evidence from Tamil–English bilingual speakers. International Journal of Bilingualism, 20(4), 473–487. https://doi.org/10.1177/1367006914567005 [Article] ×
Papagno, C., & Vallar, G. (1995). Verbal short-term memory and vocabulary learning in polyglots. The Quarterly Journal of Experimental Psychology, 48(1), 98–107. https://doi.org/10.1080/14640749508401378 [Article] [PubMed]
Papagno, C., & Vallar, G. (1995). Verbal short-term memory and vocabulary learning in polyglots. The Quarterly Journal of Experimental Psychology, 48(1), 98–107. https://doi.org/10.1080/14640749508401378 [Article] [PubMed]×
Rogers, C. L., Lister, J. J., Febo, D. M., Besing, J. M., & Abrams, H. B. (2006). Effects of bilingualism, noise, and reverberation on speech perception by listeners with normal hearing. Applied Psycholinguistics, 27(03), 465–485. https://doi.org/10.1017/S014271640606036X [Article]
Rogers, C. L., Lister, J. J., Febo, D. M., Besing, J. M., & Abrams, H. B. (2006). Effects of bilingualism, noise, and reverberation on speech perception by listeners with normal hearing. Applied Psycholinguistics, 27(03), 465–485. https://doi.org/10.1017/S014271640606036X [Article] ×
Sadat, J., Martin, C. D., Costa, A., & Alario, F. (2014). Reconciling phonological neighbourhood effects in speech production through single trial analysis. Cognitive Psychology, 68, 33–58. https://doi.org/10.1016/j.cogpsych.2013.10.001 [Article] [PubMed]
Sadat, J., Martin, C. D., Costa, A., & Alario, F. (2014). Reconciling phonological neighbourhood effects in speech production through single trial analysis. Cognitive Psychology, 68, 33–58. https://doi.org/10.1016/j.cogpsych.2013.10.001 [Article] [PubMed]×
Sheng, L., McGregor, K. K., & Marian, V. (2006). Lexical–semantic organization in bilingual children: Evidence from a repeated word association task. Journal of Speech, Language, and Hearing Research, 49(3), 572–587. https://doi.org/10.1044/1092-4388(2006/041) [Article]
Sheng, L., McGregor, K. K., & Marian, V. (2006). Lexical–semantic organization in bilingual children: Evidence from a repeated word association task. Journal of Speech, Language, and Hearing Research, 49(3), 572–587. https://doi.org/10.1044/1092-4388(2006/041) [Article] ×
Storkel, H. L. (2001). Learning new words. Phonotactic probability in language development. Journal of Speech, Language, and Hearing Research, 44(6), 1321–1337. https://doi.org/10.1044/1092-4388(2001/103) [Article]
Storkel, H. L. (2001). Learning new words. Phonotactic probability in language development. Journal of Speech, Language, and Hearing Research, 44(6), 1321–1337. https://doi.org/10.1044/1092-4388(2001/103) [Article] ×
Storkel, H. L., Armbrüster, J., & Hogan, T. P. (2006). Differentiating phonotactic probability and neighborhood density in adult word learning. Journal of Speech, Language, and Hearing Research, 49(6), 1175–1192. https://doi.org/10.1044/1092-4388(2006/085) [Article]
Storkel, H. L., Armbrüster, J., & Hogan, T. P. (2006). Differentiating phonotactic probability and neighborhood density in adult word learning. Journal of Speech, Language, and Hearing Research, 49(6), 1175–1192. https://doi.org/10.1044/1092-4388(2006/085) [Article] ×
Storkel, H. L., & Lee, S. Y. (2011). The independent effects of phonotactic probability and neighbourhood density on lexical acquisition by preschool children. Language and Cognitive Processes, 26(2), 191–211. https://doi.org/10.1080/01690961003787609 [Article] [PubMed]
Storkel, H. L., & Lee, S. Y. (2011). The independent effects of phonotactic probability and neighbourhood density on lexical acquisition by preschool children. Language and Cognitive Processes, 26(2), 191–211. https://doi.org/10.1080/01690961003787609 [Article] [PubMed]×
Van Hell, J. G., & Mahn, A. C. (1997). Keyword mnemonics versus rote rehearsal: Learning concrete and abstract foreign words by experienced and inexperienced learners. Language Learning, 47(3), 507–546. https://doi.org/10.1111/0023-8333.00018 [Article]
Van Hell, J. G., & Mahn, A. C. (1997). Keyword mnemonics versus rote rehearsal: Learning concrete and abstract foreign words by experienced and inexperienced learners. Language Learning, 47(3), 507–546. https://doi.org/10.1111/0023-8333.00018 [Article] ×
Vitevitch, M. S. (2002). The influence of phonological similarity neighborhoods on speech production. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(4), 735–747. https://doi.org/10.1037//0278-7393.28.4.735 [Article] [PubMed]
Vitevitch, M. S. (2002). The influence of phonological similarity neighborhoods on speech production. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(4), 735–747. https://doi.org/10.1037//0278-7393.28.4.735 [Article] [PubMed]×
Vitevitch, M. S., & Luce, P. A. (1999). Probabilistic phonotactics and neighborhood activation in spoken word recognition. Journal of Memory and Language, 40(3), 374–408. [Article]
Vitevitch, M. S., & Luce, P. A. (1999). Probabilistic phonotactics and neighborhood activation in spoken word recognition. Journal of Memory and Language, 40(3), 374–408. [Article] ×
Vitevitch, M. S., & Luce, P. A. (2005). Increases in phonotactic probability facilitate spoken nonword repetition. Journal of Memory and Language, 52(2), 193–204. https://doi.org/10.1006/brln.1999.2116 [Article]
Vitevitch, M. S., & Luce, P. A. (2005). Increases in phonotactic probability facilitate spoken nonword repetition. Journal of Memory and Language, 52(2), 193–204. https://doi.org/10.1006/brln.1999.2116 [Article] ×
Vitevitch, M. S., Luce, P. A., Pisoni, D. B., & Auer, E. T. (1999). Phonotactics, neighborhood activation, and lexical access for spoken words. Brain and Language, 68(1), 306–311. [Article] [PubMed]
Vitevitch, M. S., Luce, P. A., Pisoni, D. B., & Auer, E. T. (1999). Phonotactics, neighborhood activation, and lexical access for spoken words. Brain and Language, 68(1), 306–311. [Article] [PubMed]×
Wagner, R. K., Torgesen, J. K., & Rashotte, C. A. (1999). Comprehensive Test of Phonological Processing: CTOPP. Austin, TX: Pro-Ed.
Wagner, R. K., Torgesen, J. K., & Rashotte, C. A. (1999). Comprehensive Test of Phonological Processing: CTOPP. Austin, TX: Pro-Ed.×
Wang, T., & Saffran, J. R. (2014). Statistical learning of a tonal language: The influence of bilingualism and previous linguistic experience. Frontiers in Psychology, 5, 953. https://doi.org/10.3389/fpsyg.2014.00953 [PubMed]
Wang, T., & Saffran, J. R. (2014). Statistical learning of a tonal language: The influence of bilingualism and previous linguistic experience. Frontiers in Psychology, 5, 953. https://doi.org/10.3389/fpsyg.2014.00953 [PubMed]×
Yoshida, H., Tran, D. N., Benitez, V., & Kuwabara, M. (2011). Inhibition and adjective learning in bilingual and monolingual children. Frontiers in Psychology, 2, 210. https://doi.org/10.3389/fpsyg.2011.00210 [Article] [PubMed]
Yoshida, H., Tran, D. N., Benitez, V., & Kuwabara, M. (2011). Inhibition and adjective learning in bilingual and monolingual children. Frontiers in Psychology, 2, 210. https://doi.org/10.3389/fpsyg.2011.00210 [Article] [PubMed]×
Zobl, H. (1980). The formal and developmental selectivity of LI influence on L2 acquisition. Language Learning, 30(1), 43–57. https://doi.org/10.1111/j.1467-1770.1980.tb00150.x [Article]
Zobl, H. (1980). The formal and developmental selectivity of LI influence on L2 acquisition. Language Learning, 30(1), 43–57. https://doi.org/10.1111/j.1467-1770.1980.tb00150.x [Article] ×
Appendix
Stimuli characteristics, median, mean, standard deviation, definitions, and visual referent code number.
Novel words Word category PP PND PN PSF BF MSR Definitions Code number
fɒni:s HP-HD High 4 0.34 0.02 1 fɒni:s can sing beautifully and is known as the heavenly singer. Set 3 A12-C.25
pɪkɪn HP-HD High 7 0.45 0.02 1 pɪkɪn lives on Mars and owns a big crystal house. Set 2 A12-C.25
rɛdɪn HP-HD High 4 0.36 0.03 1 rɛdɪn can turn stones into diamonds. Set 2 A14-C.25
mi:lit HP-HD High 4 0.36 0.03 1 mi:lit creates water and rain in the sky. Set 1 G04-C.25
dɪtaɪz HP-HD High 7 0.32 0.04 1 dɪtaɪz can create thunder and lightning from his eyes. Set 1 G02-C.25
mɪgæk LP-LD Low 0 0.22 0.01 1 mɪgæk owns a powerful elephant which has seven heads. Set 1 F02-C.25
lɛvrəʊ LP-LD Low 0 0.17 0 1 lɛvrəʊ enjoys the beauty of the shining stars. Set 1 A03-C.25
mi:ɒp LP-LD Low 0 0.13 0 2 mi:ɒp is interested in paintings and fine arts. Set 2 C16-C.25
trɒgɛm LP-LD Low 0 0.20 0.02 1 trɒgɛm travels to earth in a carriage pulled by five horses. Set 1 A01-C.25
tæbɛk LP-LD Low 0 0.18 0.01 1 tæbɛk likes flowers and owns a beautiful garden. Set 1 K01-C.25
tɪsɪv HP-LD High Low 1 0.40 0.03 1 tɪsɪv is the eldest alien and the head of the alien family. Set 2 C10-C.25
dɪmtɛz HP-LD High Low 0 0.40 0.04 1 dɪmtɛz enjoys chocolate, milk, and sweets very much. Set 2 A11-C.25
tʃɒnid HP-LD High Low 0 0.33 0.02 1 tʃɒnid is very knowledgeable and is regarded as an experienced teacher. Set 3 D02-C.25
sɛna:k HP-LD High Low 0 0.31 0.02 1 sɛna:k is very good at healing diseases. Set 3 B12-C.25
sɪsrɛt HP-LD High Low 0 0.46 0.03 1 sɪsrɛt is fond of traveling and driving around space. Set 1 D02-C.25
HP-HD (M) 5.2 0.37 0.03
LP-LD (M) 0 0.18 0.01
HP-LD (M) 0.2 0.38 0.03
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.×
Novel words Word category PP PND PN PSF BF MSR Definitions Code number
fɒni:s HP-HD High 4 0.34 0.02 1 fɒni:s can sing beautifully and is known as the heavenly singer. Set 3 A12-C.25
pɪkɪn HP-HD High 7 0.45 0.02 1 pɪkɪn lives on Mars and owns a big crystal house. Set 2 A12-C.25
rɛdɪn HP-HD High 4 0.36 0.03 1 rɛdɪn can turn stones into diamonds. Set 2 A14-C.25
mi:lit HP-HD High 4 0.36 0.03 1 mi:lit creates water and rain in the sky. Set 1 G04-C.25
dɪtaɪz HP-HD High 7 0.32 0.04 1 dɪtaɪz can create thunder and lightning from his eyes. Set 1 G02-C.25
mɪgæk LP-LD Low 0 0.22 0.01 1 mɪgæk owns a powerful elephant which has seven heads. Set 1 F02-C.25
lɛvrəʊ LP-LD Low 0 0.17 0 1 lɛvrəʊ enjoys the beauty of the shining stars. Set 1 A03-C.25
mi:ɒp LP-LD Low 0 0.13 0 2 mi:ɒp is interested in paintings and fine arts. Set 2 C16-C.25
trɒgɛm LP-LD Low 0 0.20 0.02 1 trɒgɛm travels to earth in a carriage pulled by five horses. Set 1 A01-C.25
tæbɛk LP-LD Low 0 0.18 0.01 1 tæbɛk likes flowers and owns a beautiful garden. Set 1 K01-C.25
tɪsɪv HP-LD High Low 1 0.40 0.03 1 tɪsɪv is the eldest alien and the head of the alien family. Set 2 C10-C.25
dɪmtɛz HP-LD High Low 0 0.40 0.04 1 dɪmtɛz enjoys chocolate, milk, and sweets very much. Set 2 A11-C.25
tʃɒnid HP-LD High Low 0 0.33 0.02 1 tʃɒnid is very knowledgeable and is regarded as an experienced teacher. Set 3 D02-C.25
sɛna:k HP-LD High Low 0 0.31 0.02 1 sɛna:k is very good at healing diseases. Set 3 B12-C.25
sɪsrɛt HP-LD High Low 0 0.46 0.03 1 sɪsrɛt is fond of traveling and driving around space. Set 1 D02-C.25
HP-HD (M) 5.2 0.37 0.03
LP-LD (M) 0 0.18 0.01
HP-LD (M) 0.2 0.38 0.03
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.×
×
Footnote
1 In our calculation of bigram and trigram frequencies and phonological neighborhood density, we excluded any item for which the headword was a simple contraction, complex contraction, letter, or abbreviation, any item with a spelling containing a nonalphabetic character (e.g., hyphen, space), or a capital letter in a position other than the first. This left 65,030 unique pronunciations in the database. In cases where a pronunciation occurred in multiple entries, the frequencies were summed to get a single total frequency for that specific pronunciation.
In our calculation of bigram and trigram frequencies and phonological neighborhood density, we excluded any item for which the headword was a simple contraction, complex contraction, letter, or abbreviation, any item with a spelling containing a nonalphabetic character (e.g., hyphen, space), or a capital letter in a position other than the first. This left 65,030 unique pronunciations in the database. In cases where a pronunciation occurred in multiple entries, the frequencies were summed to get a single total frequency for that specific pronunciation.×
Figure 1.

Effects of phonotactic probability (high = stimuli with high phonotactic probability, low = stimuli with low phonotactic probability).

 Effects of phonotactic probability (high = stimuli with high phonotactic probability, low = stimuli with low phonotactic probability).
Figure 1.

Effects of phonotactic probability (high = stimuli with high phonotactic probability, low = stimuli with low phonotactic probability).

×
Figure 2.

Effects of phonological neighborhood density (high = stimuli with high phonological neighborhood density, low = stimuli with low phonological neighborhood density).

 Effects of phonological neighborhood density (high = stimuli with high phonological neighborhood density, low = stimuli with low phonological neighborhood density).
Figure 2.

Effects of phonological neighborhood density (high = stimuli with high phonological neighborhood density, low = stimuli with low phonological neighborhood density).

×
Figure 3.

Correlated comparisons (high = words with high phonotactic probability and phonological neighborhood density, low = words with low phonotactic probability and phonological neighborhood density).

 Correlated comparisons (high = words with high phonotactic probability and phonological neighborhood density, low = words with low phonotactic probability and phonological neighborhood density).
Figure 3.

Correlated comparisons (high = words with high phonotactic probability and phonological neighborhood density, low = words with low phonotactic probability and phonological neighborhood density).

×
Table 1. Demographic and background data of participants.
Demographic and background data of participants.×
Demographic variables Monolinguals Bilinguals p Value a
Age (years) 21.47 (0.938) 21.55 (0.998) .807
Nonword repetition b 70.65 (6.70) 69.55 (7.39) .625
Digit span c 68.10 (9.91) 69.15 (10.17) .746
L2 acquisition age (speaking) 6.15 (0.812)
Proficiency ratings d
Speaking 3.05 (0.394)
Listening 3.40 (0.502)
Reading 3.35 (0.489)
Writing 3.28 (0.487)
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.×
a Significance of a two-sample t test (two-tailed).
Significance of a two-sample t test (two-tailed).×
b Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
c Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
d Proficiency ratings from 0 = not proficient to 4 = highly proficient.
Proficiency ratings from 0 = not proficient to 4 = highly proficient.×
Table 1. Demographic and background data of participants.
Demographic and background data of participants.×
Demographic variables Monolinguals Bilinguals p Value a
Age (years) 21.47 (0.938) 21.55 (0.998) .807
Nonword repetition b 70.65 (6.70) 69.55 (7.39) .625
Digit span c 68.10 (9.91) 69.15 (10.17) .746
L2 acquisition age (speaking) 6.15 (0.812)
Proficiency ratings d
Speaking 3.05 (0.394)
Listening 3.40 (0.502)
Reading 3.35 (0.489)
Writing 3.28 (0.487)
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.
Note. Means and standard deviations (in parentheses). N = 20 for both participant groups. Em dashes indicate data not available. L2 = second language.×
a Significance of a two-sample t test (two-tailed).
Significance of a two-sample t test (two-tailed).×
b Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Nonword repetition (n = 18) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
c Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).
Digit span (n = 21) percentile scores (a subtest of Comprehensive Test of Phonological Processing).×
d Proficiency ratings from 0 = not proficient to 4 = highly proficient.
Proficiency ratings from 0 = not proficient to 4 = highly proficient.×
×
Table 2. Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.
Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.×
Effect Degrees of freedom F p ηp 2
Analysis 1: phonotactic probability
 Group (monolingual vs. bilingual) 1, 38 15.70 <.001* .292
 Time (immediate vs. delayed) 1, 38 35.10 <.001* .480
 PP (high vs. low) 1, 38 18.41 <.001* .326
 Time × Group 1, 38 1.40 .243
 Time × PP 1, 38 1.96 .170
 Group × PP 1, 38 0.38 .543
 Time × Group × PP 1, 38 0.04 .843
Analysis 2: phonological neighborhood density
 Group (monolingual vs. bilingual) 1, 38 12.51 .001* .248
 Time (immediate vs. delayed) 1, 38 43.41 <.001* .533
 PND (high vs. low) 1, 38 10.72 .002* .220
 Time × Group 1, 38 0.89 .353
 Time × PND 1, 38 3.22 .080
 Group × PND 1, 38 2.55 .118
 Time × Group × PND 1, 38 0.01 .932
Analysis 3: correlated comparisons
 Group (monolingual vs. bilingual) 1, 38 11.31 .002* .229
 Time (immediate vs. delayed) 1, 38 85.74 <.001* .693
 Condition (HiPP/PND vs. LowPP/PND) 1, 38 0.005 .942
 Time × Group 1, 38 1.75 .194
 Time × Condition 1, 38 0.498 .485
 Group × Condition 1, 38 1.20 .279
 Time × Group × Condition 1, 38 0.091 .764
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.×
* p < .01, ηp 2 (effect size) = partial eta square.
p < .01, ηp 2 (effect size) = partial eta square.×
Table 2. Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.
Results of phonotactic probability, phonological neighborhood density, and correlated comparisons.×
Effect Degrees of freedom F p ηp 2
Analysis 1: phonotactic probability
 Group (monolingual vs. bilingual) 1, 38 15.70 <.001* .292
 Time (immediate vs. delayed) 1, 38 35.10 <.001* .480
 PP (high vs. low) 1, 38 18.41 <.001* .326
 Time × Group 1, 38 1.40 .243
 Time × PP 1, 38 1.96 .170
 Group × PP 1, 38 0.38 .543
 Time × Group × PP 1, 38 0.04 .843
Analysis 2: phonological neighborhood density
 Group (monolingual vs. bilingual) 1, 38 12.51 .001* .248
 Time (immediate vs. delayed) 1, 38 43.41 <.001* .533
 PND (high vs. low) 1, 38 10.72 .002* .220
 Time × Group 1, 38 0.89 .353
 Time × PND 1, 38 3.22 .080
 Group × PND 1, 38 2.55 .118
 Time × Group × PND 1, 38 0.01 .932
Analysis 3: correlated comparisons
 Group (monolingual vs. bilingual) 1, 38 11.31 .002* .229
 Time (immediate vs. delayed) 1, 38 85.74 <.001* .693
 Condition (HiPP/PND vs. LowPP/PND) 1, 38 0.005 .942
 Time × Group 1, 38 1.75 .194
 Time × Condition 1, 38 0.498 .485
 Group × Condition 1, 38 1.20 .279
 Time × Group × Condition 1, 38 0.091 .764
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.
Note. Em dashes indicate data not available. PP = phonotactic probability; PND = phonological neighborhood density; Hi = high; HiPP/PND = high phonotactic probability/phonological neighborhood density.×
* p < .01, ηp 2 (effect size) = partial eta square.
p < .01, ηp 2 (effect size) = partial eta square.×
×
Novel words Word category PP PND PN PSF BF MSR Definitions Code number
fɒni:s HP-HD High 4 0.34 0.02 1 fɒni:s can sing beautifully and is known as the heavenly singer. Set 3 A12-C.25
pɪkɪn HP-HD High 7 0.45 0.02 1 pɪkɪn lives on Mars and owns a big crystal house. Set 2 A12-C.25
rɛdɪn HP-HD High 4 0.36 0.03 1 rɛdɪn can turn stones into diamonds. Set 2 A14-C.25
mi:lit HP-HD High 4 0.36 0.03 1 mi:lit creates water and rain in the sky. Set 1 G04-C.25
dɪtaɪz HP-HD High 7 0.32 0.04 1 dɪtaɪz can create thunder and lightning from his eyes. Set 1 G02-C.25
mɪgæk LP-LD Low 0 0.22 0.01 1 mɪgæk owns a powerful elephant which has seven heads. Set 1 F02-C.25
lɛvrəʊ LP-LD Low 0 0.17 0 1 lɛvrəʊ enjoys the beauty of the shining stars. Set 1 A03-C.25
mi:ɒp LP-LD Low 0 0.13 0 2 mi:ɒp is interested in paintings and fine arts. Set 2 C16-C.25
trɒgɛm LP-LD Low 0 0.20 0.02 1 trɒgɛm travels to earth in a carriage pulled by five horses. Set 1 A01-C.25
tæbɛk LP-LD Low 0 0.18 0.01 1 tæbɛk likes flowers and owns a beautiful garden. Set 1 K01-C.25
tɪsɪv HP-LD High Low 1 0.40 0.03 1 tɪsɪv is the eldest alien and the head of the alien family. Set 2 C10-C.25
dɪmtɛz HP-LD High Low 0 0.40 0.04 1 dɪmtɛz enjoys chocolate, milk, and sweets very much. Set 2 A11-C.25
tʃɒnid HP-LD High Low 0 0.33 0.02 1 tʃɒnid is very knowledgeable and is regarded as an experienced teacher. Set 3 D02-C.25
sɛna:k HP-LD High Low 0 0.31 0.02 1 sɛna:k is very good at healing diseases. Set 3 B12-C.25
sɪsrɛt HP-LD High Low 0 0.46 0.03 1 sɪsrɛt is fond of traveling and driving around space. Set 1 D02-C.25
HP-HD (M) 5.2 0.37 0.03
LP-LD (M) 0 0.18 0.01
HP-LD (M) 0.2 0.38 0.03
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.×
Novel words Word category PP PND PN PSF BF MSR Definitions Code number
fɒni:s HP-HD High 4 0.34 0.02 1 fɒni:s can sing beautifully and is known as the heavenly singer. Set 3 A12-C.25
pɪkɪn HP-HD High 7 0.45 0.02 1 pɪkɪn lives on Mars and owns a big crystal house. Set 2 A12-C.25
rɛdɪn HP-HD High 4 0.36 0.03 1 rɛdɪn can turn stones into diamonds. Set 2 A14-C.25
mi:lit HP-HD High 4 0.36 0.03 1 mi:lit creates water and rain in the sky. Set 1 G04-C.25
dɪtaɪz HP-HD High 7 0.32 0.04 1 dɪtaɪz can create thunder and lightning from his eyes. Set 1 G02-C.25
mɪgæk LP-LD Low 0 0.22 0.01 1 mɪgæk owns a powerful elephant which has seven heads. Set 1 F02-C.25
lɛvrəʊ LP-LD Low 0 0.17 0 1 lɛvrəʊ enjoys the beauty of the shining stars. Set 1 A03-C.25
mi:ɒp LP-LD Low 0 0.13 0 2 mi:ɒp is interested in paintings and fine arts. Set 2 C16-C.25
trɒgɛm LP-LD Low 0 0.20 0.02 1 trɒgɛm travels to earth in a carriage pulled by five horses. Set 1 A01-C.25
tæbɛk LP-LD Low 0 0.18 0.01 1 tæbɛk likes flowers and owns a beautiful garden. Set 1 K01-C.25
tɪsɪv HP-LD High Low 1 0.40 0.03 1 tɪsɪv is the eldest alien and the head of the alien family. Set 2 C10-C.25
dɪmtɛz HP-LD High Low 0 0.40 0.04 1 dɪmtɛz enjoys chocolate, milk, and sweets very much. Set 2 A11-C.25
tʃɒnid HP-LD High Low 0 0.33 0.02 1 tʃɒnid is very knowledgeable and is regarded as an experienced teacher. Set 3 D02-C.25
sɛna:k HP-LD High Low 0 0.31 0.02 1 sɛna:k is very good at healing diseases. Set 3 B12-C.25
sɪsrɛt HP-LD High Low 0 0.46 0.03 1 sɪsrɛt is fond of traveling and driving around space. Set 1 D02-C.25
HP-HD (M) 5.2 0.37 0.03
LP-LD (M) 0 0.18 0.01
HP-LD (M) 0.2 0.38 0.03
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.
Note. Summed positional segment frequency (minimum = 0.002, maximum = 0.805, median = 0.303, M = 0.302, SD = 0.014). Summed biphone frequency (minimum = 0, maximum = 0.137, median = 0.023, M = 0.025, SD = 0.017). Mean in parentheses indicates mean values for PN, PSF, and BF for all three novel word categories. Code number indicates the specific visual referent (alien) number corresponding to each item (all color images; Gupta et al. 2004). Link to the visual referents database http://link.springer.com/article/10.3758/BF03206540. PP = items included in the phonotactic probability manipulation; PND = items included in the phonological neighborhood density manipulation; PN = phonological neighborhood (minimum = 0, maximum = 28, median = 1, M = 2.86, SD = 3.81); PSF = positional segment frequency; BF = biphone frequency; MSR = Mandarin Similarity Rating for novel word (1 = no resemblance to Mandarin, 4 = close resemblance to Mandarin); HP-HD = high phonotactic probability–high phonological neighborhood density; LP-LD = low phonotactic probability–low phonological neighborhood density; HP-LD = high phonotactic probability–low phonological neighborhood density.×
×