Neighborhood frequency effects in late bilingual phonological neighborhoods

Neighborhood frequency effects in late bilingual phonological neighborhoods Eva Maria Luef a,b,* a Charles University, Department of Linguistics, English Language and ELT Methodology, n´ am. Jana Palacha 2, 116 38, Prague 1, Czech Republic b University of Hamburg, Institute of English and American Studies, Von-Melle-Park 6, 20146, Hamburg, Germany ARTICLE INFO Keywords: Late bilingualism Neighborhood frequency effect Phonological network Spoken word recognition Bilingual lexical processing ABSTRACT The bilingual mental lexicon is understood as a unified system containing word forms from multiple languages. Previous studies have described how cross-language phonological similarity influences bilingual lexical processing, demonstrating that lexical activation spreads both within and across languages at the segmental level. The significance of cross-linguistic phonological neighbors in bilingual speech processing is well documented, but less is known about frequency effects emerging from the cross-language neighboring words. In monolingual spoken word recognition, the ‘neighborhood frequency effect’ suggests that higher-frequency neighbors absorb more activation and/or inhibit lower-frequency neighbors within a phonological neighborhood, potentially affecting recognition latencies. This study examines whether this effect extends across languages in phonological neighborhoods of late German-English bilinguals. Results reveal that lexical frequency rates of German neighbors influence response times for English target words in an English lexical decision task. This finding supports a fully integrated mental lexicon in late bilinguals (i.e., second language learners) and highlights the role of lexical frequency in cross-language lexical processing. 1. Introduction Spoken words tend to be recognized faster when lexical competition is reduced by having fewer phonologically similar words in a speaker's mental lexicon (Luce & Pisoni, 1998; Vitevitch, 2007; Vitevitch et al., 2018; Vitevitch & Luce, 2016; Yates & Dickinson, 2023; Ziegler et al., 2003). This effect has primarily been studied from the viewpoint of phonological neighborhood density, a metric that considers the number of phonologically close words that exist in a lexicon. Traditionally, psycholinguistics has defined phonological neighbors as words differing by a single phonological segment, the Levenshtein distance (Landauer & Streeter, 1973; Levenshtein, 1966; Vitevitch & Luce, 2016), even though alternative quantifications of phonological neighborhoods can be found in the literature (Kapatsinski, 2006; Siew & Castro, 2023; Vitevitch & Castro, 2015). Within the framework of activation spreading (Luce et al., 2000; Vitevitch & Luce, 1998; Yates & Dickinson, 2023), it is theorized that phonological overlaps between words lead to co-activation of shared segments. When a word has numerous phonological neighbors, activation is distributed among them, creating a delay in the recognition of the target word within the neighborhood (Vitevitch & Luce, 2016). A higher density of phonological neighborhoods reduces the activation allocated to each individual word and leads to lower overall activation of each word, including the target word itself. While this effect can be seen as passive activation sharing across lexical candidates, theoretical accounts of active lexical competition, where an active item inhibits its neighbors, are also frequently encountered (Goldinger et al., 1989; Vitevitch & Luce, 2016; Yates & Dickinson, 2023). Chen and Mirman (2012) model activation spreading as a trade-off between inhibitory and facilitative links between lexical neighbors, where strongly activated words have an inhibitory effect and weakly activated neighbors have a facilitative effect on word recognition. Activation spreading is then dependent on how much inhibition and facilitation there is within a neighborhood. The influence of phonological neighborhood density on word recognition is well-supported by monolingual studies (see Yates & Dickinson, 2023, for a review); similar effects have been observed in multilingual studies (de Bot & Batyi, 2022; Kroll & Ma, 2017). Bilingual speakers, especially those using typologically related languages with significant word form similarities (e.g., Germanic languages, such as English, German, Dutch), experience increased activation spreading in bilingual phonological neighborhoods, as words from multiple languages become lexical competitors (e.g., Canseco-Gonzalez et al., 2010; Dijkstra et al., 2010; Frances et al., 2021; Lagrou et al., 2011; Lemh¨ ofer * Charles University, Department of Linguistics, English Language and ELT Methodology, n´ am. Jana Palacha 2, 116 38, Prague 1, Czech Republic. E-mail address: [email protected]. Contents lists available at ScienceDirect Acta Psychologica journal homepage: www.elsevier.com/locate/actpsy https://doi.org/10.1016/j.actpsy.2025.105863 Received 26 February 2025; Received in revised form 10 October 2025; Accepted 27 October 2025 Acta Psychologica 261 (2025) 105863 Available online 5 November 2025 0001-6918/© 2025 The Author. Published by Elsevier B.V. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). et al., 2008; Luef, 2025a; Marian et al., 2008; Schulpen et al., 2003). Theoretical accounts of cross-linguistic activation spreading in spoken word recognition (Lewy & Grosjean, 2008; Schwartz & Kroll, 2007; Shook & Marian, 2013) predict that lexical activation propagates through sublexical features and across languages of multilingual individuals, ultimately leading to competition effects at the lexical level (Vitevitch & Luce, 1998, 1999). This means that cross-language phonological neighbors, such as English climb [klaɪm] and German “Leim”/ Engl. glue [laɪm], have a similar effect to same-language neighbors within their neighborhoods (Luef, 2025a). The impact of activation spreading within phonological neighborhoods is modulated by the frequency of the segments and words (Luce & Large, 2001; Vitevitch & Luce, 1999), with more frequently occurring items showing higher degrees of activation. Whether lexical frequency is defined according to resting activation level (Dell, 1986; Luce & Pisoni, 1998), or whether it is seen as connection weights between a lexical representation and its segments (Chen & Mirman, 2012), higher frequency words are activated more strongly and faster, leading to shorter response latencies in word recognition. The higher baseline activation of frequent words in linguistic memory leads to less cognitive effort for their retrieval (Belke et al., 2005; Borowsky & Masson, 1999). Lexical frequency effects are commonly explained by learning effects, assuming that learning increases activation levels of word representations (Brysbaert, Mandera, & Keuleers, 2017; Coltheart et al., 2001). Alternatively, frequency effects may be considered as confounds of other lexical features such as word length and age of acquisition (Brysbaert, Mandera, & Keuleers, 2017), as these variables are typically highly correlated in first (Brysbaert & Biemiller, 2016; N. C. Ellis, 2002; Frauenfelder et al., 1993; Storkel et al., 2006) and second (Luef, 2023) languages. While lexical frequency plays a central role in models of spoken word recognition (e.g., Neighborhood Activation Model. TRACE), with its effects on word recognition well attested in different languages (e.g., Brysbaert et al., 2011, 2019; Brysbaert & New, 2009; Ferrand et al., 2010), little attention has been paid to frequency effects arising not from the target word but the neighboring words in a phonological neighborhood. Frequencies of neighboring words can have varying degrees of influence on word recognition. According to a passive lexical competition account in phonological neighborhoods, high-frequency words are suggested to constitute stronger competitors as they absorb more of the present activation (Luce & Pisoni, 1998). This diverts activation away from lower-frequency words. Depending on the frequency rate of a target word, it may be affected in different ways: if neighbors are of higher lexical frequencies than the target word, they take more of the activation, resulting in slower response latencies to the target; if target words show the highest frequency rates, they are not predicted to be delayed. Alternatively, higher-frequency words may inhibit activation of lower-frequency words, resulting in higher-frequency words monopolizing a larger share of the activation within a neighborhood. Regardless of the activation mechanics, the “neighborhood frequency effect” manifests as the influence of the frequency of phonological neighbors on the speed and accuracy with which a spoken word is recognized (Andrews, 1989; Frauenfelder et al., 1993). Neighborhood frequency effects play a central role in various cognitive linguistic tasks, including perceptual identification (Dirks et al., 2001; Pisoni et al., 1985), where word identification is typically superior when target words themselves are of higher lexical frequency but the neighborhood consists of low-frequency words. Neighborhood frequency effects are also crucial determiners for word memory, and in particular short-term memory. High-frequency words in sparse, lowfrequency neighborhoods tend to be more easily recalled than lowfrequency words in dense, high-frequency neighborhoods (Goldinger et al., 1991; Roodenrys et al., 2002), demonstrating an active role of neighborhood frequency on lexical memory. Furthermore, neighborhood frequency effects emerge in picture naming (Baus et al., 2008; Vitevitch & Sommers, 2003, but note the difference in the suggested direction of the effect), speech errors (Vitevitch, 1997), as well as speech production in first language acquisition (Newman & German, 2002). Further underscoring the importance of the neighborhood frequency effect in word recognition are studies on phonological false memories (Roediger & McDermott, 1995; Sommers & Lewis, 1999). When asked to recall a list of phonologically neighboring words, higher frequency neighbors tend to be mis-selected (e.g., falsely remembering “line” instead of “lime”). In sum, neighborhood frequency effects influence phono-lexical processing on various dimensions, making them an important aspect of lexical cognition. The majority of existing studies on the neighborhood frequency effect in lexical processing investigated speech production and visual word recognition. Studies on spoken word recognition include Luce and Pisoni (1998) who found that higher-frequency neighbors raise lexical competition because they raise the overall level of activity within a neighborhood, where multiple similar words are activated at the same time, making it harder to settle on the correct target word. Thus, a highfrequency neighborhood causes recognition delays in target words. Similar findings for French were reported by Dufour and Frauenfelder (2010). Furthermore, accuracy of word judgment seems to be affected negatively by the presence of high-frequency neighbors in Spanish phonological neighborhoods (Cervera-Crespo & Gonzalez-Alvarez, 2019). Not all studies of auditory word recognition have come to the same conclusion (e.g., Forster & Shen, 1996; Hameau et al., 2021; Sears et al., 1995, 2006; Wagenmakers et al., 2008), and in the word production domain, the opposite findings are well established: high-frequency neighbors actually facilitate target word retrieval in English and Spanish (e.g., Vitevitch & Luce, 2016; Vitevitch & Sommers, 2003). This reflects the different demands of the two modalities perception and production. Concerning bilingual word recognition, no studies to date have specifically focused on neighborhood frequency effects resulting from cross-language neighbors. The present study was designed to shed light on this issue and determine whether spoken word recognition of English-as-a-second-language (or “L2”) is influenced by German-as-afirst-language (or “L1”) in late bilinguals. The findings will provide cues as to the complexity of lexical processing in speakers of multiple languages. 1.1. Patterns of bilingual phonological neighborhoods Since activation spreads along phonological segments, phonological similarity is a crucial consideration (Gahl & Strand, 2016; Luef, 2023, 2025a). It has been proposed that closer neighbors constitute stronger competitors to target words than more distant neighbors (e.g., Chen & Mirman, 2012; Mirman & Kittredge, 2010), attributing phonological similarity major influence on spoken word recognition in monoand bilingual studies. The one-segment neighbor metric between word forms (i.e., Levenshtein distance) can be quantified based on more nuanced measures that weigh individual phonological features (e.g., frontness, nasality, etc.; see section 3.6 for more details), resulting in phonological neighborhood patterns that consider phonological closeness between word forms that differ by one phonological segment. A key factor in frequency effects is the overall frequency distribution within a phonological neighborhood. If the target word has the highest lexical frequency rate, the neighborhood is predicted to have less influence on recognition compared to cases where neighbors have higher lexical frequencies. Figs. 1 and 2 illustrate different frequency distributions within bilingual phonological neighborhoods, which are expected to lead to different response latencies for the English target words. The recognition of both “hint” and “fly” is assumed to be influenced by the lexical frequency rates of their respective neighbors and the phonological similarity of word forms, regardless of the language. Because “hint” is a low-frequency word with English neighbors that are both lower in frequency and less phonologically similar (indicated E.M. Luef Acta Psychologica 261 (2025) 105863 2 through thinner links), it is expected to be more affected by highfrequency, phonologically closer German neighbors. In contrast, “fly” is phonologically more distant from its highest-frequency neighbor (German “frei” / Engl. free) and is therefore expected to be more strongly influenced by higher-frequency and phonologically closer English neighbors such as “lie” and “flight”. Bilingual phonological neighborhoods can vary in complexity based on phonological similarity and frequency patterns in each languagespecific section of a neighborhood. Recent methodological advances in lexical sciences have introduced the innovative approach of studying the mental lexicon as a lexical network, emphasizing the interconnected nature of word forms (see Fig. 3). The position of target words within these lexical networks have been shown to influence lexical processing. This includes factors such as density of a network region (Siew & Vitevitch, 2016), the interconnectedness of the phonological neighborhood (Chan & Vitevitch, 2009; Luef, 2025a; Yates, 2013), and the key structural positions held by target words in a lexical network (Goldstein & Vitevitch, 2017; Vitevitch & Goldstein, 2014). In bilingual phonological networks, such as the L1-German-L2-English one shown in Fig. 3, the frequency distribution of neighbors in each individual language, as well as within the integrated bilingual lexicon, is a critical factor to consider. Fig. 3 illustrates the variability of phonological neighborhoods (and their extended neighborhoods) in terms of the number of neighbors, their phonological similarity, and lexical frequency (which corresponds to node size in Fig. 3). It is evident that not all neighbors have the potential to exert the same influence, and lexical processing largely depends on the specific phonological and frequency composition of a given neighborhood. The spread of co-activation can be predicted by the number of neighbors, phonological similarity between target words and neighbors, neighborhood frequency relative to target word frequency, and the integration of English and German word forms in bilingual lexica. Phonological network science provides a valuable framework for controlling these variables, thereby aiding the study of spoken word recognition. Fig. 1. A representation of the phonological neighborhood of English “hint”. Dark nodes represent German phonological neighbors; light-gray nodes represent English phonological neighbors. Node size corresponds to lexical frequency rate. The L2-English neighbors (CEFR proficiency level C1) are generally of lower lexical frequency, while the cross-language L1-German neighborhood contains various high-frequency words as well as closer phonological neighbors (indicated via line strength). Fig. 2. A representation of the phonological neighborhood of English “fly”. Dark nodes represent German phonological neighbors; light-gray nodes represent English phonological neighbors. Node size corresponds to lexical frequency rate. The L2-English neighbors (CEFR proficiency level C1) and the L2-English neighbors are of similar lexical frequency as the L1-German neighbors. The higher-frequency English neighbors are phonologically similar to the target (indicated via line strength). Fig. 3. Partial bilingual phonological network. Light-gray nodes represent L2English words (CEFR proficiency level C1), dark nodes represent L1-German words, white nodes represent word forms shared by both languages. Node size corresponds to lexical frequency rate. Link strength corresponds to phonological similarity. E.M. Luef Acta Psychologica 261 (2025) 105863 3 This study aims to broaden the scope of research on the neighborhood frequency effect by studying it from a bilingual perspective. The research question explored here concerns the influence of lexical frequency of German phonological neighbors on L2-English target words. Are German neighbors and their lexical frequency rates predictive for the retrieval of English target words? Based on the assumption that German neighbors play a significant role in bilingual word recognition, it is hypothesized that word recognition will be delayed by highfrequency German neighbors within the neighborhood, due to their effect on activation spreading. Understanding the interactions between languages in multilingual speakers allows crucial insights into lexical cognition, particularly given the growing proportion of multilingual individuals in European societies and beyond (Siemund, 2023). Dynamics of activation spreading in bilingual phonological neighborhoods serve as key evidence for an integrated mental lexicon in multilinguals. If bilingual lexical processing is driven by activation strength and the depth of memory representations shaped by lexical frequency in the dominant (first) language, this could offer new perspectives on activation spreading in second language cognition. Such findings can contribute to models of bilingual lexical organization. 2. Data availability The data and code for replication of the results can be found in the Supplementary Material and on the Zenodo repository under the link: https://doi.org/10.5281/zenodo.17431193. 3. Methods German-English late bilinguals were tested in auditory word recognition in various experimental conditions that varied the frequencies of the German neighborhoods of the English target words. Based on previous monolingual studies of neighborhood frequency effects in spoken word recognition, it was expected that higher-frequency neighbors that are phonologically more similar to the target words would lead to longer response latencies as more lexical activation is incorporated by those neighbors. In a first step, a bilingual phonological lexicon that combines L1German and L2-English was constructed. Under consideration of various network metrics, a number of target words were selected and controlled for additional factors (see section 3.6.). As a second step, participants were tested in their knowledge of all target words as well as all neighboring words that were identified in the network model. Lastly, the target words were tested in an auditory lexical decision task, which measured accuracy of word judgments as well as reaction times to the English target words. 3.1. Phonological networks The L1-German vocabulary was based on data obtained from the SUBTLEX-DE Corpus (Brysbaert et al., 2011), a corpus of German subtitles in film and TV, that is available on the Clearpond database (Marian et al., 2012). It is considered one of the best frequency measures for undergraduate student speech (Brysbaert & New, 2009). The English vocabulary was obtained from the English Vocabulary Profile database, which is a collection of L2English vocabulary based on the Cambridge Learner Corpus, the largest corpus of English as a second language (Capel, 2015). Vocabulary corresponding to proficiency level CEFR-C1 was used (see below for a description of CEFR proficiency levels). In both vocabularies, proper nouns were removed and lemmatization was applied to remove the majority of inflections. Only extremely highfrequency inflected forms were retained, such as inflections of the English words “be” and “have” and the German correspondences “sein” and “haben” and participle II forms which can function as adjectives (German “gemacht”/ Engl. “done”). German inflections related to grammatical gender of nouns (e.g., “Arzt-¨ Arztin”/ Engl. male doctorfemale doctor) were also retained. Clitics were split into their constituent parts (e.g., English “it's” ➔ “it”, “is”; German “aufs” ➔ “auf”, “das”/ Engl. “on the”). In German, infrequent loan words and compound nouns consisting of English and German words (e.g., “Actionszene”/ Engl. “action scene”) were removed. Additionally, specific medical vocabulary and abbreviations (e.g., “Dino”/ Engl. “dinosaur”) were excluded from the data, as well as swear words and numbers larger than 12 and not multiples of 10. Lemmatized German words that were not contained in the SUBTLEX-DE corpus were removed from the dataset (N =414 words, 3.7.% of the German vocabulary). This was necessary because lexical frequency rates of the German words were obtained from that corpus (i.e., via the Clearpond database). All lemmatized English word forms of the C1English vocabulary could be found in the English Clearpond database. The L1-German vocabulary consists of 14,546 words; the C1-English vocabulary contains 5387 words. The bilingual network resulted from a merging of the two vocabularies (also see Luef, 2025a). Next, phonetic transcriptions of the German and English words were copied from the Clearpond database in the form of CP-SAMPA transcriptions. The German transcriptions were manually checked and some were corrected. The transcriptions were then altered in various ways: dots between symbols were removed and transcriptions consisting of multiple symbols were replaced with single symbols (e.g., “E3” was replaced with “e”). Homophonous word forms were merged (e.g., German “bis-Biss”/ Engl. until-bite; English “see-sea”). Then, the transcribed German and English words were entered into an Oracle 12c BigData-Lite database (Bryla, 2015) and the function edit_distance of the package “utl_match” was used to compare all word forms to one another and identify one-segment phonological neighbors that differ by a single segment through either addition, deletion, or substitution (Landauer & Streeter, 1973). Table 1 shows an example result from the Oracle database. (See Table 2.) Separate analyses were conducted for the German vocabulary, the English vocabulary, and the bilingual German-English vocabulary, resulting in lists of phonological neighbors in the three vocabularies. Phonological networks were created for each of the three vocabularies, with the entirely of all words of a vocabulary being entered and all one-segment neighbors being linked to one another. The R package “igraph” (Csardi & Nepusz, 2006) and the network software “Gephi” (Bastian et al., 2009) were used for network statistics and visualizations. 3.2. Participants Sixty-nine first-language users of German (identification of 53 as female, 11 as male, and 5 as non-binary; age M =23.4 years, range 18–26) who are majoring in English Studies at a German university were recruited and gave written consent to participate in a vocabulary questionnaire and an auditory lexical decision task. Second language proficiency can be estimated based on the ‘Common European Framework of Reference’, which proposes six proficiency levels, starting with Table 1 String distances (edit distances) between the word “a” and other words in the L2English lexicon. Distance measurements are based on SAMPA transcriptions of the original phonetic transcriptions (IPA). String distances of “1” are treated as phonological neighbors. word IPA CPSAMPA tested against IPA CPSAMPA String distance a eɪeI able eɪbl eIbl 2 a eɪeI about əbaʊt 5baUt 5 a eɪeI above əbʌv 5bVv 4 a eɪeI ache eɪk eIk 1 a eɪeI add æd 1d 2 a eɪeI against əg ε nst 5gEnst 6 a eɪeI aim em eIm 1 E.M. Luef Acta Psychologica 261 (2025) 105863 4 the beginner levels A1 and A2, followed by intermediate levels B1 and B2, and the advanced levels C1 and C2 (see Council of Europe, 2018). All participants self-reported English proficiency at level C1 (advanced), which was the average proficiency required for admission to the university study program. None of them had any hearing or language disorders. Aside from German as their first and English as their second languages, if participants knew a third language, they only had basic knowledge of it. They received course credits for their participation. Data were collected between April and December 2024. The study was approved by the ethics review board of Charles University (reference number UKFF/46874/2023). 3.3. Vocabulary questionnaire 50 English words typically known by L2-English learners at lower and intermediate proficiency levels (A1, A2, B1, B2) were selected as stimuli (see Appendix Table A for an overview of the target words and their neighbors). They were controlled for a variety of factors (see section 3.6) but differed systematically in terms of their German neighborhood frequencies. The constraints that were imposed by the study design prevented more words from being eligible for testing. Requirements included target words to be of lower lexical frequency as such words are known to amplify neighborhood effects (Andrews, 1992; Luef, 2025a; Sears et al., 1995), while at the same time having at least 50 % German neighbors in the phonological neighborhood. As English and German neighborhood densities and frequencies are typically correlated (densities: r =0.5; frequencies: r =0.7), high-density/ frequency English neighborhoods are characterized by high-density/ frequency cross-language German neighborhoods. This meant that some combinations of low-frequency English target plus mediumto higher-frequent German neighbors were rarer instances to be found in the bilingual lexicon. This reduced the number of available stimuli. In addition, words from specific network density regions were chosen as targets, excluding other potential English targets. Please see section 3.6. for a description of each controlled factor. Knowledge of all involved German neighboring words was expected based on the first-language German proficiency of the participants. All participants were tested on their knowledge of the English words involved in the experiment, both target words and their phonological neighbors. In an online survey, participants were asked to rate their familiarity with the words, i.e., whether they knew a word (answer options: “I know the word”, “I have never heard the word”, “I am not sure”). The familiarity ratings were obtained from visually presented words. Words were removed from the data set when more than 30 % of participants stated that they did not know a word or were unsure. This was the case with one word, fowl, a neighbor of the target words owl and file. Fowl was taken out of neighborhood density and frequency calculations. The other target and neighboring words were well known by the participants. 80 % of the participants completed the experiment shortly before they took the vocabulary questionnaire, and in the large majority of cases both tasks were completed within one hour. It may have been possible that the 20 % who took the vocabulary questionnaire before the experiment were primed by the words they were exposed to in the questionnaire. 3.4. Pseudo-neighbors Second language perception may be influenced by the first language of listeners, leading to confusability of segmental contrasts on the lexical level (Bohn & Flege, 1992; Rocca et al., 2025). A well-researched aspect of this is the English [æ] - [ ε ] contrast that may be difficult to perceive for many German learners of English (Llompart, 2021; Llompart & Reinisch, 2020). The difficulty in distinguishing these vowels leads to additional phonological neighbors that are based on erroneous perception. For instance, the English words “beg” [b ε ɡ] and “bag” [bæɡ] may sound identical to German listeners at beginning but also intermediate and advanced levels (see Llompart, 2021), which may result in different perceptual phonological neighborhoods as compared to native speakers of English. Taking these perceptual accents into account can lead to a better estimation of a phonological neighborhood in second language learners (Amengual, 2016; Darcy et al., 2013; Weber & Cutler, 2004). This study took into account the following phonological features that were assumed to skew German L2-English learners' perception at proficiency level C1 (see Hickey, 2019): (a) the perception of [æ] as [ ε ], (b) voicing in word-initial sibilants (e.g., [zi:] for English “sea”), and (c) devoicing in word-final obstruents (e.g., [bɪk] for English “big”). Five target words were affected: egg, field, kind, loud, and tube. Using the phonetic transcriptions of the German-accented words [ ε k, fiːlt, kaɪnt, laʊt, tuːp] phonological neighbors in German and English were searched in the phonological networks of the languages and a total of 22 words were added to their respective neighborhoods. The [ ε k] neighborhood Table 2 Target words and neighborhood frequency statistics, including German, English, and pseudo neighbors in both languages. GERMAN ENGLISH Target word Sum of all frequencies within a neighborhood Standard deviation Sum of all frequencies within a neighborhood Standard deviation Belt 1391.3 249.4 664 164.4 Bike 1698.9 568.9 4821.7 1593.5 Climb 119.3 29.7 96.6 32.4 Clue 84.3 14.1 158.6 47.1 Cost 28.3 12.2 391 129.4 Dawn 12,402.4 2104.1 498.4 276.2 Dime 81.5 18.4 1989.6 1121.9 Down 2750.9 1337.4 321.8 114.9 East 352.4 164.9 466.4 130.7 Egg 21,498.6 6546.5 2357.1 693.3 False 182.5 122.1 125.7 58.3 Field 1139.9 455.7 186.5 48.8 File 1389.1 255.3 1536.8 210.3 Floor 14.8 3.3 3591.5 2059.7 Fly 203.8 59.1 221.1 48.1 Foot 37.7 10.7 5004.4 1632.2 Fuss 100.6 28.5 322 115.2 Guest 972.8 221.2 5810.7 1664 Guilt 833.8 211.6 168 12.9 Hall 424.9 100.5 2026 307 Help 161.3 28.6 513.1 260.3 Hint 1531.1 182.9 309.8 132.1 Hire 214.4 25.1 500.7 145.6 Hour_our 12,332.6 1819.5 3876.7 2228.3 How 826.9 100.4 4278.7 1325.5 Kind 1903.7 484.5 1318.8 415.6 Left 122 57.2 2460 1385 Lost 162.4 25.9 1095.5 289.9 Loud 476 75.6 4008.7 1452 Melt 1156.5 266.7 128.6 56.4 Mess 14,941.9 5611.9 3053.5 757.7 Mile 4423.8 628.4 8030 2333.9 Mint 6821.3 2077.9 494.4 175.5 Mouth 83.1 25.5 83.6 32.1 Now 2269.9 495.9 9088.6 2599.3 Ought 1800 354.4 29,609 6719.9 Out 11,970.4 1440.6 1790.2 359.4 Owl 11,045.5 1785.6 5020.1 1264.9 Piece_peace 412.8 80.5 152.3 40.9 Pile 1097.9 296.4 488.5 117.2 Shift 252.9 56.1 98.6 21.5 Shoot 3093.6 1040.1 658.7 105.9 Shout 371.4 84.5 3951.7 1918.4 Talk 946.7 221.8 239.1 118.1 Time 235.8 63.7 187.6 23.2 Town 46.1 8.9 1544.6 660.4 Tube 1306.1 506.9 8427 3737.9 Vet 5534 1101.8 7837.1 1590.7 Zoo 44,648.4 9430.8 59,371.1 15,064.6 E.M. Luef Acta Psychologica 261 (2025) 105863 5 received additional German “Ecke” (Engl. corner), German “Leck” (Engl. leak), Engl. “back”, “pack”, “lack”, sack”, “bag”. The [fiːlt] neighborhood received the additional German word “Feld” (Engl. field) and the English word “fault”. The [kaɪnt] neighborhood received additional German “Feind” (Engl. enemy), German “Kind” (Engl. child), and English “pint”. The [laʊt] neighborhood received additional German “Land” (Engl. land), “Last” (Engl. load), German “Haut” (Engl. skin), and English “route”, “shout”, “out”, and “doubt”. The [tuːp] neighborhood received additional German “taub” (Engl. deaf), German “Tipp” (Engl. tip), and English “soup”. Phonological similarities between the pseudo-neighbors and their neighborhoods were calculated and included in the neighborhood metrics. The frequencies of the neighborhoods were updated to include the individual frequencies of those pseudo-neighbors in the German and English Clearpond databases. 3.5. Lexical decision experiment An auditory lexical decision task was created with PsychoPy 2023.2.3 (Peirce et al., 2019). The presentation of stimulus words was randomized, and accuracy in word judgments as well as reaction time measurements to correct judgments were extracted for analysis. Audio stimuli of the 50 real words and 50 pseudo-words (or “non-words”) were obtained from the MALD database (Tucker et al., 2019), with all words being spoken by an American male speaker. His accent can be described as Midland or Western United States, possibly Colorado or California. His speech showed a clear cot-caught distinction and no traces of Northern Cities Vowel Shift or Southern phonetic features (as evaluated by ChatGPT based on a sample of 5 recordings). The pseudo-words differed from real English or German words by one segment. Words that could mistakenly be perceived as an accented English word by German learners of English were not selected as pseudowords (e.g., /ʧis/ which could be perceived as an acceptable pronunciation of “cheese” by German listeners). The experiment was made available online on the Pavlovia platform via a link that was given to the participants. Each participant was tested individually at their own computers; they were instructed to use headphones. Trials started with the presentation of a small cross on the screen for 500 ms. After that, target words and pseudo-words appeared in successive, randomized order and only once before participants had to decide (as quickly as possible) whether a word they heard is a real English word or a pseudo-word by pressing either “a” or “k” on their keyboards. Response times were measured from the onset of the stimulus to the onset of the key press response. Each new item was presented 500 ms. after the previous response. Inter-trial intervals were not fixed but participants were given infinite time to respond to a target word before a trial ended. There was no planned pause (or timeout) during the experiment. Prior to the experiment, participants were given seven practice trials to familiarize themselves with the procedure and to adjust the volume of the auditory stimuli. The practice trials were excluded from the analyses. The experiment lasted about 10 min. 3.6. Variables 3.6.1. English neighborhood frequency The overall frequency of the English phonological neighborhood, calculated as the sum of the individual frequencies of all neighboring words within the target word neighborhood was based on word frequencies obtained from the Clearpond (English) database (Marian et al., 2012). English neighborhood frequency values ranged between 84 and 59,371 per million. 3.6.2. German neighborhood frequency Each English target word had a minimum of 50 % German neighbors in the bilingual phonological neighborhood. The overall frequency of the German phonological neighborhood of an English target word was calculated as the sum of the individual frequencies of all German neighbors, based on the Clearpond database for German. Neighborhood frequencies ranged between 15 and 44,648 per million. A wide distribution of German neighborhood frequencies was a criterion for English target word selection. The following lexical characteristics were considered for each target word. 3.6.3. Length of words Monosyllabic words with 2 to 4 segments served as stimuli. Word length was determined by phonetically transcribing the target words (with ToPhonetics) and counting segment length with the Excel function LEN(). 3.6.4. English lexical frequency rate Frequency rates of the target words were obtained from the Clearpond database for English. All words ranged in the lower frequency spectrum of 3–3865 per million. There are indications in previous studies that neighborhood effects are amplified in low-density/ lowfrequency target words (Andrews, 1989, 1992; Forster & Shen, 1996; Luef, 2025a; Sears et al., 1995), leading to better detection of potentially milder neighborhood effects. In the present study, the target words were never the highest-frequency words within their neighborhoods. This allowed better testing of the effect of the neighbors on target recognition. 3.6.5. Phonotactic probability Biphone probabilities of the English target words were calculated with the Phonotactic Probability Calculator (Vitevitch & Luce, 2004). Biphone probability values reflect the sum of all biphone probabilities within a given word. Probabilities of target words ranged between 0.0001 and 0.0422. 3.6.6. Presence of cognates Being typologically close, English and German share many cognate word forms. In the present data, it was coded whether a cross-language phonological neighborhood contained any similar word forms with similar meanings in modern English and German (i.e., excluding historical cognates that have become phonologically non-transparent). Eight target words each had one German cognate in their neighborhood, in all cases translation terms with near identical meanings (i.e., east – Germ. “Ost”, guest – Germ. “Gast”, loud – Germ. “laut”, mile – Germ. “Meile”, mouth – Germ. “Maul”, out – Germ. “aus”, shift – Germ. “Schicht”, tube – German “Tube”), including one pseudo neighbor: field – Germ. “Feld”. 3.6.7. Phonological neighborhood density What is referred to as “degree” in phonological networks is the onesegment distance between phonological word forms (“Levenshtein distance”, see Levenshtein, 1966), which is the basis of neighborhood creation. Neighborhood density calculations were obtained from the C1English network. Only target words with sparse phonological neighborhoods of less than 10 neighbors in English were selected. The target words had between 2 and 9 English neighbors. The target words had between 2 and 18 German neighbors (the latter being quite dense neighborhoods in the German language). 3.6.8. English weighted degree The network term “weighted degree” refers to the strength of a link between two nodes. In the present study, phonological similarities between target words and their neighbors (including pseudo-neighbors) were calculated with the ALINE algorithm, computed with the “alineR” package in R (Downey et al., 2008). In the present case, the algorithm compared phonological features of the one segment that distinguished the phonological neighbors in English, including for instance manner, place, aspiration, nasality, and roundness (see Downey E.M. Luef Acta Psychologica 261 (2025) 105863 6 et al., 2017, for details). The ALINE distance score ranges between 0 (=perfect agreement) and 1 (=maximal phonological distance). It was converted to a similarity score by reverse scoring and multiplying the value by 100 (e.g., an ALINE score of 0.2 was converted to 80). This resulted in percentage-based phonological similarity scores that lead to more intuitive link strengths in a network (see Table 3). English weighted degree represented the average of all phonological similarities between a target word and its English neighbors. It ranged between 58 and 90. 3.6.9. German weighted degree ALINE scores were calculated between the English target words and their German phonological neighbors (and pseudo-neighbors), identical to the computations of English weighted degree. German weighted degree represented the average of all phonological similarities between a target word and its German neighbors. It ranged between 47 and 92. Weighted degrees and neighborhood frequencies in English showed a substantial negative correlation (r = − 0.55), which may indicate that higher frequency words are less phonologically similar to one another. The relationship between German neighborhood frequency and weighted degree was less pronounced (r = − 0.32, English), which may reflect the fact that lexical frequencies rates and phonological similarity are less tightly bound. Essentially, these correlative relationships may be artifacts of segmental and phonotactic frequencies (and probabilities) in the languages. 3.6.10. Clustering in the bilingual neighborhood The clustering coefficient (“”CC”) is a network statistic that measures the degree of interlinking of neighbors within a neighborhood (see Fig. 4). It measures how many neighbors of a target word are linked to one another. CC is based on the formula CC =2en/(kn(kn–1) ) which outputs values between 0 and 1, with the latter representing complete interlinking of all neighbors (Watts & Strogatz, 1998). In this study, only words with an intermediate CC between 0.2 and 0.6 in the bilingual German-English neighborhood were chosen as target words. 3.6.11. Network statistics All target words resided in the giant component part of the EnglishGerman phonological networks, which is the largest cluster of interconnected nodes in a network (Siew & Vitevitch, 2016). A “network density” compound variable was constructed by averaging the following five (log-transformed) bilingual network statistics: degree, weighted degree, clustering coefficient, as well as eigenvector centrality, and closeness centrality. Eigenvector centrality takes into account the centralities of neighboring nodes and assigns the highest values to those words that have highly centralized neighbors (Bonacich, 2007). Through this, words with many neighbors that have themselves many neighbors can be identified. Such a structure is indicative of the density of a particular network area. Closeness centrality is a measure of how close each node is to all other nodes in a network when measuring the shortest distance (Salavati et al., 2019). It can be understood as the inverse of the average distance from a given node to all other nodes in a network. High values indicate that a word is close to numerous other words in a network, which is indicative of the density of the network part in which a word resides (Goldstein & Vitevitch, 2017; Vitevitch & Goldstein, 2014). High values in the aggregate of the five metrics indicate denser areas in the overall bilingual network. Fig. 5 illustrates the relationship between the network statistics in a partial English phonological network where the densest region (indicated with darker nodes) can be identified using the five metrics outlined above. The highest-degree word (i.e., the one with the most phonological neighbors) is grace, followed by great, grape, grade, grave, and gray. These are also the words with the highest weighted degrees. The same words are also the ones with the highest CC in the network, as they are all linked to one another in their neighborhoods. The neighborhood around grape is also characterized by highest Eigenvector centrality, and many of its members show the highest closeness centrality values of the network (i.e., trace, grace, race). Conversely, the words trial and trail are characterized by lower values in degree, weighted degree, CC, closeness and eigenvector centralities in the network. A combination of these network statistics can inform about the density of a specific network region where a target word is located. All target words used in the lexical decision task stemmed from parts of the phonological network that are characterized by intermediate density, with logarithmic values of the compound density variable ranging between −0.13 and 0.37. Since this range is broad, the density variable was added as a control variable to the statistical models. 3.7. Statistical analysis After experimental data collection, reaction time data was cleaned and sorted in various ways. First, responses below 200 ms were removed, as they indicate non-processing of the stimulus (Whelan, 2008). Reaction times per participant were z-scored and values exceeding +3 or −3 were considered outliers and removed. Target words that elicited more than 30 % of judgment errors were deleted from the dataset. This affected two words, “owl” and “shout”. In addition, one target word, time, had to be removed as the audio file turned out to be corrupted. The overall data removed amounted to 5.6 % of the raw data. A total of 47 words remained and were processed for further analysis. One spread sheet containing all responses (N =3161) was compiled, in addition to one containing only correct responses (N = 3058). A generalized linear mixed effects model (GLM) was run in R with the package lme4 (Bates et al., 2014) to see whether the accuracy of word judgments (as real or non-words) was influenced by the frequency of the German neighborhood of a target word. As dependent variable the Table 3 Phonological similarity scores calculated with the ALINE algorithm between the word “a” and three of its phonological neighbors. word phonetic neighbor phonetic alineR output rescaled value a eɪache eɪk 0.37 63 a eɪage eɪʤ 0.37 63 a eɪair eə0.17 83 Fig. 4. Clustering in the “mass” neighborhood. The clustering coefficient of 0.52 indicates that a high number of phonological neighbors of “mass” are neighbors of one another, as indicated by the links. E.M. Luef Acta Psychologica 261 (2025) 105863 7 binary outcome “correct/incorrect” was entered. The following fixed effects were included: an interaction variable of German neighborhood frequency and German weighted degree as the variable of interest, in addition to a number of control variables, including English neighborhood frequency interacted with English weighted degree, lexical frequency rates of target words, presence of cross-language cognate (yes/ no), and density of the network part in which the target word resided. No correlations between fixed effects were detected. Random effects for target word and participant ID were specified, with random slopes for German and English neighborhood frequencies, as well as for lexical frequency rates of the target words. A binomial regression with “cloglog” links was specified in the GLM, as the data contained a larger number of correct than incorrect responses (only 3.3 % were incorrect). No convergence issues appeared during the computation. As a next step, a linear mixed effects model (LME) was computed to investigate the influence of German neighborhood frequency on reaction times in milli-seconds to target words. The LME model only included correct responses of the participants. The LME was structured in an identical way to the GLM. For the LME model, degrees of freedom and significance values can be estimated via approximations, for which the Satterthwaite's method is commonly used. Functions for it were utilized in the R package lmerTest (Kuznetsova et al., 2017). 4. Results As is typically the case in lexical decision tasks, responses to nonwords were slower than responses to real words (average non-words =2.39 s., SD =1.54; average real words =1.75 s., SD =0.81), indicating faster processing of real words. 4.1. Accuracy of word judgments Table 4 summarizes the results of the generalized linear mixed effects model that tested whether there was any difference in lexical characteristics of words that were accurately judged (e.g., real words being judged as real words) or incorrectly judged (e.g., a non-word judged as being a real word). The sample size of this accuracy model was 3161 responses. German neighbors were not shown to influence accuracy of word judgments (see Table 4). Results yielded only effects of target word frequency rate on the accuracy of the word judgments, where a significance of p =0.01 indicates that the accuracy of word judgment increases together with lexical frequency rate of target words (i.e., 0.36 higher log odds of accurate judgment, see estimate in Table 4). Correctly judged target words (i.e., real words as real words) were generally characterized by higher lexical frequency rates than those words that were judged incorrectly. Fig. 5. A partial phonological network. The darker area indicates the densest network region in terms of degree, weighted degree, clustering coefficient, closeness centrality, and Eigenvector centrality. Table 4 Generalized linear mixed effects model estimates for fixed and random effects. Judgment accuracy model. Variable Variance Estimate SD SE z p Random effects Participant German neighborhood frequency 9.264e-03 0.0962479 English neighborhood frequency 1.411e-02 0.1187986 Lexical frequency rate 4.634e-01 0.6807021 Word German neighborhood frequency 1.333e-01 0.3651588 English neighborhood frequency 1.277e-02 0.1129883 Lexical frequency rate 2.201e-05 0.0046910 Fixed effects Intercept −19.36 20.37 −0.95 0.34 German neighborhood frequency 3.24 4.55 0.71 0.48 German weighted degree 1.93 7.65 0.25 0.8 English neighborhood frequency 2.54 4.46 0.57 0.57 English weighted degree 8.88 8.46 1.05 0.29 Target lexical frequency rate 0.36 0.15 2.45 0.014 * Network density −0.65 0.75 −0.87 0.39 Cognate presence 0.26 0.19 1.41 0.16 German neighborh. Frequ. & German weighted degree (interaction) −1.75 2.4 −0.73 0.47 English neighborh. Frequ. & English weighted degree (interaction) −1.27 2.38 −0.53 0.59 * p < 0.05. E.M. Luef Acta Psychologica 261 (2025) 105863 8 4.2. Reaction times Testing reaction times to target words, the linear mixed effect model results are presented in Table 5. Incorrect responses were removed from the sample, resulting in a sample size of 3057 responses for the reaction times model. As indicated in Table 5, reaction times were influenced by German neighborhood frequency, German weighted degree, as well as the interaction between the two. The absence of any effects of English neighborhood frequency will be discussed below. The interacted variable of German neighborhood frequency and German weighted degree was associated with response latencies to target words. The positive estimate indicates that more frequent and phonologically similar German neighbors caused a delay in recognition of an English target words. Inspecting the interaction more closely, differences emerge in terms of the effect of phonological similarity and frequency (see Fig. 6). German weighted degree is the moderating variable that impacts the relationship between reaction times and German neighborhood frequency. Neighborhoods with phonological similarity values of one standard deviation above the mean (i.e., +1 SD) show a strong association between higher neighborhood frequency and longer response latencies. This means that frequent German neighbors led to longer reaction times in cases where phonological similarity was high. Phonological similarities values below the mean (i.e., −1 SD) show the reverse effect of shorter response latencies to target words with higher frequency neighborhoods, implying that phonologically dissimilar German neighbors elicited shorter response times in high-frequency German neighborhoods. Hence, the observed cross-language neighborhood effect was mediated by phonological similarity of the crosslanguage neighbors. 5. Discussion 5.1. Main findings This study investigated the effect of neighborhood frequency in bilingual phonological neighborhoods of German learners of English as a second language. Results demonstrate that higher lexical frequencies of cross-language neighbors (i.e., German neighbors of English target words), especially in connection with high phonological similarity, delay reaction times to English target words. This finding underscores the influence of a first language on a second, learned language and provides evidence for a cross-language frequency effect on the lexical level. 5.2. Implications Support for an integrated bilingual word form lexicon, where activation spreads across languages, primarily stems from neighborhood density investigations, taking into account the number of phonological neighbors a target word has (e.g., Blumenfeld & Marian, 2007; CansecoGonzalez et al., 2010; Ju & Luce, 2004; Lagrou et al., 2011; Luef, 2025a; Marian & Spivey, 2003; Shook & Marian, 2012; Weber & Cutler, 2004). This study is the first to report on neighborhood frequency effects in a cross-linguistic context. Frequency effects seem to heighten competition in neighborhoods where word forms from multiple languages share activation. Especially in late bilingualism, where a dominant first language often overshadows a second language, the L1 word forms exert strong influence in a neighborhood (Spivey & Marian, 1990; Weber & Cutler, 2004; Wen & van Heuven, 2017). It is likely that first language lexical frequencies are more deeply entrenched in speakers' mental lexica and therefore have a stronger impact than second language lexical frequencies (Diependaele et al., 2013). Cross-linguistic frequency effects change the trajectory of lexical activation spreading, which leads to different predictions regarding word recognition. This creates an imperative to consider frequency effects in studies of the bilingual lexicon, analogous to what has been proposed by Luce and Pisoni (1998) for monolingual neighborhoods. Previous research has shown that smaller vocabularies typically exhibit larger frequency effects (Brysbaert, Lagrou, & Stevens, 2017), a finding of relevance for second languages (Cop et al., 2015). According to the lexical entrenchment hypothesis, less exposure to (and proficiency in) a language leads to more pronounced frequency effects (Brysbaert, Lagrou, & Stevens, 2017). A smaller number of words being used repeatedly – as is commonly the case with second languages – leads to deeper cognitive embedding of these words and amplifies frequency effects. It is notable that no L2 neighborhood effects were observed in the present study. There are two explanations that seem likely in the present case. For one, neighborhood effects may be dominated by the first language but not there yet for L2 neighborhoods. In this context the question arises how advanced an L2 learner must be for those L2 neighborhood effects to arise. This study involved L2 learners at a quite advanced proficiency level (CEFR-C1), so it would be expected that neighborhood effects should play some role in lexical processing. In fact, L2 learners at this proficiency level were previously found to show neighborhood density effects (Luef, 2025a). An alternative explanation for the absence of English neighborhood effects could be the specific limitations imposed on target word selection. Stimuli were chosen that Table 5 Linear mixed effects model estimates for fixed and random effects. Reaction time model. Variable Variance Estimate SD SE p Random effects Participant German neighborhood frequency 8.898e-07 0.0009433 English neighborhood frequency 3.669e-07 0.0006057 Lexical frequency rate 1.181e-06 0.0010868 Word German neighborhood frequency 2.343e-05 0.0048400 English neighborhood frequency 5.636e-05 0.0075072 Lexical frequency rate 1.929e-04 0.0138890 Fixed effects Intercept 2.42 0.84 0.01 * German neighborhood frequency −0.41 0.13 0.005 ** German weighted degree −0.74 0.26 0.013* English neighborhood frequency −0.12 0.19 0.54 English weighted degree −0.42 0.38 0.29 Target lexical frequency rate −0.01 0.004 0.08 Network density −0.03 0.036 0.47 Cognate presence 0.005 0.01 0.47 German neighborh. Frequ. & German weighted degree (interacted) 0.22 0.07 0.0056 ** English neighborh. Frequ. & English weighted degree (interacted) 0.06 0.11 0.55 * p < 0.05. ** p < 0.01. E.M. Luef Acta Psychologica 261 (2025) 105863 9 Table A (continued) Target Language Neighbor IPA Translation Cognate Pseudo words/ IPA mies mis lousy es ε s it Moos mos moss mile English mail_male meɪl Post, m¨ annlich Meile may maɪk¨ onnte might maɪt k¨ onnte file faɪl Datei mine maɪn Mine pile paɪl Stapel while waɪl w¨ ahrend smile smaɪl l¨ acheln German mal_Mahl mal times, meal Beil baɪl hatchet Keil kaɪl wedge geil gaɪl cool, good heil haɪl intact, healed Meile maɪləmile yes Mais maɪs corn Seil zaɪl rope weil vaɪl because Teil taɪl part Maul maʊl mouth mint English minute mɪnət Minute Minze meant m ε nt gemeint mist mɪst Nebel hint hɪnt Hinweis German Mond mont moon Mund mʊnt mouth Rind ʁɪnt beef Mist mɪst gargabe Wind vɪnt wind Kind kɪnt child mild mɪlt mild mit mɪt with mouth English mouse maʊs Maus Mund South saʊθSüden German Maul maʊl mouth yes Mauer maʊɐ wall Maus maʊs mouse now English wow waʊwau jetzt cow kaʊKuh how haʊwie no_know noʊnein, wissen noun naʊn Nomen German Narr na fool nass nas wet nach nax after lau laʊlukewarm Schau ʃaʊshow rau ʁaʊrough Tau taʊdew Sau zaʊsow (pig) Bau baʊconstruction, building Au aʊmeadowland Naht nat seam nah na near nahe nahənear ought English all ɔl alles sollte off ɔf aus at æt bei eat it essen it ɪt es caught k α t gefangen thought θɔt Gedanke, gedacht German Ost ɔst East Ort ɔɐt place ob ɔp whether oft ɔft often Gott gɔt God Otter ɔtɐotter Pott pɔt pot out English hour_our aʊər Stunde, unser/e aus, raus owl aʊl Eule outer aʊtər¨ außere/r doubt daʊt Zweifel shout ʃaʊt Ruf (continued on next page) E.M. Luef Acta Psychologica 261 (2025) 105863 16 Table A (continued) Target Language Neighbor IPA Translation Cognate Pseudo words/ IPA put pʊt setzen, stellen, legen route rut Strecke foot fʊt Fuß German Autsch aʊtʃouch auch aʊx also acht axt eight Ast ast branch Au aʊmeadowland Aua aʊɐ ouch auf aʊf on Haut haʊt skin laut laʊt loud Abt abt abbot Art at type Schutt ʃʊt debris Eid aɪt oath Akt akt act alt alt old Amt amt office aus aʊt out yes Auto aʊtəcar owl English hour_our aʊər Stunde, unser/e Eule out aʊt aus, raus towel taʊəl Handtuch wool wʊl Wolle vowel vaʊəl Vokal pull pʊl ziehen bull bʊl Bulle foul faʊl Verstoß full fʊl voll German Aal_all al eel, space Aula aʊlɐauditorium faul faʊl lazy aus aʊs out auch aʊx also Au aʊmeadowland Aua aʊɐ ouch auf aʊf on Gaul gaʊl horse Null nʊl zero Maul maʊl mouth piece_peace English pea pi Erbse Stück, Frieden purse pɜrs Geldbeutel pass pæs Pass niece nis Nichte peel pil sch¨ alen peek_peak pik gucken, Spitze German fies fis nasty dies dis this Spießʃpis pike Pier piɐpier Pieps pips beep Piep pip beep Pass pas pass Kies kis gravel mies mis lousy pile English pine paɪn Kiefer Stapel while waɪl w¨ ahrend pale peɪl blass pipe paɪp Rohr pie paɪKuchen mile maɪl Meile file faɪl Datei pill pɪl Pille German Seil zaɪl rope Pfeil pfaɪl arrow Teil taɪl part weil vaɪl because Keil kaɪl wedge geil gaɪl cool, good heil haɪl intact, healed Beil baɪl hatchet shift English gift gɪft Geschenk Schicht lift lɪft heben German Schiff ʃɪf ship Schicht ʃɪxt shift yes (continued on next page) E.M. Luef Acta Psychologica 261 (2025) 105863 17 Table A (continued) Target Language Neighbor IPA Translation Cognate Pseudo words/ IPA Schild ʃɪlt sign Schrift ʃʁɪft writing Stift ʃtɪft pen Gift gɪft poison Lift lɪft elevator Schuft ʃʊft villain shoot English suit sut Anzug schießen sheet ʃit Laken root rut Wurzel shirt ʃɜrt Hemd shot ʃ α t Schuss shut ʃʌt geschlossen shoe ʃu Schuh German Wut vut anger Schuh ʃu shoe Hut hut hat Schub ʃup thrust Schutt ʃʊt debris gut gut good Mut mut courage Shirt ʃœɐt shirt shout English route raʊt Strecke rufen out aʊt aus, raus doubt daʊt Zweifel German Schacht ʃaxt shaft Schaum ʃaʊm foam Schau ʃaʊshow laut laʊt loud Haut haʊt skin Schutt ʃʊt debris Schauer ʃaʊɐ shower talk English toy tɔɪ Spielzeug reden, sprechen tech t ε k Technik walk wɔk Spaziergang, gehen German Schock ʃɔk shock Stock ʃtɔk stick Tag tak day toll tɔl great Dock dɔk dock Bock bɔk buck Rock ʁɔk skirt time English dime daɪm cent Zeit tie taɪbinden tyre taɪər Reifen tide taɪd Gezeiten type taɪp Art tight taɪt eng German heim haɪm home Teich taɪx pond Leim laɪm glue Teil taɪl part Reim ʁaɪm rhyme Teig taɪk dough Keim kaɪm germ, seed town English towel taʊəl Handtuch Stadt tower taʊər Turm noun naʊn Nomen down daʊn unten tone toʊn Ton German Zaun tsaʊn fence Tausch taʊʃ swap, trade taub taʊp deaf tau taʊdew Faun faʊn faun tube English to_too_two tu zu, auch, zwei R¨ ohre, Tube tool tul Werkzeug tomb tum Grab tooth tuθZahn German Tour tuɐtour tun tun do Tuba tubɐtuba Tube tubəR¨ ohre, Tube yes Pseudo neighbors Soup (Engl.) sup Suppe taub (Germ.) taʊp deaf Tipp (Germ.) tɪp tip vet English set s ε t setzen (continued on next page) E.M. Luef Acta Psychologica 261 (2025) 105863 18 Table A (continued) Target Language Neighbor IPA Translation Cognate Pseudo words/ IPA Tierarzt debt d ε t Schulden Let l ε t lassen pet p ε t Haustier get g ε t bekommen yet j ε t noch wet w ε t nass bet b ε t Wette net n ε t Netz German weh ve hurt fett f ε t fat Watt vat watt, mud flats Bett b ε t bed nett n ε t nice Welt v ε lt world wenn v ε n if wert v ε ɐt worth Wetter v ε tɐweather Wut vut anger wer_Wehr v ε ɐwho, defense zoo English do du tun Zoo sue su klagen new nu neu who hu wer you ju du to_too_two tu zu, auch, zwei shoe ʃu Schuh German Ruh ʁu rest Kuh_Coup ku cow, coup so zo so Sie zi you, they See ze lake Schuh ʃu shoe du du you, they Appendix B. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.actpsy.2025.105863. Data availability I have included the data as supplementary material References Amengual, M. (2016). The perception of language-specific phonetic categories does not guarantee accurate phonological representations in the lexicon of early bilinguals. Applied PsychoLinguistics, 37(5), 1221–1251. https://doi.org/10.1017/ S0142716415000557 Andrews, S. (1989). Frequency and neighborhood effects on lexical access: Activation or search? Journal of Experimental Psychology: Learning, Memory, and Cognition, 15(5), 802–814. https://doi.org/10.1037/0278-7393.15.5.802 Andrews, S. (1992). Frequency and neighborhood effects on lexical access: Lexical similarity or orthographic redundancy? Journal of Experimental Psychology: Learning, Memory, and Cognition, 18(2), 234–254. https://doi.org/10.1037/02787393.18.2.234 Bastian, M., Heymann, S., & Jacomy, M. (2009). Gephi: An open source software for exploring and manipulating networks (association for the advancement of artificial intelligence, ed.). Bates, D., Maechler, M., Bolker, B., & Walker, S. (2014). {lme4}: Linear mixed-effects models using Eigen and S4. R Package Version, 1, 1–7. Baus, C., Costa, A., & Carreiras, M. (2008). Neighbourhood density and frequency effects in speech production: A case for interactivity. Language & Cognitive Processes, 23(6), 866–888. https://doi.org/10.1080/01690960801962372 Belke, E., Brysbaert, M., Meyer, A. S., & Ghyselinck, M. (2005). Age of acquisition effects in picture naming: Evidence for a lexical-semantic competition hypothesis. Cognition, 96(2), B45–B54. https://doi.org/10.1016/j.cognition.2004.11.006 Blumenfeld, H. K., & Marian, V. (2007). Constraints on parallel activation in bilingual spoken language processing: Examining proficiency and lexical status using eyetracking. Language & Cognitive Processes, 22(5), 633–660. https://doi.org/10.1080/ 01690960601000746 Bohn, O. S., & Flege, J. E. (1992). The production of new and similar vowels by adult German learners of English. Studies in Second Language Acquisition, 14, 131–158. https://doi.org/10.1017/S0272263100010792 Bonacich, P. (2007). Some unique properties of eigenvector centrality. Social Networks, 29(4), 555–564. https://doi.org/10.1016/j.socnet.2007.04.002 Borowsky, R., & Masson, M. E. (1999). Frequency effects and lexical access: On the interpretation of null pseudohomophone base-word frequency effects. Journal of Experimental Psychology: Human Perception and Performance, 25(1), 270. https://doi. org/10.1037/0096-1523.25.1.270 de Bot, K., & Batyi, S. (2022). Bilingual models of speaking. Routledge. Bryla, B. (2015). Oracle database 12c handbook: Manage a scalable, secure Oracle enterprise database environment. McGraw Hill. Brysbaert, M., & Biemiller, A. (2016). Test-based age-of-acquisition norms for 44 thousand English word meanings. Behavior Research Methods, 49, 1520–1523. https://doi.org/10.3758/s13428-016-0811-4 Brysbaert, M., Buchmeier, M., Conrad, M., Jacobs, A. M., Boelte, J., & Boehl, A. (2011). The word frequency effect: A review of recent developments and implications for the voice of frequency estimates in German. Experimental Psychology, 58(5), 412. https://doi.org/10.1027/1618-3169/a000123 Brysbaert, M., Keuleers, E., & Mandera, P. (2019). Recognition times for 54 thousand Dutch words: Data from the Dutch crowdsourcing project. Psychologica Belgica, 59 (1), 281–300. https://doi.org/10.5334/pb.491 Brysbaert, M., Lagrou, E., & Stevens, M. (2017). Visual word recognition in a second language: A test of the lexical entrenchment hypothesis with lexical decision times. Bilingualism: Language and Cognition, 20(3), 530–548. https://doi.org/10.1017/ S1366728916000353 Brysbaert, M., Mandera, P., & Keuleers, E. (2017). The word frequency effect in word processing: An updated review. Current Directions in Psychological Science, 27(1), 45–50. https://doi.org/10.1177/0963721417727521 Brysbaert, M., & New, B. (2009). Moving beyond Kuˇ cera and Francis: A critical evaluation of current word frequency norms and the introduction of a new and improved word frequency measure for American English. Behavior Research Methods, 41, 977–990. https://doi.org/10.3758/BRM.41.4.977 Canseco-Gonzalez, E., Brehm, L., Brick, C. A., Brown-Schmidt, S., Fischer, K., & Wagner, K. (2010). Carpet or carcel: The effect of age of acquisition and language mode on bilingual lexical access. Language & Cognitive Processes, 25(5), 669–705. https://doi.org/10.1080/01690960903474912 Capel, A. (2015). The English vocabulary profile. In J. Harrison, & F. Barker (Eds.), English profile studies: English profile in practice (pp. 9–27). Cambridge University Press. E.M. Luef Acta Psychologica 261 (2025) 105863 19 Cervera-Crespo, T., & Gonzalez-Alvarez, J. (2019). Speech perception: Phonological neighborhood effects on word recognition persist despite semantic sentence context. Perceptual and Motor Skills, 126(6), 1047–1057. https://doi.org/10.1177/ 0031512519870032 Chan, K. Y., & Vitevitch, M. S. (2009). The influence of the phonological neighborhood clustering coefficient on spoken word recognition. Journal of Experimental Psychology: Human Perception and Performance, 35(6), 1934–1949. https://doi.org/ 10.1037/a0016902 Chen, Q., & Mirman, D. (2012). Competition and cooperation among similar representations: Toward a unified account of facilitative and inhibitory effects of lexical neighbors. Psychological Review, 119(2), 417–430. https://doi.org/10.1037/ a0027175 Chern, J., Castro, N., & Siew, C. S. Q. (2025). Evidence of community structure in phonological networks of various languages. Canadian Journal of Experimental Psychology, 79(1), 4–14. https://doi.org/10.1037/cep0000357 Coltheart, M., Rastle, K., Perry, C., Langdon, R., & Ziegler, J. (2001). A dual route cascaded model of visual word recognition and reading aloud. Psychological Review, 108, 204–256. https://doi.org/10.1037/0033-295x.108.1.204 Cop, U., Keuleers, E., Drieghe, D., & Duyck, W. (2015). Frequency effects in monolingual and bilingual natural reading. Psychonomic Bulletin & Review, 22(5), 1216–1234. https://doi.org/10.3758/s13423-015-0819-2 Council of Europe. (2018). Common European framework of reference for languages: Learning, teaching, assessment. (Companion volume with new descriptors). Csardi, G., & Nepusz, T. (2006). The igraph software package for complex network research. InterJournal: Complex Systems, 1–9, 1695. Darcy, I., Daidone, D., & Kojima, C. (2013). Asymmetric lexical access and fuzzy lexical representations in second language learners. The Mental Lexicon, 8(3), 372–420. https://doi.org/10.1075/ml.8.3.06dar Dell, G. S. (1986). A spreading-activation theory of retrieval in sentence production. Psychologial Review, 93(3), 283–321. https://doi.org/10.1037/0033-295X.93.3.283 Diependaele, K., Lemhoefer, K., & Brysbaert, M. (2013). The word frequency effect in firstand second-language word recognition: A lexical entrenchment account. Quarterly Journal of Experimental Psychology, 66(5), 843–863. https://doi.org/ 10.1080/17470218.2012.720994 Dijkstra, T., Miwa, K., Brummelhuis, B., Sappelli, M., & Baayen, R. H. (2010). How crosslanguage similarity and task demands affect cognate recognition. Journal of Memory and Language, 62, 284–301. https://doi.org/10.1016/j.jml.2009.12.003 Dirks, D. D., Takayanagi, S., & Moshfegh, A. (2001). Effects of lexical factors on word recognition among normal-hearing and hearing-impaired listeners. Journal of the American Academy of Audiology, 12(5), 233–244. Downey, S., Hallmark, B., Cox, M. P., Norquest, P., & Lansing, J. S. (2008). Computational feature-sensitive reconstruction of language relationships: Developing the ALINE distance for comparative historical linguistic reconstruction. Journal of Quantitative Linguistics, 15(4), 340–369. https://doi.org/10.1080/ 09296170802326681 Downey, S., Sun, G., & Norquest, P. (2017). AlineR: An R package for optimizing featureweighted alignments and linguistic distances. The R Journal, 9(1), 138–152. Dufour, S., & Frauenfelder, U. H. (2010). Phonological neighbourhood effects in French spoken-word recognition. Quarterly Journal of Experimental Psychology, 63(2), 226–238. https://doi.org/10.1080/17470210903308336 Ellis, A. W., & Lambon Ralph, M. A. (2000). Age of acquisition effects in adult lexical processing reflect loss of plasticity in maturing systems: Insights from connectionist networks. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26(5), 1103–1123. https://doi.org/10.1037//0278-7393.26.5.1103 Ellis, N. C. (2002). Frequency effects in language processing: A review with implications for theories of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24, 143–188. https://doi.org/10.1017/S0272263102002024 Elsherif, M. M., Preece, E., & Catling, J. C. (2023). Age-of-acquisition effects: A literature review. Journal of Experimental Psychology: Learning, Memory, and Cognition, 49(5), 812–847. https://doi.org/10.1037/xlm0001215 Ferrand, L., New, B., Brysbaert, M., & Keuleers, E. (2010). The French lexicon project: Lexical decision data for 38,840 French words and 38,840 pseudowords. Behavior Research Methods, 42(2), 488–496. https://doi.org/10.3758/BRM.42.2.488 Forster, K. I., & Shen, D. (1996). No enemies in the neighborhood: Absence of inhibitory neighborhood effects in lexical decision and semantic categorization. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22(3), 696–713. https:// doi.org/10.1037//0278-7393.22.3.696 Frances, C., Navarra-Barindelli, E., & Martin, C. D. (2021). Inhibitory and facilitatory effects of phonological and orthographic similarity on L2 word recognition across modalities in bilinguals. Scientific Reports, 11. https://doi.org/10.1038/s41598-02192259-z Frauenfelder, U. H., Baayen, R. H., Hellwis, F. M., & Schreuder, R. (1993). Neighborhood density and frequency across languages and modalities. Journal of Memory and Language, 32, 781–804. https://doi.org/10.1006/jmla.1993.1039 Gahl, S., & Strand, J. F. (2016). Many neighborhoods: Phonological and perceptual neighborhood density in lexical production and perception. Journal of Memory and Language, 89, 162–178. https://doi.org/10.1016/j.jml.2015.12.006 Goldinger, S. D., Luce, P. A., & Pisoni, D. B. (1989). Priming lexical neighbors of spoken words: Effects of competition and inhibition. Journal of Memory and Language, 28(5), 501–518. https://doi.org/10.1016/0749-596X(89)90009-0 Goldinger, S. D., Pisoni, D., & Logan, J. S. (1991). On the nature of talker variability effects on recall of spoken word lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17(1), 152–162. https://doi.org/10.1037//02787393.17.1.152 Goldstein, R., & Vitevitch, M. (2017). The influence of closeness centrality on lexical processing. Frontiers in Psychology, 8, Article 1683. https://doi.org/10.3389/ fpsyg.2017.01683 Gr´ af, T. (2017). The story of the learner corpus LINDSEI_CZ. Studies in Applied Linguistics, 8(2), 22–35. Hameau, S., Biedermann, B., & Nickels, L. (2021). Lexical activation in late bilinguals: Effects of phonological neighbourhood on spoken word production. Language, Cognition and Neuroscience, 36(4), 517–534. https://doi.org/10.1080/ 23273798.2020.1863438 Hickey, R. (2019). Persistent features in the English of German speakers. In R. Hickey (Ed.), English in the German-speaking world. Cambridge University Press. Ju, M., & Luce, P. (2004). Falling on sensitive ears: Constraints on bilingual lexical activation. Psychological Science, 15(5), 314–318. https://doi.org/10.1111/j.09567976.2004.00675.x Kapatsinski, V. (2006). Sound similarity relations in the mental lexicon: Modeling the lexicon as a complex network. Speech Research Lab Progress Report, 27, 133–152. Karimi, H., & Diaz, M. (2020). When phonological neighborhood density both facilitates and impedes: Age of acquisition and name agreement interact with phonological neighborhood during word production. Memory & Cognition, 48, 1061–1072. https://doi.org/10.3758/s13421-020-01042-4 Kroll, J. F., & Ma, F. (2017). The bilingual lexicon. In E. M. Fernandez, & H. Smith Cairns (Eds.), The handbook of psycholinguistics (pp. 294–319). Wiley. Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. (2017). LmerTest package: Tests in linear mixed effects models. Journal of Statistical Software, 82(3), 1–26. https://doi. org/10.18637/jss.v082.i13 Lagrou, E., Hartsuiker, R. J., & Duyck, W. (2011). Knowledge of a second language influences auditory word recognition in the native language. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37(4), 952–965. https://doi.org/ 10.1037/a0023217 Landauer, T. K., & Streeter, L. A. (1973). Structural differences between common and rare words: Failure of equivalence assumptions for theories of word recognition. Journal of Verbal Learning and Verbal Behavior, 12, 119–131. https://doi.org/ 10.1016/S0022-5371(73)80001-5 Lemh¨ ofer, K., Dijkstra, T., Schriefers, H., Baayen, R. H., Grainger, J., & Zwitserlood, P. (2008). Native language influences on word recognition in a second language: A megastudy. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34 (1), 12–31. https://doi.org/10.1037/0278-7393.34.1.12 Leung, C. C., & Chau, H. F. (2007). Weighted assortative and disassortative networks model. Physica A: Statistical Mechanics and its Applications, 378(2), 591–602. https:// doi.org/10.1016/j.physa.2006.12.022 Levenshtein, V. I. (1966). Binary codes capable of correcting deletions, insertions, and reversals. Soviet Physics Doklady, 10, 707–710. Lewy, N., & Grosjean, F. (2008). The Lewy and Grosjean BIMOLA model. In F. Grosjean (Ed.), Studying bilinguals (pp. 201–210). Oxford University Press. https://doi.org/ 10.1093/oso/9780199281282.003.0011. Llompart, M. (2021). Phonetic categorization ability and vocabulary size contribute to the encoding of difficult second-language phonological contrasts into the lexicon. Bilingualism: Language and Cognition, 24(3), 481–496. https://doi.org/10.1017/ S1366728920000656 Llompart, M., & Reinisch, E. (2020). The phonological form of lexical items modulates the encoding of challenging second-language sound contrasts. Journal of Experimental Psychology: Learning, Memory, and Cognition, 46(8), 1590–1610. https:// doi.org/10.1037/xlm0000832 Luce, P., Goldinger, S., Auer, E. T., & Vitevitch, M. S. (2000). Phonetic priming, neighborhood activation, and PARSYN. Perception & Psychophysics, 62(3), 615–625. https://doi.org/10.3758/BF03212113 Luce, P., & Large, N. R. (2001). Phonotactics, density, and entropy in spoken word recognition. Language And Cognitive Processes, 16, 565–581. https://doi.org/ 10.1080/01690960143000137 Luce, P., & Pisoni, D. B. (1998). Recognizing spoken words: The neighborhood activation model. Ear and Hearing, 19(1), 1–36. https://doi.org/10.1097/00003446199802000-00001 Luef, E. M. (2023). Obsolescence effects in second language phonological networks. Memory & Cognition, 52(4), 771–792. https://doi.org/10.3758/s13421-023-01500-9 Luef, E. M. (2025a). Modeling the bilingual lexicon as a multiplex phonological network. Canadian Journal of Experimental Psychology, 79(1), 41–60. https://doi.org/10.1037/ cep0000351 Luef, E. M. (2025b). Empirical models of phonological networks and their growth in English. Karolinum Press. Marian, V., Bartolotti, J., Chabal, S., & Shook, A. (2012). CLEARPOND: Cross-linguistic easy access resource for phonological and orthographic neighborhood densities. PLoS One, 7(8), Article e43230. https://doi.org/10.1371/journal.pone.0043230 Marian, V., Blumenfeld, H. K., & Boukrina, O. V. (2008). Sensitivity to phonological similarity within and across languages. Journal of Psycholinguistic Research, 37(3), 141–170. https://doi.org/10.1007/s10936-007-9064-9 Marian, V., & Spivey, M. (2003). Competing activation in bilingual language processing: Withinand between-language competition. Bilingualism: Language and Cognition, 6, 97–115. https://doi.org/10.1017/S1366728903001068 Mirman, D. (2011). Effects of near and distant semantic neighbors on word production. Cognitive, Affective, & Behavioral Neuroscience, 11, 32–43. https://doi.org/10.3758/ s13415-010-0009-7 Mirman, D., & Kittredge, A. K. (2010). Effects of near and distant phonological neighbors on picture naming. Proceedings of the Annual Meeting of the Cognitive Science Society, 32, 1447–1452. Mirman, D., & Magnuson, J. S. (2008). Attractor dynamics and semantic neighborhood density: Processing is slowed by near neighbors and speeded by distant neighbors. E.M. Luef Acta Psychologica 261 (2025) 105863 20 Journal of Experimental Psychology: Learning, Memory, and Cognition, 34(1), 65–79. https://doi.org/10.1037/0278-7393.34.1.65 Newman, R. S., & German, D. J. (2002). Effects of lexical factors on lexical access among typical language-learning children and children with word-finding difficulties. Language and Speech, 43, 285–317. https://doi.org/10.1177/ 00238309020450030401 Peirce, J. W., Gray, J. R., Simpson, S., MacAskill, M. R., H¨ ochenberger, R., Sogo, H., … Lindeløv, J. (2019). PsychoPy2: Experiments in behavior made easy. Behavior Research Methods, 51, 195–203. https://doi.org/10.3758/s13428-018-01193-y Pisoni, D. B., Nusbaum, H. C., Luce, P. A., & Slowiaczek, L. M. (1985). Speech perception, word recognition and the structure of the lexicon. Speech Communication, 4(1), 75–95. https://doi.org/10.1016/0167-6393(85)90037-8 Rocca, B., Llompart, M., & Darcy, I. (2025). Phonological neighborhood density, phonetic categorization, and vocabulary size differentially affect the phonolexical encoding of easy and difficult L2 segmental contrasts. Bilingualism: Language and Cognition, 28(3), 662–675. https://doi.org/10.1017/S1366728924000865 Roediger, H. L., & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(4), 803–814. https://doi.org/10.1037/0278-7393.21.4.803 Roodenrys, S., Hulme, C., Lethbridge, A., Hinton, M., & Nimmo, L. M. (2002). Wordfrequency and phonological-neighborhood effects on verbal short-term memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(6), 1019–1034. https://doi.org/10.1037//0278-7393.28.6.1019 Salavati, C., Abdollahpouri, A., & Manbari, Z. (2019). Ranking nodes in complex networks based on local structure and improving closeness centrality. Neurocomputing, 336, 36–45. https://doi.org/10.1016/j.neucom.2018.04.086 Schulpen, B., Dijkstra, T., Schriefers, H., & Hasper, M. (2003). Recognition of interlingual homophones in bilingual auditory word recognition. Journal of Experimental Psychology: Human Perception and Performance, 29(6), 1155–1178. https://doi.org/ 10.1037/0096-1523.29.6.1155 Schwartz, A. I., & Kroll, J. F. (2007). Language processing in bilingual speakers. In M. J. Traxler, & M. A. Gernsbacher (Eds.), Hanbook of psycholinguistics (pp. 963–995). Elsevier. https://doi.org/10.1016/B978-012369374-7/50026-2. Sears, C. R., Campbell, C. R., & Lupker, S. J. (2006). Is there a neighborhood frequency effect in English? Evidence from reading and lexical decision. Journal of Experimental Psychology: Human Perception and Performance, 32(4), 1040–1062. https://doi.org/ 10.1037/0096-1523.32.4.1040 Sears, C. R., Hino, Y., & Lupker, S. J. (1995). Neighborhood size and neighborhood frequency effects in word recognition. Journal of Experimental Psychology: Human Perception and Performance, 21(4), 876–900. https://doi.org/10.1037/00961523.21.4.876 Shook, A., & Marian, V. (2012). Bimodal bilinguals co-activate both languages during spoken comprehension. Cognition, 124(3), 314–324. https://doi.org/10.1016/j. cognition.2012.05.014 Shook, A., & Marian, V. (2013). The bilingual language interaction network for comprehension of speech. Bilingualism: Language and Cognition, 16(2). https://doi. org/10.1017/S136672891200046 Siemund, P. (2023). Multilingual development: English in a global context. Cambridge University Press. Siew, C. S. Q. (2013). Community structure in the phonological network. Frontiers in Psychology, 4, 553. https://doi.org/10.3389/fpsyg.2013.00553 Siew, C. S. Q., & Castro, N. (2023). Phonological similarity judgments of word pairs reflect sensitivity to large-scale structure of the phonological lexicon. Journal of Experimental Psychology: Learning, Memory, and Cognition, 49(12), 1989–2002. https://doi.org/10.1037/xlm0001271 Siew, C. S. Q., & Vitevitch, M. S. (2016). Spoken word recognition and serial recall of words from components in the phonological network. Journal of Experimental Psychology: Learning, Memory, and Cognition, 42(3), 394–410. https://doi.org/ 10.1037/xlm0000139 Sommers, M. S., & Lewis, B. P. (1999). Who really lives next door: Creating false memories with phonological neighbors. Journal of Memory and Language, 40(1), 83–108. https://doi.org/10.1006/jmla.1998.2614 Spivey, M., & Marian, V. (1990). Cross-talk between native and second languages: Partial activation of an irrelevant lexicon. Psychological Science, 10, 281–284. https://doi. org/10.1111/1467-9280.00151 Storkel, H. L., Armbruster, J., & Hogan, T. P. (2006). Differentiating phonotatic probability and neighborhood density in adult word learning. Journal of Speech, Language, and Hearing Research, 49, 1175–1192. https://doi.org/10.1044/10924388(2006/085) Tucker, B. V., Brenner, D., Danielson, D. K., Kelley, M. C., Nenadic, F., & Sims, M. (2019). The massive auditory lexical decition (MALD) database. Behavior Research Methods, 51, 1187–1204. https://doi.org/10.3758/s13428-018-1056-1 Vitevitch, M. (1997). The Neighborhood Characteristics of Malapropisms. Language and Speech, 40, 211–228. https://doi.org/10.1177/002383099704000301 Vitevitch, M. (2007). The spread of the phonological neighborhood influences spoken word recognition. Memory & Cognition, 35, 166–175. https://doi.org/10.3758/ bf03195952 Vitevitch, M., & Castro, N. (2015). Using network science in the language sciences and clinic. International Journal of Speech-Language Pathology, 17, 13–25. https://doi.org/ 10.3109/17549507.2014.987819 Vitevitch, M., & Goldstein, R. (2014). Keywords in the mental lexicon. Journal of Memory and Language, 73, 131–147. https://doi.org/10.1016/j.jml.2014.03.005 Vitevitch, M., & Luce, P. (1998). When words compete: Levels of processing in perception of spoken words. Psychological Science, 9, 325–329. https://doi.org/10.1111/14679280.00064 Vitevitch, M., & Luce, P. (1999). Probabilistic phonotactics and neighborhood activation in spoken word recognition. Journal of Memory and Language, 40, 374–408. https:// doi.org/10.1006/jmla.1998.2618 Vitevitch, M., & Luce, P. (2004). A web-based interface to calculate phonotactic probability for words and nonwords in English. Behavior Research Methods, Instruments, & Computers, 36, 481–487. https://doi.org/10.3758/bf03195594 Vitevitch, M., & Luce, P. (2016). Phonological neighborhood effects in spoken word perception and production. Annual Review of Linguistics, 2, 75–94. https://doi.org/ 10.1146/annurev-linguistics-030514-124832 Vitevitch, M., Siew, C. S. Q., & Castro, N. (2018). Spoken word recognition. In S.- A. Rueschemeyer, & M. G. Gaskell (Eds.), The Oxford handbook of psycholinguistics (pp. 31–47). Oxford University press. https://doi.org/10.1093/oxfordhb/ 9780198786825.013.2. Vitevitch, M., & Sommers, M. (2003). The facilitative influence of phonological similarity and neighborhood frequency in speech production. Memory & Cognition, 31, 491–504. https://doi.org/10.3758/BF03196091 Wagenmakers, E. J., Ratcliff, R., Gomez, P., & McKoon, G. (2008). A diffusion model account of criterion shifts in the lexical decision task. Journal of Memory and Language, 58(1), 140–159. https://doi.org/10.1016/j.jml.2007.04.006 Watts, D. J., & Strogatz, S. H. (1998). Collective dynamics of ‘small-world’ networks. Nature, 393, 440. https://doi.org/10.1038/30918 Weber, A., & Cutler, A. (2004). Lexical competition in non-native spoken-word recognition. Journal of Memory and Language, 50(1), 1–25. https://doi.org/10.1016/ S0749-596X(03)00105-0 Wen, Y., & van Heuven, W. J. B. (2017). Non-cognate translation priming in masked priming lexical decision experiments: A meta-analysis. Psychonomic Bulletin & Review, 24(3), 879–886. https://doi.org/10.3758/s13423-016-1151-1 Whelan, R. (2008). Effective analysis of reaction time data. Psychological Review, 58, 475–482. https://doi.org/10.1007/BF03395630 Yates, M. (2013). How the clustering of phonological neighbors affects visual word recognition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 39 (5), 1649–1656. https://doi.org/10.1037/a0032422 Yates, M., & Dickinson, D. (2023). How similarity influences word recognition. In J. Guendouzi, F. Loncke, & M. J. Williams (Eds.), The Routledge international handbook of psycholinguistic and cognitive processes (pp. 273–290). Routledge. Ziegler, J. C., Muneaux, M., & Grainger, J. (2003). Neighborhood effects in auditory word recognition: Phonological competition and orthographic facilitation. Journal of Memory and Language, 48, 779–793. https://doi.org/10.1016/S0749-596X(03) 00006-8 E.M. Luef Acta Psychologica 261 (2025) 105863 21