Figures
1.2Average formant trajectories of monophthongs by dialect in the F2–F1 and F2–F3 space.
1.3Sample spectrograms of /y/ and /ø/, as produced by Seoul and Dandong speakers.
1.5Formant rises for F1, F2, and F3 for /y/ and /ø/ by age, gender, and dialect.
1.8Average formant tracks of Mandarin /y/ by Mandarin and heritage Korean speakers in Dandong and Hunchun.
1.9Self-reported language background by dialect, gender, and year of birth.
2.2Results across groups for American English voiceless stops.
3.1Comparison of a phrase in three types of Japanese script and in hiragana only.
4.1A typical broad focus VSO utterance from a female speaker of Samoan, from Calhoun (Reference Calhoun2015, p. 215).
4.2A typical subject focus SVO utterance in Samoan, from Calhoun (Reference Calhoun2015, p. 218).
4.3Picture for Na toso e Sione le maea analeilā (‘Sione pulled a rope’), adapted from Calhoun (Reference Calhoun2015, p. 212).
4.4Typical object focus realization for Alofa and Lupe: ‘Sione ate the jam last night.’
4.5Typical subject focus realization for Alofa and Lupe: ‘It was Kalolo who painted the bike earlier.’
4.6A broad focus utterance showing both H- and an innovative high plateau: ‘Iona went to catch the dog at the park.’
4.7Typical object focus realization for Rosita: ‘No, Moana swept the courtyard yesterday.’
4.8Typical subject focus realization for Rosita: ‘Kalolo hugged the girl yesterday.’
4.9Typical broad focus realization for Rosita: ‘Moana swept the courtyard yesterday.’
4.10Example of an object focus realization by Leilani: ‘Iona carried the spade.’
4.11Example of a subject focus realization by Leilani: ‘Kalolo mixed the cement last night.’
4.12Example of a broad focus realization by Leilani: ‘Leona carried the chair.’
5.1Count of each loanword adaptation pattern type by speaker.
5.2Waveform and spectrogram of ‘elephant’ illustrating two adaptation patterns, as produced by speakers f2 and f3.
5.4Proportion of pattern C and D tokens (out of the total of C and D tokens produced) by speaker.
5.5Scatterplot of vocabulary size and count of pattern D tokens by speaker.
5.6Overlapping constraints on a continuous ranking scale in three different scenarios.
6.1Boxplot of z-scored F2–F1 values by position for each generation*language combination.
6.2Conditional inference tree fitted to z-scored F2–F1 Punjabi data with the predictors ‘position’ and ‘generation.’
6.3Conditional inference tree fitted to z-scored F2–F1 English data with the predictors ‘position’ and ‘generation.’
6.4Conditional inference trees fitted to raw F2–F1 (Hz) data with the predictor variables ‘position’ and ‘language.’
7.1Spectrogram showing the realization of word-initial /ɡ/ with partial prevoicing in the Russian item god (‘year’).
7.2Range of VOT values for the plosives /p/, /t/, and /k/ in heritage Polish and Russian compared to monolingual VOT values mentioned in previous literature.
7.3Frequency of voicing types (i.e., prevoiced, partially prevoiced, and devoiced) in the production task with lenis stops.
8.1Waveform, spectrogram, and F0 trace illustrating stress shift in heritage Spanish (contrastive focus on the DO).
8.2Waveform, spectrogram, and F0 trace illustrating p-movement in heritage Spanish (information focus on the DO).
8.3Waveform, spectrogram, and F0 trace illustrating a simple cleft in heritage Spanish (information focus on the S).
8.4Waveform, spectrogram, and F0 trace illustrating a simple cleft in heritage Spanish (information focus on the S).
8.5Waveform, spectrogram, and F0 trace illustrating stress shift in heritage Spanish (information focus on the S).
9.1Sample waveform and spectrogram for a full word C1VC2V and a /ɡu/–/ku/ VOT continuum of C2.
9.2Mean percentages of Korean, Mandarin, English, and heritage speakers’ judgments along the VOT continuum of Korean stops.
9.3Mean percentages of responses along the VOT continuum of Korean stops for individual heritage speakers.
9.4Mean percentages of responses along the VOT continuum of Korean stops for the four speaker groups by stop type.
10.1Variable groups and categorical variables (by level) in MFA1.
10.2Representations of individuals according to demographic variables in MFA1.
10.3Variable groups, categorical variables (by level), and numeric variables in MFA2.
10.4Representations of heritage speaker individuals according to demographic variables in MFA2.
10.5Variable groups and Mandarin phonetic (segmental and tonal) variables in MFA3.
10.6Representations of individuals according to Mandarin phonetic variables in MFA3.
10.7Representations of heritage speaker and native Mandarin individuals by Mandarin variety in MFA3.
10.9Representations of individuals according to English phonetic variables in MFA4.
11.5Words from the perception experiment by mean nativeness rating.
11.6Duration of singleton and geminate consonants produced by each speaker group.
11.7Duration of singleton and geminate consonants based on voicing.
11.8Duration of singleton and geminate consonants based on manner of articulation.
12.1Distribution of Icelandic nuclear pitch accent types in declaratives per group.
12.2Modern Icelandic declarative with the nuclear contour L+H* L%.
12.3Distribution of Icelandic nuclear pitch accent types in polar questions per group.
12.4Distribution of Icelandic boundary tones in polar questions per group.
12.5Distribution of Icelandic nuclear contours in polar questions per group.
12.6Modern Icelandic polar question with an L+H* nuclear accent and an L% boundary tone.
12.7North American Icelandic polar question with an L* nuclear accent and an H% boundary tone.
13.2Boxplots showing /y/~/u/ Pillai scores by generational group.
13.3The /y/~/u/ Pillai scores at the group level based on dominant language.
13.4Boxplots showing /a/~/ɔ/ Pillai scores by generational group.
13.5The /a/~/ɔ/ Pillai scores at the group level based on dominant language.
13.7Plot for C2F21B, the Gen2 speaker with the highest Pillai score.
13.8Plot for C2M22A, the Gen2 speaker with the lowest Pillai score.
14.1Tonal space for Gen0, Gen1, and Gen 2 normalized by duration and pitch.
15.1Polish [ɡ] in <tego> (‘this’) with partial voicing in the closure.
15.2Unnormalized voicing percentage in voiced versus voiceless stops by generation.
15.3Speaker-normalized preceding vowel duration before voiced versus voiceless stops by generation.
15.4Speaker-normalized closure duration before voiced versus voiceless stops by generation.
15.5Speaker-normalized release duration by place of articulation and generation.
15.6Speaker-normalized release duration by voicing and generation.
16.4Mean individual VOT values of Portuguese voiceless stops produced by the heritage speaker group in pseudowords.
16.5Mean individual percent fit indexes of voiceless Portuguese stops to English stop categories.
17.1Rhotic durations (and standard errors) by position in the syllable and stress.
17.2Rhotic durations (and standard errors) in singleton onsets by position in the word and stress.
17.3Rhotic durations (and standard errors) in codas by position in the word and stress.
17.4Production of a trill in the phrase na recia (‘an ear’) in Brazilian Veneto.
17.5Production of a tap in the word bareta (‘cap’) in Brazilian Veneto.
17.6Production of an approximant in the word naransa (‘orange’) in Brazilian Veneto.
17.7Production of a fricative in the phrase un rospo (‘a toad’) in Brazilian Veneto.
