Word Pairs Research Articles

Emotion schema effects on associative memory differ across emotion categories at the behavioural, physiological and neural level

Previous behavioural and neuroimaging studies have consistently reported that memory is enhanced for associations congruent or incongruent with the structure of prior knowledge, termed as schemas. However, it remains unclear if similar effects arise with emotion-related associations, and whether they depend on the type of emotions.Here, we addressed this question using a novel face-word pair association paradigm combined with fMRI and eye-tracking techniques. In two independent studies, we demonstrated and replicated that both congruency with emotion schemas and emotion category interact to affect associative memory. Overall, memory retrieval was higher for faces from pairs congruent vs. incongruent with emotion schemas, paralleled by a greater recruitment of left inferior frontal gyrus (IFG) during successful encoding. However, emotion schema effects differed across two negative emotion categories. Disgust was remembered better than fear, and only disgust activated left IFG stronger during encoding of congruent vs. incongruent pairs, suggestive of deeper semantic processing for the associations. On the contrary, encoding of congruent fear vs. disgust-related pairs was accompanied with greater activity in right fusiform gyrus (FG), suggesting a stronger sensory processing of faces. In addition, successful memory formation for congruent disgust pairs was associated with a higher pupil dilation index related to sympathetic activation, longer gaze time on words compared to faces, and more gaze switches between paired words and faces. This was reversed for fear-related congruent pairs where the faces attracted longer gaze time (compared to words).Overall, our results provide converging evidence from behavioural, physiological, and neural measures to suggest that congruency with available emotion schemas influence memory associations in a similar manner to semantic schemas. However, these effects vary across distinct emotion categories, pointing to a differential role of semantic processing and visual attention processes in the modulation of memory by disgust and fear, respectively.

Assessing the Effect of Restudying Anatomy with Flashcards versus Handouts in Study Groups Equalized by Memory Test Scores

IntroductionRetrieval practice through testing can strengthen long‐term memory. Experimentally, this concept known as the “testing effect” has often been demonstrated using word pairs. As many health science students quiz with flashcards to enhance recall, can the testing effect be detected within the more complex study of anatomy? It was hypothesized that participants utilizing flashcards for review would score higher on tests than those restudying a handout.MethodsStudents were recruited from Pacific Northwest University of Health Sciences (PNWU). On day 1 of the study, subjects were given 3 memory tests (assessing memory of printed words, pictures, and spoken words), the scores of which were combined and used to assign participants to groups with approximately equal “memory.” Participants studied a handout on ruminant digestive anatomy, a topic novel to most students. After the initial study session, subjects were divided into a “flashcard group” (FG) or a “handout group” (HG) and reviewed the learned material via their assigned method. Participants then took a 15‐question test, indicating whether they were confident or not for each answer, and took a similar test 1 and 3 weeks later. Participants completed surveys on demographics, existing knowledge of ruminants and anatomical terms, wellness, and study behaviors on day 1 of the study. Data were analyzed using Mann Whitney U test, Fischer’s exact test, Pearson Correlation Coefficient, and repeated measures ANOVA using SPSSv25.0 software with p<0.05 considered statistically significant. This study was deemed exempt by PNWU IRB (#2021‐015).ResultsThough not statistically so, mean scores were higher in the FG than in the HG for each of the ruminant digestive anatomy assessments. For day 1 test score means (FG 10.67 ± 1.50 vs HG 9.00 ± 2.68), a large effect size (ηp2=0.16) was detected. While the FG started with more experience with anatomical terms (p=0.046), this experience was not significantly correlated with day 1 test scores (n=17). Though the FG tended to perform better on day 1 tests, the HG was more confident in their answer choices (mean confidence scores HG 10.00 ± 1.10 vs FG 7.78 ± 1.72), with a large effect size present (ηp2=0.38). Memory scores were not significantly different between the FG and the HG. Though not significant, most memory scores were negatively correlated with age. Sleep was negatively correlated with memory scores, which was significant for memory of printed words (r=‐0.56; p=0.019). Female memory scores consistently ranked higher than male scores.ConclusionsParticipants who restudied with flashcards tended to outperform those restudying handouts on tests covering a novel anatomy topic, suggesting that the testing effect may have occurred. Interestingly, the opposite pattern was seen in confidence, with those studying handouts seemingly more confident in their test answers. Utilizing memory test scores resulted in evenly distributed groups with respect to memory, and some intriguing potential differences were seen related to sleep on the night before day 1 of the study, age, and gender. This study was limited by a small sample size and should be expanded to further investigate the effect of various restudy techniques, such as using flashcards or handouts while learning anatomy.

Transfer of Test-Enhanced Learning.

Memory for paired-associate words is facilitated by interim testing relative to restudy. According to the mediator effectiveness hypothesis, the benefit of retrieval practice is a consequence of the activation of a mediator word linking the cue and target. Evidence for the activation of cue-related mediators stems from the finding that mediators are more effective at prompting recall of target words than are words not associated with the original cue, a pattern that is larger following testing than restudy. The benefit of testing for the unstudied cues at the final test is referred to as transfer of test-enhanced learning. One goal of the current study was to examine whether the activation of mediators leads to the recall of targets indirectly via the original cues in a process known as backward chaining. We indexed backward chaining with the probability of incorrectly recalling a trial-specific original cue in place of a target. The second goal was to explore whether testing would yield a transfer effect for cues associated with target words. In four experiments, following an initial study of weakly related word pairs (e.g., Mother-CHILD), participants either restudied the pairs or attempted to recall the target given the original cue (e.g., Mother). On a final cued-recall test, participants were presented with unstudied cues that were related to either the original cue (semantic mediators, e.g., Father) or the target (target-related cues, e.g., Baby). The type of new cue presented on the final test was varied either between subjects (Experiment 1) or mixed within a list (Experiments 2, 3, and 4). Mixing mediators and target-related cues reduced the transfer of test-enhanced learning and increased the likelihood of recalling the original cues when shown a mediator. These results challenge the assumptions of the mediator effectiveness hypothesis.

Variasi Konsonan Bahasa Jawa pada Masyarakat Kelas Bawah di Daerah Istimewa Yogyakarta

This study aims to determine the variation of Javanese consonants in the lower-class community in RT 34-35 Giwangan Village, Umbulharjo District, Special Region of Yogyakarta. This type of research is a case study, data obtained utilizing interviews and literature study. The study results indicate that there are variations in consonants that occur in the lower society. The consonant variations are a form of neutral in the word pairs /b/ and /p/, which occurs in the word "landep", which means sharp; in the word pairs /d/ and /t/ occurs in the word "reged" which in Indonesian means dirty, and the word pairs /g/ and /k/ occurs in the word "ambleg" which means to fall. The next variation is the insertion and addition of consonants /n/, /ŋ/, and /m/in the word "montor", which means motorbikes. The next variation is the insertion of semi-vowels /w/ and /y/, which occurs in the word "kebuwang", which means wasted, and in the word "lagèYane", which has the meaning of style. The next variation is the variation of the consonants /z/, /dz/ and /v/, /f/ that occurs in the word "jikir", which means remembrance and the word TV, which is pronounced by with the word "tipi". Next is a shortening variation that occurs in the sentence "no money", which is pronounced by "ra de det" or "ora nduwe duwit". The last consonant variation is the omission of sounds that occur in the word "government", which means government.

Acquisition of emphatic consonants by Ammani Arabic-speaking children

ABSTRACT This article investigates the production of the emphatic consonants /ṭ, ḍ, ṣ/ by typically developing Jordanian children. Sixty typically developing monolingual Ammani Arabic-speaking children (30 boys and 30 girls) with ages ranging from 2 to 7;11 years were recruited in a production experiment. In the experiment, they were asked to produce 18 minimal pair words with emphatic consonants and their corresponding plain coronals in all word positions. Contrary to Amayreh & Dyson (1998) and Hamdan & Amayreh (2007), >50% of Ammani Arabic-speaking children produced emphatic consonants by the age of 3, >75% produced them by the age of 4, and >90% produced them by the age of 5. Acoustically, they produce them in an adult-like manner at the age of 6 word-initially and medially and at the age of 7 word-finally. The acoustic measurements confirm that children’s productions become increasingly adult-like with age. Compared with nonemphatic consonants, the appearance of emphatics tends to be delayed in Ammani Arabic. This delay in the production of emphatics, which are among the least frequently occurring consonants in Arabic, is likely due to their articulatory complexity, which involves a secondary co-articulation (the so-called pharyngealization). The current findings help clarify the developmental trajectory of children’s acquisition of emphatic consonants and should be informative for both researchers and clinicians.

The role of sleep for episodic memory consolidation: Stabilizing or rescuing?

Prior studies suggest a role for sleep in memory consolidation, with specific contributions from slow oscillations and sleep spindles (Rasch & Born, 2013). However, recent studies failed to replicate sleep’s superiority over wake in strengthening memory against interference (Cordi & Rasch, 2021). The goal of the current study is to investigate whether sleep protects newly formed memory from unspecific interference induced by daytime experiences over 24 h, as well as to elucidate the sleep features that are involved. 56 healthy adults were randomly assigned to either the Sleep First or Wake First group. The Sleep First group encoded word pairs at night before sleep, while the Wake First group encoded word pairs in the morning before a day of wakefulness. Memory was tested 30 min, 12 h, and 24 h after encoding for both groups. The Sleep First group performed significantly better 12 h after encoding, replicating prior findings that memory is better after a period of sleep compared to wake. However, after 24 h, the two groups performed similarly. The Wake First group showed a positive correlation between overnight memory improvement and the theta and delta band power during slow wave sleep, an effect not found in the Sleep First group. These correlations suggest the possibility that after a day of waking interference, the brain recruits extra sleep resources to rescue memories from further forgetting. Our results are not consistent with prior studies showing a significant role for sleep in stabilizing memory from future interference, but they may suggest that sleep rescues memories after interference has occurred.

Semantic flow and its relation to controlled semantic retrieval deficits in the narrative production of people with aphasia

Aphasia has had a profound influence on our understanding of how language is instantiated within the human brain. Historically, aphasia has yielded an in vivo model for elucidating the effects of impaired lexical-semantic access on language comprehension and production. Aphasiology has focused intensively on single word dissociations. Yet, less is known about the integrity of combinatorial semantic processes required to construct well-formed narratives. Here we addressed the question of how controlled lexical-semantic retrieval deficits (a hallmark of aphasia) might compound over the course of longer narratives. We specifically examined word-by-word flow of taxonomic vs. thematic semantic distance in the storytelling narratives of individuals with chronic post-stroke aphasia (n = 259) relative to age-matched controls (n = 203). We first parsed raw transcribed narratives into content words and computed inter-word semantic distances for every running pair of words in each narrative (N = 232,490 word transitions). The narratives of people with aphasia showed significant reductions in taxonomic and thematic semantic distance relative to controls. Both distance metrics were strongly predictive of offline measures of semantic impairment and aphasia severity. Since individuals with aphasia often exhibit perseverative language output (i.e., repetitions), we performed additional analyses with repetitions excluded. When repetitions were excluded, group differences in semantic distances persisted and thematic distance was still predictive of semantic impairment, although some findings changed. These results demonstrate the cumulative impact of deficits in controlled word retrieval over the course of narrative production. We discuss the nature of semantic flow between words as a novel metric of characterizing discourse and elucidating the nature of lexical-semantic access impairment in aphasia at multiword levels.

Children's disambiguation of novel words varies by the number and position of phonological contrasts.

Young children often make pragmatic assumptions when learning new words. For example, they assume that a speaker who uses different words intends to refer to different things - the so-called principle of contrast. We used a standard disambiguation task to explore whether children's assumptions about contrast depend on how much words differ. Three- to 6-year-olds heard pairs of words that differed in terms of the number, position, and types of phonological contrasts. Results indicate that children were less likely to disambiguate words differing by one phoneme than words differing by two or more phonemes, particularly when those one-phoneme differences were located at the beginning or end of the words (as in fim/vim). Overall, the findings suggest that children's pragmatic assumptions about two contrasting words depend not only on if words differ, but also on how they differ.

Lexical restructuring stimulates phonological awareness among emerging English-French bilingual children's literacy.

This longitudinal study investigated how lexical restructuring can stimulate emerging bilingual children's phonological awareness in their first (L1) and second (L2) languages. Sixty-two English (L1) - French (L2) bilingual children (Mage = 75.7 months, SD = 3.2) were taught new English and French word pairs differing minimally in phonological contrast. The results indicated that increasing lexical specificity in English mediated the relationship between English vocabulary and English phonological awareness both concurrently and longitudinally at the end of Grade 1. A longitudinal relationship was established among French vocabulary, French lexical specificity, and French phonological awareness at the end of Grade 1. Notably, cross-language transfer from English lexical specificity was a better predictor of development in French phonological awareness, especially for words that contained phonological contrasts that occurred in both languages. The results from this study highlight the phonological foundations of early literacy and extend the lexical restructuring hypothesis to emerging bilingual children.

When the sense of fluency triggers an attentional bias.

Spatial attention can be captured automatically by an exogenous stimulus (e.g., digital interruption) or by an endogenous stimulus (e.g., valence of the stimulus). In this study, we investigated whether a non-perceptual characteristic (e.g., sense of fluency) has an impact on attention. To this end, we used the conceptual fluency paradigm developed by Whittlesea combined with the dot-probe task developed by MacLeod et al. In three experiments, we measured the response times for each experimental situation (i.e., Valid and Non-valid situations). At each trial, participants were presented in three consecutive displays on a screen: (1) an incomplete and predictive sentence stem; (2) a pair of words, one of which was semantically compatible with the previous sentence stem; and (3) a circle appeared at the spatial location of one of the words. Then, participants had to perform a Go (i.e., a filled circle) and No-go (i.e., an empty circle) task. The analysis found that response times were significantly faster when the Go stimulus appeared at the same location as the semantically compatible word (i.e., Valid situations). Overall, our results show that the sense of fluency triggers attentional capture. Thus, they replicate those of Gardner et al. using another experimental paradigm. Our finding might be helpful to better understand the consequences of digital interruptions on behavioural performance.

The frequency of word gender as a variable for lexical access in Spanish

Abstract The endings of Spanish nouns reflect gender with varying degrees of frequency and regularity. The most common and regular endings are -o for masculine nouns and -a for feminine nouns, -o being more frequent and less closely associated with a specific gender category (masculine) than -a. Pairs of words occurring with both gender categories differ in the frequencies with which they are used as masculine or feminine forms: médic-o/médic-a ‘doctor’ (m.)/‘doctor’ (f.) is a clear example of a masculine-dominant pair, whereas enfermer-o/enfermer-a ‘nurse’ (m.)/ ‘nurse’ (f.) is a feminine-dominant pair. Adult readers of Spanish are faster in recognizing feminine forms of feminine-dominant pairs, and masculine forms of masculine-dominant pairs (Dominguez, Alberto, Fernando Cuetos & Juan Segui. 1999. The processing of grammatical gender and number in Spanish. Journal of Psycholinguistic Research 28(5). 485–498). This study aims to test the dominant frequency effect in third and sixth grade children, as well as in adults. Children were faster in recognizing masculine forms in masculine-dominant pairs, but not feminine forms in feminine-dominant pairs. Adults, by contrast, tended to respond faster to higher frequency words, irrespective of gender, indicating that they have independent representations for both genders. The dominance of masculine forms in children could be a consequence of the statistical distribution of gender dominance and regularity in Spanish. The experience of skilled adult readers seems to make them less dependent on this statistical pattern.

Affection Enhanced Relational Graph Attention Network for Sarcasm Detection

Sarcasm detection remains a challenge for numerous Natural Language Processing (NLP) tasks, such as sentiment classification or stance prediction. Existing sarcasm detection studies attempt to capture the subtle semantic incongruity patterns by using contextual information and graph information through Graph Convolutional Networks (GCN). However, direct application of dependence may inevitably introduce noisy information and inferiorly in modeling long-distance or disconnected words in the dependency tree. To better learn the sentiment inconsistencies between terms, we propose an Affection Enhanced Relational Graph Attention network (ARGAT) by jointly considering the affective information and the dependency information. Specifically, we use Relational Graph Attention Networks (RGAT) to integrate relation information guided by a trainable matrix of relation types and synchronously use GCNs to integrate affection information explicitly donated by affective adjacency matrixes. The employment of RGAT contributes to information interaction of structural relevant word pairs with a long distance. With the enhancement of affective information, the proposed model can capture complex forms of sarcastic expressions. Experimental results on six benchmark datasets show that our proposed approach outperforms state-of-the-art sarcasm detection methods. The best-improved results of accuracy and F1 are 4.19% and 4.33%, respectively.

Influences of talker gender on second language speech perception

This study examined how talker gender information may influence second language speech perception. The participants were ten right-handed male Japanese adults with normal hearing. The speech stimuli were two minimal pairs of English words, “heed-hod” and “head-had”, recorded from twenty male and twenty female native English speakers. The stimuli were delivered binaurally at a sensation level of 50 dB. In the gender recognition task, the subjects were asked to judge talker gender. In the vowel recognition task, they were asked to identify the vowel category. Magnetoencephalography data were recorded during the experiment. For gender recognition, there was no significant difference in identification accuracy between the two word pairs while reaction time was significantly faster for the difficult “head-had” pair. In the phoneme recognition task, there was a significant phoneme*gender interaction for the difficult “head-had” pair but not for the easy “heed-hod” pair. The MEG data for gender recognition showed right hemisphere dominance, and the phoneme recognition task showed overall bilateral activation. Time-frequency analysis further revealed distinctions between male and female voices and the difficulty differences for the two word pairs in the phoneme recognition task. The results suggest a complex role of paralinguistic gender information in L2 speech perception.

Freeing capacity in working memory (WM) through the use of long-term memory (LTM) representations.

Previous research indicates that long-term memory (LTM) may contribute to performance in working memory (WM) tasks. Across 3 experiments, we investigated the extent to which active maintenance in WM can be replaced by relying on information stored in episodic LTM, thereby freeing capacity for additional information in WM. First, participants encoded word pairs into LTM, and then completed a WM task, also involving word pairs. Crucially, the pairs presented in each WM trial comprised varying numbers of new pairs and the previously learned LTM pairs. Experiment 1 showed that recall performance in the WM task was unaffected when memory set size increased through the addition of LTM pairs, but that it deteriorated when set size increased through adding new pairs. In Experiment 2, we investigated the robustness of this effect, orthogonally manipulating the number of new and LTM pairs used in the WM task. When WM load was low, performance declined with the addition of LTM pairs but remained superior to performance with the matched set size comprising only new pairs. By contrast, when WM load was higher, adding LTM pairs did not affect performance. In Experiment 3, we found that the benefit of LTM representations arises from retrieving these during the WM test, leading them to suffer from typical interference effects. We conclude that individuals can outsource workload to LTM to optimize performance, and that the WM system negotiates the exchange of information between WM and LTM depending on the current memory load. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Enhancing secondary voicing cues in words with minimal pairs

Lexical competition conditions phonetic variation. For example, words with more phonological neighbors are produced with more hyperarticulated vowels than words with fewer neighbors, and /p,t,k/-initial words with voiced stop minimal pairs have longer VOT. Such effects enhance contrastive features and may be aimed at improving perceptibility of more potentially confusable words. Secondary features may also help distinguish words, as with lengthened vowels that cue coda consonant voicing in English. Our study investigates whether secondary features are enhanced in the production of lexically confusable words, comparing vowel duration in words with final voiced consonants between words with and without minimal pair competitors (e.g., bag with minimal pair back vs. gag with no pair gack). Fifteen American English speakers produced 30 monosyllabic English words with voiced final consonants, half with minimal pair competitors and half without (matched for frequency and neighborhood density). Pre-voiced vowels were found to be longer in minimal pair words than in non-minimal pair words; interactions indicate that the effect is stronger for some vowels. These findings show that it is not just contrastive features, but secondary features too that are subject to enhancement in lexical competition contexts, since they too can contribute to successful lexical perception.

Periodic words, common subsequences and frogs

Let W(n) be the n-letter word obtained by repeating a fixed word W, and let Rn be a random n-letter word over the same alphabet. We show several results about the length of the longest common subsequence (LCS) between W(n) and Rn; in particular, we show that its expectation is γWn−O(n) for an efficiently-computable constant γ W. This is done by relating the problem to a new interacting particle system, which we dub “frog dynamics”. In this system, the particles (“frogs”) hop over one another in the order given by their labels. Stripped of the labeling, the frog dynamics reduces to a variant of the PushTASEP. In the special case when all symbols of W are distinct, we obtain an explicit formula for the constant γW and a closed-form expression for the stationary distribution of the associated frog dynamics. In addition, we propose new conjectures about the asymptotic of the LCS of a pair of random words. These conjectures are informed by computer experiments using a new heuristic algorithm to compute the LCS. Through our computations, we found periodic words that are more random-like than a random word, as measured by the LCS.

Perception and imitation of prevoicing across language backgrounds

The present study investigates languages users' ability to perceive and imitate prevoicing, testing speakers of languages differing its contrastive status: (1) those where prevoicing serves as a primary cue to the laryngeal contrast (e.g., Portuguese); (2) those where prevoicing never occurs, with aspiration being the primary cue to the contrast (e.g., Cantonese); and (3) English, where prevoicing occurs in free variation. After hearing pairs of words differing in presence/absence of prevoicing, participants were asked to imitate them, followed by an ABX discrimination task. Preliminary findings show above-chance discrimination accuracy as well as significant imitation of the difference, in participants all language backgrounds, with place of articulation, consonantal, and vocalic cues all affecting the extent of prevoicing in production. As expected, speakers of languages where prevoicing is absent showed less overall prevoicing than other groups; however, there were not clear differences across language groups in the faithfulness of imitation, or in discrimination ability. While faithful imitation was usually associated with accurate discrimination, good discrimination was also found on many tokens that were not faithfully imitated, indicating that factors other than perception are necessary to account for variability in imitative ability.

Metamaterials you can talk to: Speech recognition with elastic neural networks

To detect spoken commands, smart devices (for example, a speaker with Alexa or Siri) continuously convert acoustic waves to electronic signals, translate them into the digital domain, and analyze them in a signal processor. Each of these steps constantly consumes energy, imposing the need for tethered operation or large batteries. We propose to solve this problem using elastic neural networks, metamaterials consisting of arrays of coupled (potentially nonlinear) resonators. The frequencies and couplings of the resonators are optimized to maximise the speech classification accuracy (energy transmitted when excited with one word but not another). Even in purely linear metastructures, we observe binary classification accuracies exceeding 90% for a large number of pairs of words. This is demonstrated on a dataset from a large and diverse group of speakers. To attain these results, we have developed refined modelling techniques involving localised oscillations and machine learning. A unique feature of metamaterial-based speech processing is that speech classification is entirely passive, requiring no external energy. This is possible due to the very low energy dissipation of elastic waves.

Optimal Learning Under Time Constraints: Empirical and Simulated Trade-offs Between Depth and Breadth of Study.

Learners are often constrained by their available study time, typically having to make a trade-off between depth and breadth of learning. Classic experimental paradigms in memory research treat all items as equally important, but this is unlikely the case in reality. Rather, information varies in importance, and people vary in their ability to distinguish what is more or less important. We test the impact of this trade-off in the study of Graduate Record Examination (GRE)-synonym word pairs. In our empirical Study 1, we split our stimuli set, with some items (focal) being afforded more rounds of retrieval practice than other items (non-focal). All conditions had the same total number of trials (i.e., constant study time), but differed in the number of focal words (breadth) and repetitions (depth). The conditions differed significantly in both mean performance and variance on the day-delayed test. Using this empirical data as a base, we then conducted a simulation (Study 2) modeling depth-breadth trade-offs under various conditions of learner forecasting accuracy and test coverage. In Study 2, we found that a medium-depth medium-breadth strategy was appropriate for most of the learning situations covered by our simulation, but that learners with a well-calibrated understanding of importance may benefit from a more targeted high-depth, low-breadth approach. Our results highlight the complexity of navigating the depth-breadth trade-off. Models of learning strategy optimization will need to account for learner forecasting sensitivity, which itself is likely an interaction between relatively stable individual differences and shifting contextual factors.

An ultrasound study of high vowel devoicing in Tokyo Japanese: Evidence for the vowel gesture retention

High vowels in Tokyo Japanese are typically devoiced between voiceless obstruents, but controversy remains over whether vowel gestures persist when devoiced or are instead deleted. A previous ultrasound study (Iwasaki et al., 2020) showed that the lingual articulation of the release burst of /kV/ differs by vowel context even when the vowels are devoiced. This study uses tongue surface contours derived from midsagittal ultrasound images to investigate the effects of vowel devoicing on changes in quantified tongue shape over time. Native speakers of Tokyo Japanese produced word pairs (/C1VC2e/) that contrasted in the voicing of V, which was either /i/ or /u/. Tongue shape was characterized by Fourier transforming tangent angles along each contour (Dawson et al., 2016). Time-normalized trajectories over the /C1VC2e/ sequence were compared by vowel context (/i/ versus /u/) and voicing environment (devoiced versus voiced). Preliminary results show that the real component of the first Fourier coefficient is sensitive todetecting evolving shape differences between the two vowel contexts over the sequence, not just when the vowels are voiced, but also when they are devoiced. Based on these results, the high vowel contrast persists even in the devoiced environment, suggesting that devoiced vowels are not deleted.