Learning linguistic diversity: Listeners have race-based linguistic expectations, but only for phonological variation.

In three artificial language experiments, we explored the rate at which adults learned associations between linguistic variation and speaker characteristics. Within each of the experiments, we observed that listeners sociolinguistic learning occurred, regardless of whether the speaker characteristic is social (race and sex/gender) or nonsocial (hat wearing), or whether they heard a phonological or morphological variant. However, we found that listener's initial expectations of what social properties were predictive of linguistic variation differed, impacting overall performance. First, participants were much more likely to assume that a phonological variant was predicted by a social property than a nonsocial property (Experiment 1). Most interestingly, participants were more likely to privilege speaker race than sex/gender, but only in the case of a phonological variant (Experiments 2 and 3). The same effect was found in both White and Black participants, though White participants were more likely to correctly articulate which speaker characteristic explained the variation, suggesting that sociolinguistic learning hinges on real-world experiences with language and social diversity. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

20-month-olds selectively generalize newly learned word meanings based on cues to linguistic community membership.

A growing body of work suggests that speaker-race influences how infants and toddlers interpret the meanings of words. In two experiments, we explored the role of speaker-race on whether newly learned word-object pairs are generalized to new speakers. Seventy-two 20-month-olds were taught two word-object pairs from a familiar race speaker, and two different word-object pairs from an unfamiliar race speaker (four new pairs total). Using an intermodal preferential looking procedure, their interpretation of these new word-object pairs was tested using an unpictured novel speaker. We found that toddlers did not generalize word meanings taught by an unfamiliar race speaker to a new speaker (Experiment 1), unless given evidence that the unfamiliar race speaker was a member of the child's linguistic community through affiliative behaviour and linguistic competence (Experiment 2). In both experiments, generalization was observed for the word-object pairs taught by the familiar race speaker. These experiments indicate that children attend to speakers' non-linguistic properties, and this, in turn, can influence the perceived relevance of speakers' labels.

Toddlers link social and speech variation during word learning.

Within a language, there is considerable variation in the pronunciations of words owing to social factors like age, gender, nationality, and race. In the present study, we investigate whether toddlers link social and linguistic variation during word learning. In Experiment 1, 24- to 26-month-old toddlers were exposed to two talkers whose front vowels differed systematically. One talker trained them on a word-referent mapping. At test, toddlers saw the trained object and a novel object; they heard a single novel label from both talkers. Toddlers responded differently to the label as a function of talker. The following experiments demonstrate that toddlers generalize specific pronunciations across speakers of the same race (Experiment 2), but not across speakers who are simply an unfamiliar race (Experiment 3). They also generalize pronunciations based on previous affiliative behavior (Experiment 4). When affiliative behavior and race are pitted against each other, toddlers' linguistic interpretations are more influenced by affiliative behavior (Experiment 5). These experiments suggest that toddlers attend to and link social and speech variation in their environment. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Putting Mutual Exclusivity in Context: Speaker Race Influences Monolingual and Bilingual Infants' Word-Learning Assumptions.

Previous work indicates mutual exclusivity in word learning in monolingual, but not bilingual toddlers. We asked whether this difference indicates distinct conceptual biases, or instead reflects best-guess heuristic use in the absence of context. We altered word-learning contexts by manipulating whether a familiar- or unfamiliar-race speaker introduced a novel word for an object with a known category label painted in a new color. Both monolingual and bilingual infants showed mutual exclusivity for a familiar-race speaker, and relaxed mutual exclusivity and treated the novel word as a category label for an unfamiliar-race speaker. Thus, monolingual and bilingual infants have access to similar word-learning heuristics, and both use nonlinguistic social context to guide their use of the most appropriate heuristic.

The Role of Audiovisual Speech in Fast-Mapping and Novel Word Retention in Monolingual and Bilingual 24-Month-Olds.

Three experiments examined the role of audiovisual speech on 24-month-old monolingual and bilinguals' performance in a fast-mapping task. In all three experiments, toddlers were exposed to familiar trials which tested their knowledge of known word-referent pairs, disambiguation trials in which novel word-referent pairs were indirectly learned, and retention trials which probed their recognition of the newly-learned word-referent pairs. In Experiment 1 (n = 48), lip movements were present during familiar and disambiguation trials, but not retention trials. In Experiment 2 (n = 48), lip movements were present during all three trial types. In Experiment 3 (bilinguals only, n = 24), a still face with no lip movements was present in all three trial types. While toddlers succeeded in the familiar and disambiguation trials of every experiment, success in the retention trials was only found in Experiment 2. This work suggests that the extra-linguistic support provided by lip movements improved the learning and recognition of the novel words.

Preschoolers are sensitive to accent distance.

Can children tell how different a speaker's accent is from their own? In Experiment 1 (N = 84), four- and five-year-olds heard speakers with different accents and indicated where they thought each speaker lived relative to a reference point on a map that represented their current location. Five-year-olds generally placed speakers with stronger accents (as judged by adults) at more distant locations than speakers with weaker accents. In contrast, four-year-olds did not show differences in where they placed speakers with different accents. In Experiment 2 (N = 56), the same sentences were low-pass filtered so that only prosodic information remained. This time, children judged which of five possible aliens had produced each utterance, given a reference speaker. Children of both ages showed differences in which alien they chose based on accent, and generally rated speakers with foreign accents as more different from their native accent than speakers with regional accents. Together, the findings show that preschoolers perceive accent distance, that children may be sensitive to the distinction between foreign and regional accents, and that preschoolers likely use prosody to differentiate among accents.

Children's accent-based inferences depend on geographic background.

Young children make inferences about speakers based on their accents. Here, we show that these accent-based inferences are influenced by information about speakers' geographic backgrounds. In Experiment 1, 4- to 6-year-olds (N = 60) inferred that a speaker would be more likely to have the same cultural preferences as another speaker with the same accent than a speaker with a different accent; in Experiment 2 (N = 90), children made similar inferences about speakers' friendship preferences. Critically, in both experiments, children were less likely to make accent-based inferences when they were told that the speakers all came from different places (both experiments) or from the same place (Experiment 2). These results suggest that young children's accent-based inferences hinge on information about geographic background and provide insight into how and why children make accent-based inferences. These findings are also the first to show that young children use accent to infer other people's social preferences.

And then I saw her race: Race-based expectations affect infants' word processing.

How do our expectations about speakers shape speech perception? Adults' speech perception is influenced by social properties of the speaker (e.g., race). When in development do these influences begin? In the current study, 16-month-olds heard familiar words produced in their native accent (e.g., "dog") and in an unfamiliar accent involving a vowel shift (e.g., "dag"), in the context of an image of either a same-race speaker or an other-race speaker. Infants' interpretation of the words depended on the speaker's race. For the same-race speaker, infants only recognized words produced in the familiar accent; for the other-race speaker, infants recognized both versions of the words. Two additional experiments showed that infants only recognized an other-race speaker's atypical pronunciations when they differed systematically from the native accent. These results provide the first evidence that expectations driven by unspoken properties of speakers, such as race, influence infants' speech processing.

Accent, Language, and Race: 4-6-Year-Old Children's Inferences Differ by Speaker Cue.

Three experiments examined 4- to 6-year-olds' use of potential cues to geographic background. In Experiment 1 (N = 72), 4- to 5-year-olds used a speaker's foreign accent to infer that they currently live far away, but 6-year-olds did not. In Experiment 2 (N = 72), children at all ages used accent to infer where a speaker was born. In both experiments, race played some role in children's geographic inferences. Finally, in Experiment 3 (N = 48), 6-year-olds used language to infer both where a speaker was born and where they currently live. These findings reveal critical differences across development in the ways that speaker characteristics are used as inferential cues to a speaker's geographic location and history.

Read my lips: Visual speech influences word processing in infants.

What do infants hear when they read lips? In the present study, twelve-to-thirteen-month-old infants viewed a talking face produce familiar and unfamiliar words. The familiar words were of three types: in Experiment 1, they were produced correctly (e.g., "bottle"); in Experiment 2, infants saw and heard mispronunciations in which the altered phoneme either visually resembled the original phoneme (visually consistent, e.g. "pottle"), or did not visually resemble the original phoneme (visually inconsistent, e.g., "dottle"). Infants in the correct and consistent conditions differentiated the familiar and unfamiliar words, but infants in the inconsistent condition did not. Experiment 3 confirms that infants were sensitive to the mispronunciations in the consistent condition with auditory-only words. Thus, although infants recognized the consistent mispronunciations when they saw a face articulating the words, they did not with the auditory information alone. These results provide the first evidence that visual articulatory information affects word processing in infants.

Where are you from? Preschoolers infer background from accent.

For adults, accent is an obvious indicator of a speaker's geographical background. The current study investigated whether preschoolers are sensitive to the relationship between background and accent. Experiment 1 shows that 3- to 5-year-olds believe that two speakers who share the same accent live in the same place but do not share the same personal preferences. Experiment 2 demonstrates that 4- and 5-year-olds believe that two speakers with the same accent share cultural norms associated with a particular place, but that two speakers with different accents have different cultural norms. As in Experiment 1, children did not think that personal preferences were related to accent. These findings show early awareness of the relationship between accent and geographical background.