Journal of the Acoustical Society of America, Mar 1, 2023
Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—some... more Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—sometimes within minutes of exposure. Two major questions remain open: (a) How does adaptation commence within the initial moments of an encounter?; and (b) How does the adaptation continue via repeated exposure to the same talker/accent? The current study addressed these questions by assessing incremental changes of native English listeners’ recognition of L2-accented speech (Mandarin-accented English) over two timescales: within the first few minutes and over a month. Specifically, we tested how native listeners’ recognition of a word-final /d/, which is initially confusable with a /t/, may change over these two time scales. The target versus control groups both heard Mandarin-accented speech, but only the target group heard words containing the critical contrast. Using a new repeated exposure-test paradigm, trajectories of adaptation were assessed three times within an initial session (Exp.1) as well as over five sessions spanning a month (Exp.2). In addition to mixed-effect analysis on the effect of exposure at a group level (i.e., target versus control), computational simulations of distinct mechanisms (cue-based normalization, Bayesian belief updating, changes of decision-making criteria) will examine whether adaptation seen across the timescales can stem from the same mechanism.
Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—some... more Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—sometimes within minutes of exposure. Two major questions remain open: (a) How does adaptation commence within the initial moments of an encounter?; and (b) How does the adaptation continue via repeated exposure to the same talker/accent? The current study addressed these questions by assessing incremental changes of native English listeners’ recognition of L2-accented speech (Mandarin-accented English) over two timescales: within the first few minutes and over a month. Specifically, we tested how native listeners’ recognition of a word-final /d/, which is initially confusable with a /t/, may change over these two time scales. The target versus control groups both heard Mandarin-accented speech, but only the target group heard words containing the critical contrast. Using a new repeated exposure-test paradigm, trajectories of adaptation were assessed three times within an initial session (Ex...
journal homepage: www.elsevier.com/locate/COGNIT Zipfian frequency distributions facilitate word ... more journal homepage: www.elsevier.com/locate/COGNIT Zipfian frequency distributions facilitate word segmentation
The acquisition and development of the topic marker wa in L1 Japanese The role of NP-wa? 1 in mot... more The acquisition and development of the topic marker wa in L1 Japanese The role of NP-wa? 1 in mother-child interaction* Chigusa Kurumada Department of Linguistics, Stanford University 1. Introduction 348 2. Method 351 2.1 Data 351 2.2 Coding 352 2.3 Data analysis 353 3. ...
The object of this presentation is to introduce a prototype of the Digital Museum Project in our ... more The object of this presentation is to introduce a prototype of the Digital Museum Project in our attempt at the documentation of Ikema, one of the endangered dialects of Southern Ryukyuan, spoken on Miyakojima Island, Okinawa, Japan. The language is no longer ...
This project adopts computational approaches to explore mechanisms underlying exposure-driven cha... more This project adopts computational approaches to explore mechanisms underlying exposure-driven changes of speech perception.
We present an eye-tracking experiment investigating the time course with which listeners derive p... more We present an eye-tracking experiment investigating the time course with which listeners derive pragmatic inferences from contextual information. We used as a test case the construction "It looks like an X" pronounced either with (a) a nuclear pitch accent on the final noun, or (b) a contrastive L+H* pitch accent and a rising boundary tone, a contour that can support a complex contrastive inference (e.g., It LOOKS like a zebra...(but it is not)). The contrastive intonational contour elicited higher proportions of fixations to non-prototypical target pictures (e.g., a zebra-like animal) during the earliest moments of processing the target noun. Further, when the display only contained a single related pair of pictures, effects of the contrastive accent on "looks" emerged prior to the target noun, indicating that efficient referential resolution is supported by rapidly generated inferences based on visual and prosodic context .
We investigate pre-schoolers' ability in drawing pragmatic inferences based on prosodic infor... more We investigate pre-schoolers' ability in drawing pragmatic inferences based on prosodic information. Previous work has found that young children are generally oblivious to intona-tional meaning of utterances. In particular, the ability to comprehend contrastive prosody develops late during language acquisition (after the age of 6). In three experiments, we show that preschoolers can engage in prosody-based pragmatic inferences if the context provides supports for them. Furthermore , we find that preschoolers' interpretation of prosody involves complex counter-factual reasoning ('what the speaker would have said if she had intended another meaning'). The picture emerging from our studies contrasts with previous work: Through rich contextual inferences, four-year olds are able to bootstrap their interpretation of prosodic information, and achieve adult like performance in intonation interpretation.
Journal of the Acoustical Society of America, Mar 1, 2023
Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—some... more Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—sometimes within minutes of exposure. Two major questions remain open: (a) How does adaptation commence within the initial moments of an encounter?; and (b) How does the adaptation continue via repeated exposure to the same talker/accent? The current study addressed these questions by assessing incremental changes of native English listeners’ recognition of L2-accented speech (Mandarin-accented English) over two timescales: within the first few minutes and over a month. Specifically, we tested how native listeners’ recognition of a word-final /d/, which is initially confusable with a /t/, may change over these two time scales. The target versus control groups both heard Mandarin-accented speech, but only the target group heard words containing the critical contrast. Using a new repeated exposure-test paradigm, trajectories of adaptation were assessed three times within an initial session (Exp.1) as well as over five sessions spanning a month (Exp.2). In addition to mixed-effect analysis on the effect of exposure at a group level (i.e., target versus control), computational simulations of distinct mechanisms (cue-based normalization, Bayesian belief updating, changes of decision-making criteria) will examine whether adaptation seen across the timescales can stem from the same mechanism.
Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—some... more Perceptual difficulties associated with unfamiliar talkers or accents are known to dissipate—sometimes within minutes of exposure. Two major questions remain open: (a) How does adaptation commence within the initial moments of an encounter?; and (b) How does the adaptation continue via repeated exposure to the same talker/accent? The current study addressed these questions by assessing incremental changes of native English listeners’ recognition of L2-accented speech (Mandarin-accented English) over two timescales: within the first few minutes and over a month. Specifically, we tested how native listeners’ recognition of a word-final /d/, which is initially confusable with a /t/, may change over these two time scales. The target versus control groups both heard Mandarin-accented speech, but only the target group heard words containing the critical contrast. Using a new repeated exposure-test paradigm, trajectories of adaptation were assessed three times within an initial session (Ex...
journal homepage: www.elsevier.com/locate/COGNIT Zipfian frequency distributions facilitate word ... more journal homepage: www.elsevier.com/locate/COGNIT Zipfian frequency distributions facilitate word segmentation
The acquisition and development of the topic marker wa in L1 Japanese The role of NP-wa? 1 in mot... more The acquisition and development of the topic marker wa in L1 Japanese The role of NP-wa? 1 in mother-child interaction* Chigusa Kurumada Department of Linguistics, Stanford University 1. Introduction 348 2. Method 351 2.1 Data 351 2.2 Coding 352 2.3 Data analysis 353 3. ...
The object of this presentation is to introduce a prototype of the Digital Museum Project in our ... more The object of this presentation is to introduce a prototype of the Digital Museum Project in our attempt at the documentation of Ikema, one of the endangered dialects of Southern Ryukyuan, spoken on Miyakojima Island, Okinawa, Japan. The language is no longer ...
This project adopts computational approaches to explore mechanisms underlying exposure-driven cha... more This project adopts computational approaches to explore mechanisms underlying exposure-driven changes of speech perception.
We present an eye-tracking experiment investigating the time course with which listeners derive p... more We present an eye-tracking experiment investigating the time course with which listeners derive pragmatic inferences from contextual information. We used as a test case the construction "It looks like an X" pronounced either with (a) a nuclear pitch accent on the final noun, or (b) a contrastive L+H* pitch accent and a rising boundary tone, a contour that can support a complex contrastive inference (e.g., It LOOKS like a zebra...(but it is not)). The contrastive intonational contour elicited higher proportions of fixations to non-prototypical target pictures (e.g., a zebra-like animal) during the earliest moments of processing the target noun. Further, when the display only contained a single related pair of pictures, effects of the contrastive accent on "looks" emerged prior to the target noun, indicating that efficient referential resolution is supported by rapidly generated inferences based on visual and prosodic context .
We investigate pre-schoolers' ability in drawing pragmatic inferences based on prosodic infor... more We investigate pre-schoolers' ability in drawing pragmatic inferences based on prosodic information. Previous work has found that young children are generally oblivious to intona-tional meaning of utterances. In particular, the ability to comprehend contrastive prosody develops late during language acquisition (after the age of 6). In three experiments, we show that preschoolers can engage in prosody-based pragmatic inferences if the context provides supports for them. Furthermore , we find that preschoolers' interpretation of prosody involves complex counter-factual reasoning ('what the speaker would have said if she had intended another meaning'). The picture emerging from our studies contrasts with previous work: Through rich contextual inferences, four-year olds are able to bootstrap their interpretation of prosodic information, and achieve adult like performance in intonation interpretation.
Uploads
Papers by Chigusa Kurumada