Amy Schafer | University of Hawaii at Manoa (original) (raw)

Papers by Amy Schafer

Language and Speech, 2017

The Journal of the Acoustical Society of America, 2016

Previous work [1]–[5] has provided evidence that the perception of lexical tones by native speake... more Previous work [1]–[5] has provided evidence that the perception of lexical tones by native speakers of Mandarin can be more categorical than that of naïve foreign listeners, for the tone pairings T1-T2, T2-T4, T1-T4 and T3-T4. The present study extended this work by testing Mandarin and naïve English listeners' perception of all six possible pairwise combinations of the four Mandarin tones, using identification and AXB discrimination tasks. The results revealed significant differences in perception across tone pairs, highlighting potential challenges for language learning. They also confirmed that Mandarin listeners were more categorical in their perception than the naïve listeners, who showed higher discrimination accuracy than Mandarin listeners. The latter finding contrasts with previous results for naïve French speakers' perception of Mandarin tone [2], and is consistent with an effect of native-language supra-segmental patterns on foreign-language tone perception.

Discourse-level factors, such as event structure and the form of referential expressions, play an... more Discourse-level factors, such as event structure and the form of referential expressions, play an important role in native speakers' referential processing. This paper presents an experiment with Japanese-and Korean-speaking learners of English, investigating the extent to which discourse-level biases that have gradient effects in L1 speakers are also implicated in L2 speakers' coreference choices. Results from a story continuation task indicate that biases involving referential form were remarkably similar for L1 and L2 speakers. In contrast, event structure, indicated by perfective versus imperfective aspect, had a more limited effect on L2 speakers' referential choices. The L2 results are discussed in light of existing accounts of L1 reference processing, which assume that referential choices are shaped by speakers' continually updated expectations about what is likely to be mentioned next, and argued to reflect L2 speakers' reduced reliance on expectations.

Listeners rapidly process tonal composition and pitch accent placement within an utterance to cre... more Listeners rapidly process tonal composition and pitch accent placement within an utterance to create expectations about its pragmatic meaning and information structure. It is still unknown whether the nuclear pitch accent alone or a combination of pitch accent and the following edge tone are needed in order to process intonational meaning in French. This study investigates the online comprehension of the French (L)H*L% rise-fall “implication” contour, which evokes a contrast meaning. Twenty-nine speakers participated in an eye-tracking experiment. The critical stimuli were sentences whose interpretation could be anticipated by successfully processing the implied meaning evoked by the (L)H*L% rise-fall contour on the critical word (hereafter CW). The results showed that participants are able to associate the implication contour with a contrast meaning, and that they start doing this only after the H* peak of the rise-fall intonation movement has been processed, hence when part of the L% falling movement has been perceived.

The Journal of the Acoustical Society of America, 2008

ABSTRACT In the work of Hwang et al. (2007), native English speakers showed overall poor accuracy... more ABSTRACT In the work of Hwang et al. (2007), native English speakers showed overall poor accuracy in distinguishing initially rising versus level (e.g., L(*)L(*)H- H(*)L-L% vs L(*)L(*)L- H(*)L-L%) or initially falling versus level (e.g., H(*)H(*)L- H(*)L-L% vs H(*)H(*)H- H(*)L-L%) contour contrasts on English phrases in an AX discrimination task. Results not reported in that paper found that it was easier to discriminate when a more complex F0 contour occurred second than when it occurred first. Several orders of presentation effects in the perception of intonation have been reported (e.g., L. Morton (1997); S. Lintfert (2003); Cummins et al. (2006)] but no satisfying account has been provided. This study investigated these asymmetries more systematically. The order effect was significant for falling-level contrast pairs: pairs with a more complex F0 contour last were discriminated more easily than the reverse order. Rising versus level contrasts showed a similar tendency. The results thus extend intonational discrimination asymmetries to these additional contours. They suggest that the cause of the asymmetries may depend more on F0 complexity than on F0 peak.

Journal of Psycholinguistic Research, 2009

Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phr... more Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phrase length on overt and implicit prosody. Both experiments controlled non-prosodic length factors by using long versus short proper names that occurred before the syntactically critical material. Experiment 1 found that long phrases induce different prosodic phrasing than short phrases in a read-aloud task and change the preferred interpretation of globally ambiguous sentences. It also showed that speakers who have been told of the ambiguity can provide significantly different prosody for the two interpretations, for both lengths. Experiment 2 verified that prosodic patterns found in first-pass pronunciations predict self-paced reading patterns for silent reading. The results extend the coverage of the Implicit Prosody Hypothesis [Fodor, J Psycholinguist Res 27:285-319, 1998; Prosodic disambiguation in silent reading. In M. Hirotani (Ed.), NELS 32 (pp. 113-132). Amherst, MA: GLSA Publications, 2002] to another construction and to Korean. They further indicate that strong syntactic biases can have rapid effects on the formulation of implicit prosody.

Coreference choices are influenced by multiple factors, including information structural categori... more Coreference choices are influenced by multiple factors, including information structural categories such as topic and focus. These information structural categories can be indicated by intonation, yet few studies have investigated how intonation affects subsequent choices for coreference. Using a story continuation experiment with aurally presented stimuli, we show that the location of contrastive focus in Mainstream American English significantly affects the preferred referent for the subject of the next sentence in a short discourse.

We report on two experiments that ask when and under what linguistic conditions comprehenders con... more We report on two experiments that ask when and under what linguistic conditions comprehenders construct detailed shape representations of mentioned objects, and whether these can change over the course of a sentence when new information contradicts earlier expectations. We used Japanese because the verb-final word order of Japanese presented a revealing test case where information about objects can radically change with a subsequent verb. The results show that language understanders consistently generate a distinct and detailed shape for an object by integrating the semantic contributions of different sentential elements. These results first confirm that the tendency to generate specific shape information about objects that are involved in described events is not limited to English, but is also present in Japanese, a typologically and genetically distinct language. But more importantly, they shed light on the processing mechanism of object representation, showing that mental representations are initiated sentence medially, and are rapidly revised if followed by a verb that implies a change to an object shape. This work contributes to ongoing research on incremental language processingcomprehenders appear to construct extremely detailed semantic representations early in a sentence, and modify them as needed.

A series of speech production and categorization experiments demonstrates that naïve speakers and... more A series of speech production and categorization experiments demonstrates that naïve speakers and listeners reliably use correspondences between prosodic phrasing and syntactic constituent structure to resolve standing and temporary ambiguity. Materials obtained from a co-operative gameboard task show that prosodic phrasing effects (e.g., the location of the strongest break in an utterance) are independent of discourse factors that might be expected to influence the impact of syntactic ambiguity, including the availability of visual referents for the meanings of ambiguous utterances and the use of utterances as instructions versus confirmations of instructions. These effects hold across two dialects of English, spoken in the American Midwest, and New Zealand. Results from PP-attachment and verb transitivity ambiguities indicate clearly that the production of prosody-syntax correspondences is not conditional upon situational disambiguation of syntactic structure, but is rather more directly tied to grammatical constraints on the production of prosodic and syntactic form. Differences between our results and those reported elsewhere are best explained in terms of differences in task demands.

Little experimental evidence exists for how prosodic/ intonational information might affect the g... more Little experimental evidence exists for how prosodic/ intonational information might affect the generation of an implicature. We provide online evidence that the combination of an L+H* pitch accent and an L−H% boundary tone work together to imply a contradiction, and that this contour has distinct effects from an L+H* L−L% tune. We also compare the online processing of changes in meaning suggested by prosody versus explicit negation. The results highlight the importance of intonational information in sentence understanding, and the differences in processing prosodically cued contrastive information versus lexical negation.

Journal of Psycholinguistic Research, 2009

Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phr... more Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phrase length on overt and implicit prosody. Both experiments controlled non-prosodic length factors by using long versus short proper names that occurred before the syntactically critical material. Experiment 1 found that long phrases induce different prosodic phrasing than short phrases in a read-aloud task and change the preferred interpretation of globally ambiguous sentences. It also showed that speakers who have been told of the ambiguity can provide significantly different prosody for the two interpretations, for both lengths. Experiment 2 verified that prosodic patterns found in first-pass pronunciations predict self-paced reading patterns for silent reading. The results extend the coverage of the Implicit Prosody Hypothesis [Fodor, J Psycholinguist Res 27:285–319, 1998; Prosodic disambiguation in silent reading. In M. Hirotani (Ed.), NELS 32 (pp. 113–132). Amherst, MA: GLSA Publications, 2002] to another construction and to Korean. They further indicate that strong syntactic biases can have rapid effects on the formulation of implicit prosody.

A major obstacle to the early diagnosis of language loss and to the assessment of language mainte... more A major obstacle to the early diagnosis of language loss and to the assessment of language maintenance efforts is the absence of an easy-to-use psycholinguistic measure of language strength. In this paper, we describe and discuss a body-part naming task being developed as part of the Hawai'i Assessment of Language Access (HALA) project. This task, like the others in the HALA inventory, exploits the fact that the speed with which bilingual speakers access lexical items and structure-building operations in their two languages offers a sensitive measure of relative language strength. In a pilot study conducted with Korean-English bilinguals, we were able to establish a strong correlation between language strength and naming times even in highly fluent bilingual speakers, in support of the central assumption underlying the HALA tests. We discuss the implications of this finding for the broader study of language strength as well as for the practical problems associated with work on language loss, maintenance, and revitalization.

Previous work has shown that advanced Korean learners of English (L2ers) are less effective than ... more Previous work has shown that advanced Korean learners of English (L2ers) are less effective than native English speakers (L1ers) at using English intermediate phrases (ips) to establish syntactic boundaries . This study investigated whether the effect is due to perceptual differences between L1ers and L2ers, based on the interplay between phonology and perception (e.g., ). L1ers and L2ers listened to pairs of phrases in an AX task that crossed boundary strength with intonational contour. Little variation was found between L1ers' and L2ers' discrimination patterns, which correlated highly with each other. Both groups were more sensitive to falling vs. level contour contrasts than rising vs. level contrasts (in the context tested) and were more responsive to contrasts in contour than in boundary strength. The results suggest that the L2ers' poor use of ips in comprehension likely rests primarily on difficulty with prosody-syntax mappings.

The Informative Boundary Hypothesis (IBH: ) claims that a prosodic boundary is interpreted relati... more The Informative Boundary Hypothesis (IBH: ) claims that a prosodic boundary is interpreted relative to preceding boundaries. This study tests predictions of the IBH with Korean learners of English (L2ers) and English native speakers (L1ers) in a prosody experiment on the resolution of an Early vs. Late Closure ambiguity in spoken English sentences. A control experiment assessed and controlled for English morpho-syntactic knowledge in the main experiment. The main experiment presented the syntactically ambiguous portion of sentences in a forced-choice continuation-selection task. The results showed that 1) Korean L2ers at all levels used relative boundary size to disambiguate sentences, like L1ers; 2) intonation phrase boundaries provided stronger evidence for syntactic boundaries than intermediate phrase boundaries, especially for the L2ers; and 3) the IBH's 3-way categorization of relative boundary size -larger/same-size/smaller -appears insufficient for this syntactic structure.

Language and Speech, 2017

The Journal of the Acoustical Society of America, 2016

Previous work [1]–[5] has provided evidence that the perception of lexical tones by native speake... more Previous work [1]–[5] has provided evidence that the perception of lexical tones by native speakers of Mandarin can be more categorical than that of naïve foreign listeners, for the tone pairings T1-T2, T2-T4, T1-T4 and T3-T4. The present study extended this work by testing Mandarin and naïve English listeners' perception of all six possible pairwise combinations of the four Mandarin tones, using identification and AXB discrimination tasks. The results revealed significant differences in perception across tone pairs, highlighting potential challenges for language learning. They also confirmed that Mandarin listeners were more categorical in their perception than the naïve listeners, who showed higher discrimination accuracy than Mandarin listeners. The latter finding contrasts with previous results for naïve French speakers' perception of Mandarin tone [2], and is consistent with an effect of native-language supra-segmental patterns on foreign-language tone perception.

Discourse-level factors, such as event structure and the form of referential expressions, play an... more Discourse-level factors, such as event structure and the form of referential expressions, play an important role in native speakers' referential processing. This paper presents an experiment with Japanese-and Korean-speaking learners of English, investigating the extent to which discourse-level biases that have gradient effects in L1 speakers are also implicated in L2 speakers' coreference choices. Results from a story continuation task indicate that biases involving referential form were remarkably similar for L1 and L2 speakers. In contrast, event structure, indicated by perfective versus imperfective aspect, had a more limited effect on L2 speakers' referential choices. The L2 results are discussed in light of existing accounts of L1 reference processing, which assume that referential choices are shaped by speakers' continually updated expectations about what is likely to be mentioned next, and argued to reflect L2 speakers' reduced reliance on expectations.

Listeners rapidly process tonal composition and pitch accent placement within an utterance to cre... more Listeners rapidly process tonal composition and pitch accent placement within an utterance to create expectations about its pragmatic meaning and information structure. It is still unknown whether the nuclear pitch accent alone or a combination of pitch accent and the following edge tone are needed in order to process intonational meaning in French. This study investigates the online comprehension of the French (L)H*L% rise-fall “implication” contour, which evokes a contrast meaning. Twenty-nine speakers participated in an eye-tracking experiment. The critical stimuli were sentences whose interpretation could be anticipated by successfully processing the implied meaning evoked by the (L)H*L% rise-fall contour on the critical word (hereafter CW). The results showed that participants are able to associate the implication contour with a contrast meaning, and that they start doing this only after the H* peak of the rise-fall intonation movement has been processed, hence when part of the L% falling movement has been perceived.

The Journal of the Acoustical Society of America, 2008

ABSTRACT In the work of Hwang et al. (2007), native English speakers showed overall poor accuracy... more ABSTRACT In the work of Hwang et al. (2007), native English speakers showed overall poor accuracy in distinguishing initially rising versus level (e.g., L(*)L(*)H- H(*)L-L% vs L(*)L(*)L- H(*)L-L%) or initially falling versus level (e.g., H(*)H(*)L- H(*)L-L% vs H(*)H(*)H- H(*)L-L%) contour contrasts on English phrases in an AX discrimination task. Results not reported in that paper found that it was easier to discriminate when a more complex F0 contour occurred second than when it occurred first. Several orders of presentation effects in the perception of intonation have been reported (e.g., L. Morton (1997); S. Lintfert (2003); Cummins et al. (2006)] but no satisfying account has been provided. This study investigated these asymmetries more systematically. The order effect was significant for falling-level contrast pairs: pairs with a more complex F0 contour last were discriminated more easily than the reverse order. Rising versus level contrasts showed a similar tendency. The results thus extend intonational discrimination asymmetries to these additional contours. They suggest that the cause of the asymmetries may depend more on F0 complexity than on F0 peak.

Journal of Psycholinguistic Research, 2009

Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phr... more Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phrase length on overt and implicit prosody. Both experiments controlled non-prosodic length factors by using long versus short proper names that occurred before the syntactically critical material. Experiment 1 found that long phrases induce different prosodic phrasing than short phrases in a read-aloud task and change the preferred interpretation of globally ambiguous sentences. It also showed that speakers who have been told of the ambiguity can provide significantly different prosody for the two interpretations, for both lengths. Experiment 2 verified that prosodic patterns found in first-pass pronunciations predict self-paced reading patterns for silent reading. The results extend the coverage of the Implicit Prosody Hypothesis [Fodor, J Psycholinguist Res 27:285-319, 1998; Prosodic disambiguation in silent reading. In M. Hirotani (Ed.), NELS 32 (pp. 113-132). Amherst, MA: GLSA Publications, 2002] to another construction and to Korean. They further indicate that strong syntactic biases can have rapid effects on the formulation of implicit prosody.

Coreference choices are influenced by multiple factors, including information structural categori... more Coreference choices are influenced by multiple factors, including information structural categories such as topic and focus. These information structural categories can be indicated by intonation, yet few studies have investigated how intonation affects subsequent choices for coreference. Using a story continuation experiment with aurally presented stimuli, we show that the location of contrastive focus in Mainstream American English significantly affects the preferred referent for the subject of the next sentence in a short discourse.

We report on two experiments that ask when and under what linguistic conditions comprehenders con... more We report on two experiments that ask when and under what linguistic conditions comprehenders construct detailed shape representations of mentioned objects, and whether these can change over the course of a sentence when new information contradicts earlier expectations. We used Japanese because the verb-final word order of Japanese presented a revealing test case where information about objects can radically change with a subsequent verb. The results show that language understanders consistently generate a distinct and detailed shape for an object by integrating the semantic contributions of different sentential elements. These results first confirm that the tendency to generate specific shape information about objects that are involved in described events is not limited to English, but is also present in Japanese, a typologically and genetically distinct language. But more importantly, they shed light on the processing mechanism of object representation, showing that mental representations are initiated sentence medially, and are rapidly revised if followed by a verb that implies a change to an object shape. This work contributes to ongoing research on incremental language processingcomprehenders appear to construct extremely detailed semantic representations early in a sentence, and modify them as needed.

A series of speech production and categorization experiments demonstrates that naïve speakers and... more A series of speech production and categorization experiments demonstrates that naïve speakers and listeners reliably use correspondences between prosodic phrasing and syntactic constituent structure to resolve standing and temporary ambiguity. Materials obtained from a co-operative gameboard task show that prosodic phrasing effects (e.g., the location of the strongest break in an utterance) are independent of discourse factors that might be expected to influence the impact of syntactic ambiguity, including the availability of visual referents for the meanings of ambiguous utterances and the use of utterances as instructions versus confirmations of instructions. These effects hold across two dialects of English, spoken in the American Midwest, and New Zealand. Results from PP-attachment and verb transitivity ambiguities indicate clearly that the production of prosody-syntax correspondences is not conditional upon situational disambiguation of syntactic structure, but is rather more directly tied to grammatical constraints on the production of prosodic and syntactic form. Differences between our results and those reported elsewhere are best explained in terms of differences in task demands.

Little experimental evidence exists for how prosodic/ intonational information might affect the g... more Little experimental evidence exists for how prosodic/ intonational information might affect the generation of an implicature. We provide online evidence that the combination of an L+H* pitch accent and an L−H% boundary tone work together to imply a contradiction, and that this contour has distinct effects from an L+H* L−L% tune. We also compare the online processing of changes in meaning suggested by prosody versus explicit negation. The results highlight the importance of intonational information in sentence understanding, and the differences in processing prosodically cued contrastive information versus lexical negation.

Journal of Psycholinguistic Research, 2009

Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phr... more Two sentence processing experiments on a dative NP ambiguity in Korean demonstrate effects of phrase length on overt and implicit prosody. Both experiments controlled non-prosodic length factors by using long versus short proper names that occurred before the syntactically critical material. Experiment 1 found that long phrases induce different prosodic phrasing than short phrases in a read-aloud task and change the preferred interpretation of globally ambiguous sentences. It also showed that speakers who have been told of the ambiguity can provide significantly different prosody for the two interpretations, for both lengths. Experiment 2 verified that prosodic patterns found in first-pass pronunciations predict self-paced reading patterns for silent reading. The results extend the coverage of the Implicit Prosody Hypothesis [Fodor, J Psycholinguist Res 27:285–319, 1998; Prosodic disambiguation in silent reading. In M. Hirotani (Ed.), NELS 32 (pp. 113–132). Amherst, MA: GLSA Publications, 2002] to another construction and to Korean. They further indicate that strong syntactic biases can have rapid effects on the formulation of implicit prosody.

A major obstacle to the early diagnosis of language loss and to the assessment of language mainte... more A major obstacle to the early diagnosis of language loss and to the assessment of language maintenance efforts is the absence of an easy-to-use psycholinguistic measure of language strength. In this paper, we describe and discuss a body-part naming task being developed as part of the Hawai'i Assessment of Language Access (HALA) project. This task, like the others in the HALA inventory, exploits the fact that the speed with which bilingual speakers access lexical items and structure-building operations in their two languages offers a sensitive measure of relative language strength. In a pilot study conducted with Korean-English bilinguals, we were able to establish a strong correlation between language strength and naming times even in highly fluent bilingual speakers, in support of the central assumption underlying the HALA tests. We discuss the implications of this finding for the broader study of language strength as well as for the practical problems associated with work on language loss, maintenance, and revitalization.

Previous work has shown that advanced Korean learners of English (L2ers) are less effective than ... more Previous work has shown that advanced Korean learners of English (L2ers) are less effective than native English speakers (L1ers) at using English intermediate phrases (ips) to establish syntactic boundaries . This study investigated whether the effect is due to perceptual differences between L1ers and L2ers, based on the interplay between phonology and perception (e.g., ). L1ers and L2ers listened to pairs of phrases in an AX task that crossed boundary strength with intonational contour. Little variation was found between L1ers' and L2ers' discrimination patterns, which correlated highly with each other. Both groups were more sensitive to falling vs. level contour contrasts than rising vs. level contrasts (in the context tested) and were more responsive to contrasts in contour than in boundary strength. The results suggest that the L2ers' poor use of ips in comprehension likely rests primarily on difficulty with prosody-syntax mappings.

The Informative Boundary Hypothesis (IBH: ) claims that a prosodic boundary is interpreted relati... more The Informative Boundary Hypothesis (IBH: ) claims that a prosodic boundary is interpreted relative to preceding boundaries. This study tests predictions of the IBH with Korean learners of English (L2ers) and English native speakers (L1ers) in a prosody experiment on the resolution of an Early vs. Late Closure ambiguity in spoken English sentences. A control experiment assessed and controlled for English morpho-syntactic knowledge in the main experiment. The main experiment presented the syntactically ambiguous portion of sentences in a forced-choice continuation-selection task. The results showed that 1) Korean L2ers at all levels used relative boundary size to disambiguate sentences, like L1ers; 2) intonation phrase boundaries provided stronger evidence for syntactic boundaries than intermediate phrase boundaries, especially for the L2ers; and 3) the IBH's 3-way categorization of relative boundary size -larger/same-size/smaller -appears insufficient for this syntactic structure.