Edward Carney

Neural coding of formant-exaggerated speech in the infant brain

Speech scientists have long proposed that formant exaggeration in infant-directed speech plays an important role in language acquisition. This event-related potential (ERP) study investigated neural coding of formant-exaggerated speech in 6-12-month-old infants. Two synthetic /i/ vowels were presented in alternating blocks to test the effects of formant exaggeration. ERP waveform analysis showed significantly enhanced N250 for formant exaggeration, which was more prominent in the right hemisphere than the left. Time-frequency analysis indicated increased neural synchronization for processing formant-exaggerated speech in the delta band at frontal-central-parietal electrode sites as well as in the theta band at frontal-central sites. Minimum norm estimates further revealed a bilateral temporal-parietal-frontal neural network in the infant brain sensitive to formant exaggeration. Collectively, these results provide the first evidence that formant expansion in infant-directed speech enhances neural activities for phonetic encoding and language learning.

Statistical properties of responses to dichotic listening with CV nonsense syllables

Twenty-four listeners received 20 dichotic listening runs of 30 pairs of natural CV nonsense syllables per run. Left- and eight-ear responses are represented by two partially overlapping normal distributions with equal variance. Observed ear advantages across listening runs are also distributed normally. The means of the distribution of ear advantage are listener specific, but the standard deviations are 10.8% +/- 2.5. The origin of reversals in direction of ear advantage among listening runs is statistical--a small mean/sigma ratio--and approximately six listening runs (180 listening trials) are required to achieve a split-half reliability coefficient of + 0.90.

Predictions of recall and study strategy decisions after diffuse brain injury

Objective: To investigate relationships between self-monitoring operationalized by predicting recall and study strategy decisions made by adults with diffuse, acquired brain injury (ABI) and adults without ABI. Research design and methods: Eighteen adults with ABI and 16 without ABI studied two lists of unrelated noun-pairs, made item-by-item immediate and delayed recall predictions and selected items (after predictions) to restudy again. The computer selected items for restudy based on the lowest prediction rating (i.e. unlikely to recall). A mixed list design was used to balance item selection (self or computer, within-lists) by type of prediction (immediate or delayed, between-lists). Recall was tested prior to and after restudying. Hypotheses: Delayed recall predictions would be more accurate than immediate recall predictions; participants would select items for restudy that corresponded with ‘lower’ delayed predictions (i.e. less likely to recall) and ‘higher’ immediate predictions (i.e. more likely to recall); for adults with ABI, recall would improve the most from restudying items selected after delayed predictions; and that recall predictive accuracy and study selection decisions are independent processes. Results: Delayed recall predictions were more accurate than immediate recall predictions, though adults with ABI tended to be less accurate than controls. Both groups selected items for restudy that had relatively low prediction ratings irrespective of prediction timing. Of adults with ABI, those with low recall prior to restudy selected items that had ‘high’ immediate predictions (i.e. likely to recall). However, there was no greater benefit to recall using this strategy. For adults with ABI, recall improved the most from restudying items that were self-selected after delayed predictions, whereas controls’ recall improved, irrespective of prediction and selection timing. Between-person correlations revealed no relationship between recall predictive accuracy and study selection strategies. Conclusions: These findings imply that adults with ABI should base strategy decisions on delayed predictions rather than on ‘in the moment’ immediate ones, selecting items that they have predicted will be difficult to recall. Continued efforts to disambiguate self-monitoring from strategy decisions are required before direct clinical applications can be made.

Multi-unit mapping of acoustic stimuli in gerbil inferior colliculus

Multi-unit peristimulus time (MU-PST) histograms were recorded in the gerbil inferior colliculus (IC) in response to tone burst stimuli. Histograms were collected every 100 microns as the recording electrode was advanced along the tonotopic axis of the central nucleus of the IC. Space/time maps of neural activity were constructed from these data. In most of our sample the pattern of response changed systematically as the stimulating frequency was increased in octave steps. At low frequencies (< 500 Hz) the pattern of response was broadly distributed spatially and phase-locked to the stimulus frequency. At higher frequencies (> 1 kHz) the pattern of response was more localized and showed no evidence of phase locking. The location of the maximum response to tones from 1 to 32 kHz moved ventrally along the tonotopic axis at an approximate rate of 230 microns/stimulus octave. The patterns of response were localized near stimulus threshold and spread over a larger region as level increased. This method of collecting and displaying multi-unit response maps provides an overview of ensemble activity that allows concurrent observation of spatial and temporal variations in activity patterns. The quantitative analysis of components of MU-PST Maps are consistent with trends illustrated with single-unit tuning and level functions. This perspective of IC activity suggests potential processing mechanisms that are congruent with single-unit reconstructions.

Corpora of Vietnamese Texts: Lexical effects of intended audience and publication place

This article has two primary aims. The first is to introduce a new Vietnamese text-based corpus. The Corpora of Vietnamese Texts (CVT; Tang, 2006a) consists of approximately 1 million words drawn from newspapers and children’s literature, and is available online at www.vnspeechtherapy.com/vi/CVT. The second aim is to investigate potential differences in lexical frequency and distributional characteristics in the CVT on the basis of place of publication (Vietnam or Western countries) and intended audience: adult-directed texts (newspapers) or child-directed texts (children’s literature). We found clear differences between adult- and child-directed texts, particularly in the distributional frequencies of pronouns or kinship terms, which were more frequent in children’s literature. Within child- and adult-directed texts, lexical characteristics did not differ on the basis of place of publication. Implications of these findings for future research are discussed.

Waveform editing and spectrographic software.

Programs for interactive editing and spectrographic display of digitized speech waveform files have been developed for use on IBM‐PC compatible computers. The waveform editing program (WavEd) reads and writes waveform files of several types, including ILS sampled data. WavEd allows the user to perform the following operations: (1) record and play waveforms; (2) place labels at various points, and save these labels; (3) copy, delete, and insert waveform segments; and (4) do acoustic analyses, such as computation of durations, formant frequencies and bandwidths, and linear prediction coefficients. The spectrographic display program (SPECTO) produces spectrograms from ILS sampled data files using filters of user‐specified bandwidths. Parameters that determine spectrogram appearance (such as darkness, contrast, and frequency and time range) are user controlled. Spectrograms can be sent to the PC monitor or to dot‐matrix or laser printers. Both programs are designed to be easy to install and to use on any PC, ...

Cortical processing of audiovisual speech perception in infancy and adulthood

The ability to detect auditory-visual correspondence in speech is an early hallmark of typical language development. Infants are able to detect audiovisual mismatches for spoken vowels such as /a/ and /i/ as early as 4 months of age. While adult event-related potential (ERP) data have shown an N300 associated with the detection of audiovisual incongruency in speech, it remains unclear whether similar responses can be elicited in infants. The present study collected ERP data in congruent and incongruent audiovisual presentation conditions for /a/ and /i/ from 21 typically developing infants (6~11 month of age) and 12 normal adults (18~45 years). The adult data replicated the N300 in the parietal electrode sites for detecting audiovisual incongruency in speech, and minimum norm estimation (MNE) showed the primary neural generator in the left superior temporal cortex for the N300. Unlike the adults, the infants showed a later N400 response in the centro-frontal electrode sites, and scalp topography as well a...

Neural coding of formant‐exaggerated speech in infants and adults.

Speech scientists have long proposed that formant‐exaggerated speech plays an important role in phonetic learning and language acquisition. However, there have been very little neurophysiological data on how the infant brain and adult brain respond to formant exaggeration in speech. We employed event‐related potentials (ERPs) to investigate neural coding of formant‐exaggerated speech sounds. Two synthetic /i/ vowels were modeled after infant‐directed speech data and presented in alternating blocks to test the effects of formant exaggeration. The fundamental frequencies of the two sounds were kept identical to avoid interference from exaggerated pitch level and range. For adult subjects, non‐speech homologs were also created by using the center frequencies of the formants to additionally test whether the effects were speech‐specific. In the infants (6 to 12‐month olds), ERP waveforms showed significantly enhanced N250 and sustaining negativity responses for processing formant‐exaggerated speech. In adults,...

A comparison of Levitt and Zwislocki decision rules for use with forced‐choice adaptive procedures

Forced‐choice adaptive procedures enjoy widespread use for the measurement of detection and discrimination thresholds. Zwislocki et al. [J. Acoust. Soc. Am. 30, 254–262 (1958)] proposed an adaptive procedure with a decision rule that targets 75% correct several years before these procedures gained acceptance in psychophysics, and even today little is known about the statistical properties of this decision rule. This paper evaluates, using computer simulations, the bias and efficiency of Zwislocki’s proposed decision rule in 2‐alternative forced‐choice (AFC) and 3‐AFC procedures under conditions of full attention and inattenion. The results for this decision rule compare favorably with two popular rules proposed by Levitt [J. Acoust. Soc. Am. 49, 467–477 (1971)] that target 70.7% correct and 79.4% correct. In summary, the rule that targets 75% correct (Zwislocki’s rule) produces less biased threshold estimates than the rule that targets 70.7% correct and it is affected less by inattention than the rule tha...

Pure-Tone–Spondee Threshold Relationships in Functional Hearing Loss: A Test of Loudness Contribution

Purpose The purpose of this article is to examine explanations for pure-tone average-spondee threshold differences in functional hearing loss. Method Loudness magnitude estimation functions were obtained from 24 participants for pure tones (0.5 and 1.0 kHz), vowels, spondees, and speech-shaped noise as a function of level (20-90 dB SPL). Participants listened monaurally through earphones. Loudness predictions were obtained for the same stimuli by using a computational, dynamic loudness model. Results When evaluated at the same SPL, speech-shaped noise was judged louder than vowels/spondees, which were judged louder than tones. Equal-loudness levels were inferred from fitted loudness functions for the group. For the clinical application, the 2.1-dB difference between spondees and tones at equal loudness became a 12.1-dB difference when the stimuli were converted from SPL to HL. Conclusions Nearly all of the pure-tone average-spondee threshold differences in functional hearing loss are attributable to references for calibration for 0 dB HL for tones and speech, which are based on detection and recognition, respectively. The recognition threshold for spondees is roughly 9 dB higher than the speech detection threshold; persons feigning a loss, who base loss magnitude on loudness, do not consider this difference. Furthermore, the dynamic loudness model was more accurate than the static model.