Electrophysiological indices of discrimination of long-duration, phonetically similar vowels in children with typical and atypical language development.

PURPOSE: The authors investigated the neurophysiological bases of vowel perception in children with specific language impairment (SLI) compared with typical language development (TLD) controls using 250-ms phonetically similar vowels. In a previous study, children with SLI showed a poor neurophysiological response (the mismatch negativity [MMN]) to 50-ms versions of these vowels, regardless of whether attention was directed to (attend) or away (passive) from the auditory modality (V. Shafer, M. Morr, H. Datta, D. Kurtzberg, & R. Schwartz, 2005). They hypothesized that longer vowels would allow for improved speech perception. METHOD: Brain responses were elicited to the vowel contrast in 2 conditions: (a) attend and (b) passive. Behavioral discrimination and identification responses were also examined. RESULTS: Both SLI and TLD groups showed evidence of discrimination (MMN) of the vowels in the attend and passive conditions. Only subtle differences in the scalp topography of a late negative (LN) brain component were observed between groups. The SLI compared with the TLD group showed significantly poorer identification of these long vowels, as found previously with the shorter vowels. CONCLUSIONS: Increased vowel duration can improve discrimination in children with SLI. However, poor identification of these longer vowels by some children with SLI suggests a deficit in long-term phonological representations or accessing these representations.

Neurophysiological indexes of speech processing deficits in children with specific language impairment.

We used neurophysiological and behavioral measures to examine whether children with specific language impairment (SLI) have deficits in automatic processing of brief, phonetically similar vowels, and whether attention plays a role in such deficits. The neurophysiological measure mismatch negativity (MMN) was used as an index of discrimination in two tasks; one in which children ignored the auditory stimuli and watched a silent video and a second in which they attended to the auditory modality. Children with SLI showed good behavioral discrimination, but significantly poorer behavioral identification of the brief vowels than the children with typical language development (TLD). For the TLD children, two neurophysiological measures (MMN and a later negativity, LN) indexed discrimination of the vowels in both tasks. In contrast, only the LN was elicited in either task for the SLI group. We did not see a direct correspondence between the absence of MMN and poor behavioral performance in the children with SLI. This pattern of findings indicates that children with SLI have speech perception deficiencies, although the underlying cause may vary.

Electrophysiological indices of brain activity to "the" in discourse.

In a first experiment, we recorded event-related-potentials (ERPs) to "the" followed by meaningful words (Story) versus "the" followed by nonsense syllables (Nonse). Left and right lateral anterior positivities (LAPs) were seen from the onset of "the" up to 200 ms in both conditions. Later than 200 ms following the onset of "the", the left and right LAPs continued for "the" in the Story, but were replaced by a negativity in the Nonse Condition. In a second experiment, ERPs were recorded to "the" in the Story and Nonse contexts mixed together under two different task instructions (attend to the auditory stimuli versus ignore the auditory stimuli). The same pattern of findings as Experiment 1 were observed for the Story and Nonse contexts when the participants attended to the auditory stimuli. Ignoring the auditory stimuli led to an attenuation of the right LAP, supporting the hypothesis that it is an index of discourse processing.

Effects of sensorineural hearing loss and personal hearing AIDS on cortical event-related potential and behavioral measures of speech-sound processing.

OBJECTIVE: To systematically investigate the combined effects of sensorineural hearing loss and prescribed personal hearing aid(s) on cortical event-related potentials (ERPs) (waves N1, MMN, N2b, and P3b) and their related behavioral measures of discrimination (d-prime sensitivity and reaction time) to the speech sounds /ba/ and /da/ presented at 65 and 80 dB peak-to-peak equivalent SPL. DESIGN: Cortical ERPs were recorded to /ba/ and /da/ speech stimuli presented at 65 and 80 dB peak-to-peak equivalent SPL from 20 normal-hearing adults and 14 adults with sensorineural hearing losses. The degree of sensorineural impairment at 1000 to 2000 Hz ranged from moderate losses (50 to 74 dB HL) to severe-profound losses (75 to 120 dB HL). The speech stimuli were presented in an oddball paradigm and cortical ERPs were recorded in both active and passive listening conditions at both stimulus intensities. The adults with hearing impairments were tested in the unaided and aided conditions at each stimulus intensity. Electroacoustic and real-ear testing was performed on each subject's hearing aid(s) before electrophysiology testing to ensure that the hearing aids were functioning at the time of testing. RESULTS: The use of personal hearing aids substantially improved the detectability of all the cortical ERPs and behavioral d-prime performance scores at both stimulus intensities. This was especially true for individuals with severe-profound hearing losses. At 65 dB SPL, mean ERP amplitudes and d-prime sensitivity scores were all significantly higher or better in the aided versus unaided condition. At 80 dB SPL, only the N1 amplitudes and d-prime sensitivity scores were significantly better in the aided condition. Even though the majority of the hearing-impaired subjects showed increased amplitudes, decreased latencies, and better waveform morphology in the aided condition, the amount of response change (improvements) seen in these measures showed considerable variability across subjects. When compared with the responses obtained from the normal-hearing subjects, both hearing-impaired groups had significantly prolonged aided RT latencies at both stimulus intensities and N2b latencies at the higher stimulus intensities. CONCLUSIONS: These results suggest that hearing-impaired individuals' brains process speech stimuli with greater accuracy and in a more effective manner when these individuals use their personal hearing aids. This is especially true at the lower stimulus intensity. The effects of sensorineural hearing loss and personal hearing aids on cortical ERPs and behavioral measures of discrimination are dependent on the degree of sensorineural loss, the intensity of the stimuli, and the level of cortical auditory processing that the response measure is assessing. The possible clinical significance of these cortical ERP and behavioral findings is discussed.

Language-specific memory traces of consonants in the brain.

This study examined whether experience with a native language affected processing of a place-of-articulation feature. In Experiment 1, 10 stimuli from a continuum of synthesized Hindi bilabial, dental and retroflexed stops were presented to English and Hindi speakers to examine discrimination and identification across the bilabial/dental and dental/retroflexed boundaries. In an oddball task designed to elicit mismatch negativity (MMN), subjects ignored these stimuli while their brain activity was recorded. All participants showed similar behavioral discrimination and identification. However, the English subjects were slower that the Hindi subjects on the discrimination task. All subjects were less accurate and slower at identifying and discriminating the dental and retroflexed compared to the bilabial sounds. A small MMN was observed for some of the bilabial-dental contrasts, but not for the dental-retroflexed contrasts. No group differences were found. In Experiment 2, MMN was observed to a greater stimulus difference (bilabial-retroflexed) and was earlier when the bilabial rather than the retroflexed sound served as the frequent stimulus for both groups. The MMN was also earlier for the Hindi than the English groups when the retroflexed sound served as the frequent stimulus. These results indicate that the Hindi speakers used detailed acoustic-phonetic information for more rapid brain discrimination than the English participants and that the dental-retroflexed discrimination is more difficult than the bilabial-dental discrimination for all speakers.

Maturation of mismatch negativity: a scalp current density analysis.

OBJECTIVE: Auditory evoked potentials provide the opportunity to better understand the central processing of auditory stimuli, which is the basis of speech and language perception. The purpose of this study was to examine maturational changes in the topography of one of these auditory evoked potentials, the mismatch negativity (MMN), using scalp current density (SCD) analysis. DESIGN: Subjects were children ages 4 to 11 yr (N = 53), and adults (N = 12). Stimuli were 85 dB peSPL 1000 Hz standard tones and 1200 Hz deviant tones (deviant probability = 0.15). Auditory evoked potentials were recorded using surface electrodes placed at 32 locations on the head while subjects ignored the stimuli by watching a silent video. RESULTS: Significant maturational changes in topography of MMN were seen over frontal and left lateral sites. CONCLUSIONS: Differences in MMN for the children compared to adults indicate that the MMN generators or their orientation, and thus the neural processes underlying discrimination of simple tones, are not yet mature by 11 yr of age.

Effects of sensorineural hearing loss on cortical event-related potential and behavioral measures of speech-sound processing.

OBJECTIVE: To investigate systematically the effects of sensorineural hearing loss on cortical event-related potentials (ERPs) N1, MMN, N2 and P3 and their associated behavioral measures (d' sensitivity and reaction time) to the speech sounds /ba/ and /da/ presented at 65 and 80 dB ppe SPL. DESIGN: Cortical ERPs were recorded to /ba/ and /da/ speech stimuli presented at 65 and 80 dB ppe SPL from 20 normal-hearing adults and 20 adults who are hearing impaired. The degree of sensorineural impairments at 1000 to 2000 Hz ranged from mild losses (defined as 25 to 49 dB HL) to severe/profound losses (75 to 120 dB HL). The speech stimuli were presented in an oddball paradigm and the cortical ERPs were recorded in both active and passive listening conditions for each stimulus intensity. RESULTS: Both ERP amplitudes and behavioral discrimination (d') scores were lower for listeners with sensorineural hearing loss than for those with normal hearing. However, these differences in response strength were evident only for those listeners whose average hearing loss at 1000 to 2000 Hz exceeded 60 dB HL for the lower intensity stimuli and exceeded 75 dB HL for the higher intensity stimuli. In contrast, prolongations in the ERP and behavioral latencies, relative to responses from normal-hearing subjects, began with even mild (25 to 49 dB HL) threshold elevations. The amplitude and latency response changes that occurred with sensorineural hearing loss were significantly greater for the later ERP peaks (N2/P3) and behavioral discrimination measures (d' and RT) in comparison with earlier (N1, MMN) responses. CONCLUSIONS: The results indicate that latency measures are more sensitive indicators of the early effects of decreased audibility than are response strength (amplitude, d' or percent correct) measures. Sensorineural hearing loss has a greater impact on higher level or "nonsensory" cortical processing in comparison with lower level or "sensory" cortical processing. Possible physiologic mechanisms within the cortex that may be responsible for these response changes are presented. Lastly, the possible clinical significance of these ERP and behavioral findings is discussed.

Maturation of mismatch negativity in typically developing infants and preschool children.

OBJECTIVE: 1) To determine whether an adult-like mismatch negativity (MMN) can be reliably elicited in typically developing awake infants and preschool children, and if so 2) to examine whether maturational changes exist in MMN latencyand amplitude. DESIGN: Two experiments were designed to elicit MMN using an "oddball" paradigm. In Experiment 1, a 1000-Hz tone served as the standard stimulus and a 1200-Hz tone as the deviant. In Experiment 2, a 1000-Hz standard stimulus and a 2000-Hz deviant were presented. Infants' ages ranged from 2 to 47 and 3 to 44 mo in Experiments 1 and 2, respectively. RESULTS: In Experiment 1, a negativity was not elicited in the majority of the infants and preschoolers tested. In Experiment 2, a negativity was reliably elicited in the infants and preschoolers across all ages. A significant negative correlation was observed between age and latency, but not for age and amplitude for this negativity. This negativity was found to decrease at a rate of 1 msec/mo. Infants younger than 12 mo of age showed a significantly larger positivity to the deviant than to the standard between 150-300 and 200-300 msec in Experiments 1 and 2, respectively. CONCLUSIONS: The discriminative processes indexed by MMN in response to frequency changes areimmature in infants and preschool children. Although there is convincing evidence that the negativity elicited in Experiment 2 is an immature MMN, the possibility that it may be an "obligatory effect" indexing recovery from refractoriness cannot be ruled out at this time. The results from these experiments suggest that the MMN component haslimited use as a clinical tool at this time for infants and young children.