Kernel Learning for Sound Field Estimation with L1 and L2 Regularizations

Purpose This study investigated the nature of the association of lexical–grammatical abilities within and across languages in Latino dual language learners (DLLs) with specific language impairment (SLI) using language-specific and bilingual measures. Method Seventy-four Spanish/English–speaking preschoolers with SLI from preschools serving low-income households participated in the study. Participants had stronger skills in Spanish (first language [L1]) and were in the initial stages of learning English (second language [L2]). The children's lexical, semantic, and grammar abilities were assessed using normative and researcher-developed tools in English and Spanish. Hierarchical linear regressions of cross-sectional data were conducted using measures of sentence repetition tasks, language-specific vocabulary, and conceptual bilingual lexical and semantic abilities in Spanish and English. Results Results indicate that language-specific vocabulary abilities support the development of grammar in L1 and L2 in this population. L1 vocabulary also contributes to L2 grammar above and beyond the contribution of L2 vocabulary skills. However, the cross-linguistic association between vocabulary in L2 and grammar skills in the stronger or more proficient language (L1) is not observed. In addition, conceptual vocabulary significantly supported grammar in L2, whereas bilingual semantic skills supported L1 grammar. Conclusions Our findings reveal that the same language-specific vocabulary abilities drive grammar development in L1 and L2 in DLLs with SLI. In the early stages of L2 acquisition, vocabulary skills in L1 also seem to contribute to grammar skills in L2 in this population. Thus, it is critical to support vocabulary development in both L1 and L2 in DLLs with SLI, particularly in the beginning stages of L2 acquisition. Clinical and educational implications are discussed.

Download Full-text

Measuring Word Learning Ability in Sequential Bilingual Children

Perspectives on Communication Disorders and Sciences in Culturally and Linguistically Diverse Populations ◽

10.1044/cds17.1.25 ◽

2010 ◽

Vol 17 (1) ◽

pp. 25-32 ◽

Cited By ~ 3

Author(s):

Pui Fong Kan

Keyword(s):

Word Learning ◽

Learning Ability ◽

Bilingual Children ◽

Word Knowledge ◽

Typically Developing ◽

Learning Skills ◽

Word Forms ◽

The Individual ◽

Novel Word Learning ◽

L1 And L2

Abstract The purpose of this article is to look at the word learning skills in sequential bilingual children—children who learn two languages (L1 and L2) at different times in their childhood. Learning a new word is a process of learning a word form and relating this form to a concept. For bilingual children, each concept might need to map onto two word forms (in L1 and in L2). In case studies, I present 3 typically developing Hmong-English bilingual preschoolers' word learning skills in Hmong (L1) and in English (L2) during an 8-week period (4 weeks for each language). The results showed gains in novel-word knowledge in L1 and in L2 when the amount of input is equal for both languages. The individual differences in novel word learning are discussed.

Download Full-text

Decay of Temporary Threshold Shift in Noise

Journal of Speech and Hearing Research ◽

10.1044/jshr.1602.267 ◽

1973 ◽

Vol 16 (2) ◽

pp. 267-270 ◽

Cited By ~ 5

Author(s):

John H. Mills ◽

Seija A. Talo ◽

Gloria S. Gordon

Keyword(s):

Sound Field ◽

Threshold Shift ◽

Temporary Threshold Shift ◽

Octave Band ◽

Band Noise ◽

Lower Asymptote

Groups of monaural chinchillas trained in behavioral audiometry were exposed in a diffuse sound field to an octave-band noise centered at 4.0 k Hz. The growth of temporary threshold shift (TTS) at 5.7 k Hz from zero to an asymptote (TTS ∞ ) required about 24 hours, and the growth of TTS at 5.7 k Hz from an asymptote to a higher asymptote, about 12–24 hours. TTS ∞ can be described by the equation TTS ∞ = 1.6(SPL-A) where A = 47. These results are consistent with those previously reported in this journal by Carder and Miller and Mills and Talo. Whereas the decay of TTS ∞ to zero required about three days, the decay of TTS ∞ to a lower TTS ∞ required about three to seven days. The decay of TTS ∞ in noise, therefore, appears to require slightly more time than the decay of TTS ∞ in the quiet. However, for a given level of noise, the magnitude of TTS ∞ is the same regardless of whether the TTS asymptote is approached from zero, from a lower asymptote, or from a higher asymptote.

Download Full-text

Modified Earpieces and CROS for High Frequency Hearing Losses

Journal of Speech and Hearing Research ◽

10.1044/jshr.1101.204 ◽

1968 ◽

Vol 11 (1) ◽

pp. 204-218 ◽

Cited By ~ 9

Author(s):

Elizabeth Dodds ◽

Earl Harford

Keyword(s):

Hearing Loss ◽

Hearing Aids ◽

High Frequency ◽

Hearing Aid ◽

Sound Field ◽

Intensity Level ◽

Test Results ◽

High Frequency Hearing Loss ◽

Increased Sensitivity ◽

Difficult Cases

Persons with a high frequency hearing loss are difficult cases for whom to find suitable amplification. We have experienced some success with this problem in our Hearing Clinics using a specially designed earmold with a hearing aid. Thirty-five cases with high frequency hearing losses were selected from our clinical files for analysis of test results using standard, vented, and open earpieces. A statistical analysis of test results revealed that PB scores in sound field, using an average conversational intensity level (70 dB SPL), were enhanced when utilizing any one of the three earmolds. This result was due undoubtedly to increased sensitivity provided by the hearing aid. Only the open earmold used with a CROS hearing aid resulted in a significant improvement in discrimination when compared with the group’s unaided PB score under earphones or when comparing inter-earmold scores. These findings suggest that the inclusion of the open earmold with a CROS aid in the audiologist’s armamentarium should increase his flexibility in selecting hearing aids for persons with a high frequency hearing loss.

Download Full-text