Maria V. Kondaurova

Analysis of categorical response data: Use logistic regression rather than endpoint-difference scores or discriminant analysis

Example of a typical second-language (L2) speech perception experiment: Synthetic vowel stimuli from a two-dimensional grid of points in which acoustic properties vary systematically in duration and spectral properties are classified as English /i/ or /I/ by L2-English listeners. In a number of studies, the data from such experiments have been analyzed using endpoint-difference scores or discriminant analysis. The current letter describes theoretical problems inherent in the first procedure in general, and in the application of the second procedure to data of this type in particular. Logistic regression is proposed as an alternative, which does not suffer from these problems.

Age-Related Changes in Prosodic Features of Maternal Speech to Prelingually Deaf Infants with Cochlear Implants

This study investigated prosodic and structural characteristics of infant-directed speech to hearing-impaired infants as they gain hearing experience with a cochlear implant over a 12-month period of time. Mothers were recorded during a play interaction with their HI infants (N = 27, mean age 18.4 months) at 3, 6, and 12 months post-implantation. Two separate control groups of mothers with age-matched normal-hearing infants (NH-AM) (N = 21, mean age 18.1 months) and hearing experience-matched normal-hearing infants (NH-EM) (N = 24, mean age 3.1 months) were recorded at three testing sessions. Mothers produced less exaggerated pitch characteristics, a larger number of syllables per utterance, and faster speaking rate when interacting with NH-AM as compared to HI infants. Mothers also produced more syllables and demonstrated a trend suggesting faster speaking rate in speech to NH-EM relative to HI infants. Age-related modifications included decreased pitch standard deviation and increased number of syllables in speech to NH-AM infants and increased number of syllables in speech to HI and NH-EM infants across the 12-month period. These results suggest that mothers are sensitive to the hearing status of their infants and modify characteristics of infant-direct speech over time.

Phonetic Modification of Vowel Space in Storybook Speech to Infants up to 2 Years of Age

PURPOSE A large body of literature has indicated vowel space area expansion in infant-directed (ID) speech compared with adult-directed (AD) speech, which may promote language acquisition. The current study tested whether this expansion occurs in storybook speech read to infants at various points during their first 2 years of life. METHOD In 2 studies, mothers read a storybook containing target vowels in ID and AD speech conditions. Study 1 was longitudinal, with 11 mothers recorded when their infants were 3, 6, and 9 months old. Study 2 was cross-sectional, with 48 mothers recorded when their infants were 3, 9, 13, or 20 months old (n=12 per group). The 1st and 2nd formants of vowels /i/, /ɑ/, and /u/ were measured, and vowel space area and dispersion were calculated. RESULTS Across both studies, 1st and/or 2nd formant frequencies shifted systematically for /i/ and /u/ vowels in ID compared with AD speech. No difference in vowel space area or dispersion was found. CONCLUSIONS The results suggest that a variety of communication and situational factors may affect phonetic modifications in ID speech, but that vowel space characteristics in speech to infants stay consistent across the first 2 years of life.

Communicative intent and affect in mothers’ speech to hearing-impaired infants with cochlear implants

Emotional properties of infant-directed speech influence normal-hearing (NH) infants’ attention to speech sounds. The current study examines communicative intent/affect in speech to hearing-impaired (HI) infants following the first year of cochlear implantation. Mothers of HI infants (HI group, ages 13.3—25.5 months), NH age-matched infants (NH-AM group, ages 13.5—25.7 months) and NH experience-matched infants (NH-EM group, ages 2.3—3.6 months) were recorded playing with their infants at three sessions over the course of one year. 25-second speech samples were low-pass filtered, leaving pitch but not speech information intact. Twelve adults rated stimuli along five scales of communicative intent/affect: Positive/Negative Affect, Intention to Express Affection, Encourage Attention, Comfort/Sooth and Direct Behavior. ANOVAs demonstrated main effects of Group and/or Session for all scales (ps = 0.01 to 0.07). Speech to HI and NH-EM infants was more positive, affective, encouraging, and comforting than speech...

Russian and Spanish listener’s perception of the English tense/lax vowel contrast: Contributions of native language allophony and individual experience

We examined the influence of listeners native phonology on the perception of American English tense and lax front unrounded vowels ([i] and [■]). These vowels are distinguishable according to both spectral quality and duration. Nineteen Russian, 18 Spanish, and 16 American English listeners identified stimuli from a beat‐bit continuum varying in nine spectral and nine duration steps. English listeners relied predominantly on spectral quality when identifying these vowels, but also showed some reliance on duration. Russian and Spanish speakers relied entirely on duration. Three additional tests examined listeners allophonic use of vowel duration in their native languages. Duration was found to be equally important for the perception of lexical stress for all three language groups. However, the use of duration as a cue to postvocalic consonant voicing differed due to phonotactic differences across the three languages. Group results suggest that non‐native perception of the English tense/lax vowel contrast i...

Vocal turn-taking between mothers and their hearing-impaired infants with cochlear implants

Normal-hearing (NH) infants participate in social exchanges soon after birth. What does the temporal organization of vocal turn-taking (VTT) look like in infants with hearing loss with cochlear implants? This study examined VTT during spontaneous play in eight dyads of mothers and their hearing-impaired (HI) or age-matched NH infants (mean age 17.4 m at time 1). Dyads came to two sessions, corresponding to 3 and 12 months post-implantation. Analyses demonstrated that although HI infants vocalized less than NH infants, the proportion of vocalizations involved in VTT exceeded that from NH infants at time 1 and was equal at time 2. For vocalizations involved in VTT, there was a higher proportion of simultaneous speech in HI dyads compared to NH dyads at time 1, but the direction reversed at time 2. At time 1, the number of turns was greater in the HI group but decreased compared to the NH group at time 2. Duration of between-speaker pauses (BSP) was shorter in infant-mother compared to mother-infant turns in...

Vocal matching in interactions between mothers and their normal-hearing and hearing-impaired twins

Vocal matching, the ability to imitate phonetic properties of speech, was examined in interactions between mothers and their normal-hearing (NH) and hearing-impaired twins who used hearing aids (HAs) or a cochlear implant (CI). Vocalizations of three mother-twin triads were recorded in three sessions over 12 months. In one triad, the twins were 15.8 months old and NH. In another triad, the twins were 11.8 months; one was NH while the other had HAs. In the third triad, both twins were 14.8 months; one was NH while the other had a CI. A vocal match was defined as an instance of perceptual and acoustic similarity between adjacent maternal and infant utterances in relation to pitch height and contour, utterance duration, rhythm, or vowels and consonants. Reciprocal vocal matching occurred in 28% to 38% of infant vocalizations across triads. At session three, CI and HA infants’ reciprocal vocal matches increased compared to two previous sessions and to those of NH siblings; reciprocal vocal matches in the NH d...

Acoustic characteristics of infant-directed speech to normal-hearing and hearing-impaired twins with hearing aids and cochlear implants: A case study

The study examined acoustic characteristics of maternal speech to normal-hearing (NH) and hearing-impaired (HI) twins who received hearing aids (HAs) or a unilateral cochlear implant (CI). A mother of female-male NH twins (NH-NH; age 15.8 months), a mother of two male twins, one NH and another HI with HAs (NH-HA; age 11.8 months) and a mother of a NH female twin and a HI male twin with a CI (NH-CI; age 14.8 months) were recorded playing with their infants during three sessions across a 12-month period. We measured pitch characteristics (normalized F0 mean, F0 range, and F0 SD), utterance and pause duration, syllable number, and speaking rate. ANOVAs demonstrated that speech to NH-NH twins was characterized by lower, more variable pitch with greater pitch range as compared to speech to NH-HA and NH-CI pairs. Mothers produced more syllables, had faster speaking rate and longer utterance duration in speech to NH-NH than the other pairs. The results suggest that the pediatric hearing loss in one sibling affec...

Prosodic characteristics of speech directed to adults and to infants with and without hearing loss.

Studies have shown that infant‐direct speech (IDS) is distinguished from adult‐directed speech (ADS) via a variety of acoustic‐prosodic characteristics, including fundamental frequency and rate. However, little is known about how IDS and ADS may be distinguished in terms of linguistically relevant prosodic constructs, such as pitch accents, nor how these constructs map onto acoustic differences previously identified for IDS vs ADS. To investigate these issues, longitudinal recordings of mothers reading a storybook to their typically developing infants and to an experimenter were made when the infants were approximately 3, 6, or 9 months old. Trained analysts used an annotation system to code prosodic characteristics, including rhythmic prominences, pitch accents, and intonation phrase boundaries. Preliminary analyzes of labeled corpora revealed that mothers produced significantly more prominences in IDS than in ADS. In addition, mothers showed substantial individual differences in rates of prominence prod...

Comparison of acoustic characteristics of American English tense and lax vowels in maternal speech to prelingually deaf infants, normal‐hearing infants, and adults.

Recent studies have demonstrated that mothers exaggerate phonetic properties of infant‐directed (ID) speech. However, these studies focused on a single acoustic dimension, i.e., frequency, whereas speech sounds are composed of multiple acoustic cues. Moreover, little is known how mothers adjust phonetic properties of speech to children with hearing loss. This study examined the mothers’ production of frequency and duration cues to the American English tense/lax vowel contrast in speech to profoundly deaf (N = 12) and normal‐hearing (N = 12) infants and to an adult experimenter. First and second formant frequencies and vowel duration of tense ([i, [u]) and lax ([i], [u]) vowels were measured. Results demonstrated that mothers exaggerated vowel duration in ID relative to adult‐directed speech. However, only a trend suggesting an exaggeration of vowel space in ID speech was observed. These findings suggest that although both spectral and duration cues to the tense/lax distinction are modified in a systematic...