Slide Slam O13
Acoustic and language-specific processing of phonological sequence onsets in the low gamma- and theta frequency bands
Monica Wagner1, Silvia Ortiz-Mantilla2, Mateusz Rusiniak3, April Benasich2, Valerie Shafer4, Mitchell Steinschneider5; 1St. John's University, Queens, New York, USA, 2Rutgers University, Newark, New Jersey, USA, 3BESA, GmbH, Gräfelfing, Germany, 4The Graduate Center of the City University of New York, New York, USA, 5Albert Einstein College of Medicine, Bronx, New York, USA
Temporal processing models suggest that the rate of acoustic change of syllabic level- and phonemic level-structures give rise to entrainment in the theta- and low gamma (LG) frequency bands, respectively. It has been suggested that this hierarchical-coordinated decoding of speech facilitates speech perception. However, few studies have investigated whether language exposure to phonotactic patterns modulates sensory processing in the theta- and LG bands. To elucidate processes that facilitate speech perception, time-frequency (T-F) analyses of EEGs obtained from 24 native speakers of English and 24 native speakers of Polish (late language learners of English) were conducted. Participants listened to same and different nonword pairs within counterbalanced attend- and passive-listening conditions, with each testing condition separated by a minimum of two months. Nonwords within the pairs contained the phonological sequence onsets /pt/, /pət/, /st/, and /sət/ that occur in both the Polish and English languages with the exception of /pt/, which never occurs in English without a preceding vowel. A source localization model to the first word in the pairs was created from auditory-evoked potentials (AEP) that transformed the 64-channel EEG into brain source-level channels for T-F analyses (BESA Research 7.1). The model was created from ~53,000 single AEP trials filtered between 1-55 Hz with 1500 ms epochs (-500 pre- 1000 post-onset). Goodness of fit for the model for each onset, language group and condition was greater than 98%. T-F measures of spectral power and inter-trial phase locking (ITPL) in the LG- and theta bands were analyzed from two bilateral, auditory cortical-source channels. Between-language-group analyses to each of the four onsets examined, separately, revealed a single significant language effect. Larger spectral power in LG was found for the English listeners to the unfamiliar /pt/ onsets in the right hemisphere, at early cortical stages during the passive condition. The English group response to the /pt/ onset, which has zero probability of occurrence in the English language, might reflect a right-hemisphere novelty effect. Sensory processing appeared bilateral for the English listeners, whereas the Polish listeners showed a clear left asymmetry to the /pt/ onsets in LG. Onset contrasts that differed by one phoneme (/pt-pət/, /st-sət/, /pt-st/, /pət-sət/) were also examined in each language group, separately. The ITPL pattern to each phonological sequence within the contrast was specific in time and frequency sub-bands within theta. Further, the ITPL patterns were highly replicable across language groups, but appeared modulated by frequency of occurrence of the onset contrast within the language. For example, the pattern of ITPL differences to the /pt-pət/ contrast was similar across language groups, however, differences in ITPL values reached significance only for the Polish group. Thus, phase-locking may be more variable for phonotactic patterns that are absent or rare in one’s language. Language-specific processing in LG and acoustic- and language-specific processing in theta support temporal processing models that view phonemic-level segmentation in LG, along with coordinated activations between theta and LG, to supports speech perception. This study has translational relevance for assessment of individuals with developmental language disorder, dyslexia, and auditory processing disorder.