2023 Seminars
February 2023
| HRC Seminar with Zhenghan Qi | February 24th |
Zhenghan Qi, Northeastern University
Title: Statistical learning: A multi-faceted process sculptured by brain development
Abstract: Statistical learning (SL) is an implicit and multi-faceted process, through which we detect and acquire the structured patterns in our environment, often without awareness. From infants to adults, humans are equipped with robust sensitivity to statistical information embedded in sensory inputs. The recent research on SL has brought fresh perspectives that emphasize the componential and complex nature of the learning mechanisms above and beyond a domain-general view. The work, motivated by the behavioral differences in SL seen in neurodevelopmental disorders, investigates the underlying cognitive and neural mechanisms through the lens of human development. We ask how statistical information is processed in mature and developing brains that differ in both higher-order cognition and language experiences. Our findings suggest shifting roles of attention and language networks during SL across development, implicating different processing mechanisms underlying similar learning outcomes.
March 2023
| HRC Seminar with Emily Stephen | March 17th |
Title: Multivariate temporal receptive fields in speech perception reflect low-dimensional dynamics
Abstract: Research into the cortical basis of auditory speech perception has successfully modeled how high gamma (70-150Hz) responses in electrocorticography (ECoG) over the superior temporal gyrus (STG) encode phonetic features such as consonants and vowels. Furthermore, some STG populations respond to sentence onsets and acoustic onset edges (“peak rate” events), which represent sentence-level and syllable-level timing, respectively. In this talk, I will present work suggesting that these timing representations are shared across populations in a low-dimensional latent state. These spatially distributed timing signals could serve to provide temporal context for, and possibly bind across time, the concurrent processing of individual phonetic features, to compose higher-order phonological (e.g., word-level) representations.
April 2023
| HRC Seminar with Erika Skoe | April 21st |
Erika Skoe, University of Connecticut
Title: Social and environmental influences on hearing: Insights from circulating blood and wearable technologies
Abstract: Real-world auditory environments have rich acoustic properties that fluctuate in frequency and intensity over multiple time scales, from milliseconds to hours to days. In the face of this complexity, how does the auditory system respond to the prevailing sound conditions in different behavioral and social contexts? Detailed investigations of how the auditory system is influenced by environmental sound conditions have largely been undertaken almost exclusively in nonhuman mammals. Whether results generalize to humans is currently poorly understood. This knowledge gap is due partly to the technical challenges of obtaining nuanced acoustic measures of a listener’s long-term auditory environment. Across a series of recent investigations, we have been working to address this gap by using portable, wearable sound recorders to characterize the acoustic characteristics of a person’s everyday soundscape. This ecoacoustic data has helped reveal surprising sound exposure patterns, dissociate the effects of auditory training and trauma on auditory function, and — when combined with novel blood-based biomarkers — has given insight into where along the auditory pathway, environment-dependent changes first occur.
| HRC Seminar with Bertrand Delgutte | April 28th |
Bertrand Delgutte, Massachusetts Eye and Ear Infirmary, Harvard Medical School
Title: In search of the missing resolved harmonics: Neural and behavioral studies of pitch discrimination
Abstract: The pitch of harmonic complex tones (HCT) plays important roles in speech and music perception, and in the perceptual organization of sound. Human listeners rely primarily on the spectral pattern of harmonics individually resolved by the frequency analysis in the cochlea to derive a pitch percept. Yet, evidence for a robust neural representation of resolved harmonics in animal models has been elusive, and some species appear to rely on temporal envelope cues rather than on resolved harmonics to discriminate pitch. We performed both behavioral experiments on pitch discrimination by rabbits and single-unit recordings from the auditory nerve (AN) and inferior colliculus (IC) using HCTs with equal-amplitude harmonics in which the fundamental frequency (F0) was varied over a wide range. We found that rabbits can discriminate the pitch of HCT using the spectral pattern of resolved harmonics for F0s within the range of conspecific vocalizations. Correspondingly, many IC neurons contribute to a rate-place code for the frequencies of resolved harmonics for F0s > 500 Hz, and this code is more robust to variations in stimulus level than that found in the AN. A remaining challenge is to identify the neural mechanisms underlying enhanced rate-place coding in IC.
May 2023
| HRC Seminar with Andrew Oxenham | May 5th |
Andrew Oxenham, University of Minnesota
Title: Context effects in auditory and speech perception
Abstract: Our perception depends strongly on our expectations and the context in which stimuli occur. The way our auditory system adapts to ongoing sounds produces context effects that may help not only in enhancing novel sounds in the environment, but also in maintaining the perceptual constancy of sounds in the face of variable acoustics produced by different talkers, different room environments, and different background noise. Despite their importance and ubiquity, context effects in hearing have received much less attention than analogous effects in the visual domain. This talk will review a number of auditory and speech context effects and will examine the potential underlying neural mechanisms, as well as the effects of hearing loss, including cochlear implants, on the influence of context.
