Emily Cibelli

Postdoctoral research associate
Department of Linguistics
Northwestern University

Current projects
Speech biomarkers of mental illness
Collaborators: Vijay Mittal, Matt Goldrick, Joseph Keshet

Psychotic disorders such as schizophrenia are known to impact motor control circuits. Effects of this disruption are observable in head and limb movement both in individuals diagnosed with such a disorder, and to some extent in young adults who are at high risk, but have not yet received a diagnosis. It is plausible that these disruptions also impact speech motor control; evidence of such an effect could point to speech as a novel biomarker for psychotic disorders, ultimately facilitating early intervention. In this project, we are using automatic phonetic analysis tools to process a corpus of spoken data from high-risk youth to look for prosodic and phonetic correlates of motor control disruptions in speech in this population.

Sichlinger, L., Cibelli, E., Mittal, V., & Goldrick, M. (In press). Clinical correlates of aberrant conversational turn-taking in youth at clinical high-risk for psychosis. Schizophrenia Research.
Presented at: The Acoustical Society of America (2017, Boston); the Individual Differences Workshop (2018, Northwestern University)


Phonetic echoes of cognitive disruptions
Collaborators: Matt Goldrick, Rhonda McClain, Joseph Keshet, Yossi Adi

This project is comprised of a set of studies designed to investigate cognitive control of lexical access during speech production. We use phonetic properties of speech produced in the lab to look for traces of lexical competition in populations which may be particularly susceptible to the cognitive demands of competition: older speakers, and bilinguals. In addition to our theoretical goals, this project is also concerned with the development of tools to automatically align speech and generate phonetic measurements for rapid processing of speech production data.

Adi, Y., Keshet, J., Cibelli, E., Gustafson, E., Clopper, C., & Goldrick, M. (2016). Automatic measurement of vowel duration via structured prediction. The Journal of the Acoustical Society of America 140 (6), 4517-4527.
Adi, Y., Keshet, J., Cibelli, E., & Goldrick, M. (2017, March). Sequence segmentation using joint RNN and structured prediction models. In the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 2422-2426). IEEE.
Tools tutorial: Deep Phonetic Tools tutorial


Tongue twisters and articulatory variability
Collaborators: Erin Madigan, Gia Eapen, Matt Goldrick

This project looks at the articulatory variability inherent in speech production data generated during tongue twister tasks. Our study is aimed to test and replicate existing effects indicating that multiple phonetic targets are active and leave traces in articulation during twister repetition. In addition, we plan to test the effects of different baseline tasks on the extent of articulatory varability reported in these tasks, and to compare different approaches to modeling this varaibility on a single data set.

Past projects
Color perception and categorization
Collaborators: Yang Xu, Joe Austerweil, Tom Griffiths, Terry Regier

Current and former members of the Language and Cognition Lab and the Computational Cognitive Science Lab at Berkeley are probing the nature and stability of color categorization and perception. We're testing categorical effects on color perception within a Bayesian framework in order to better understand how perceivers handle stimuli that are prototypical or marginal members of categories.

Cibelli, E., Xu, Y., Austerweil, J. L., Griffiths, T. L., & Regier, T. (2016). The Sapir-Whorf Hypothesis and Probabilistic Inference: Evidence from the Domain of Color. PloS one, 11(7), e0158725.
(*Co-first authors.)


Acquisition of non-native phonemes

The sound system of your native language strongly influences the structure of your speech perception and production systems. In some cases, this can make it challenging to acquire sounds in a new language. In my dissertation, I explored the types of information that can contribute to the stable acquisition of a new phonemic category in a second language, and how those novel categories are represented neurally.

My dissertation: Aspects of articulatory and perceptual learning in novel phoneme acquisition
Presented at: The Acoustical Society of America (2014, Providence; 2015, Pittsburgh); The Linguistic Society of America (2016, Washington, D.C.)


Neural pathways of lexical representations
Collaborators: Matt Leonard, Keith Johnson, Eddie Chang

With collaborators at UCSF, I investigated the neural pathways of speech using ECoG. Our focus was on the influence of stored lexical knowledge during spoken language processing in the left temporal lobe.

Published as: Cibelli, E.S., Leonard, M.K., Johnson, K., & Chang, E.F. (2015). The influence of lexical statistics on temporal lobe cortical dynamics during spoken word listening. Brain and Language 147, 66-75.


Metrical timing
Collaborators: Sam Tilsen, John Houde, Shinae Kang

Berkeley linguists and members of the UCSF Biomagnetic Imaging Lab collaborated on behavioral and MEG studies designed to measure the effect of rhythmic regularity on production, and what that reveals about the nature of speech planning.

Manuscript: Tilsen, S.,Cibelli, E., Kang, S., Houde, J., and Nagarajan, S. (2012). Acoustic analysis of the effects of metrical regularity on interval durations. Cornell Working Papers in Phonetics and Phonology, 38-50.


Speech and aging
Collaborators: Susanne Gahl, Kat Hall, Ronald Sprouse

Our team developed a longitudinal corpus of speech from the Up Series, a set of documentaries which have interviewed several individuals in Britain every seven years since the age of seven. The corpus provides source material to investigate the development of speech over the entire lifespan, with particular attention to the understudied productions of middle-age speakers.

Published as: Gahl, S., Cibelli, E., Hall, K., and Sprouse, R. (2014). The "Up" corpus: A corpus of speech samples across adulthood. Corpus Linguistics and Linguistic Theory 10(2), 315-328.


Heritage language attrition and reclamation

I worked with heritage speakers of the dialect of Finnish spoken by Americans of Finnish descent (variously known as "Finglish," fingelska, and amerikansuomi). In addition to documenting lexical items that vary from standard Finnish, I measured the phonetic qualities of the speech of heritage speakers who are relearning Finnish in adulthood. The goal was to examine what native-like phonetic aspects of the phonological system are maintained as a function of early childhood exposure, even when the language has not been used for long periods in adulthood.

Poster presented at: The Road Less Travelled workshop on heritage languages, University of Toronto, October 2012
Talk presented at: The Society for the Advancement of Scandinavian Study, San Francisco, May 2013


Cross-linguistic effects on speech perception
Collaborators: Keith Johnson, Shinae Kang

Our group investigated how native language experience is integrated with cues from both auditory and visual information to investigate compensation for coarticulation.

Talk presented at: The Acoustic Society of America, San Francisco, December 2013


The pitch effects of prenasalized consonants

In a recent paper, I measured the intrinsic phonetic effects of prenasalized consonants on the realization of pitch in Chichewa syllables, in an effort to shed light on a puzzling phonological pattern in tone languages.

Published in: Cibelli, E. (2015). The phonetic basis of a phonological pattern. The Phonetics-Phonology Interface: Representations and methodologies, 335, 171-192.
Talk presented at: Phonetics and Phonology in Iberia, Universitat Rovira i Virgili, Tarragona, 2011


Phonetic convergence in bilinguals

At Wellesley, I conducted a behavioral study which measured the degree of phonetic convergence to a speaker with a non-native accent. Participants included monolingual English speakers, balanced English-Spanish bilinguals, and bilinguals with one language more dominant than the other, in order to determine the extent to which language history and dominance influenced the tendency to converge to accented speech.

Download the thesis here
Presented at: The Mellon 23 Faculty Workshop in Linguistics, Swarthmore, 2009


Perception of non-native speech sounds
Collaborators: Hayley Sutherland, Andrea Levitt, Cathi Best

While at Wellesley, I worked on a series of behavioral experiments and fMRI data sets aimed at achieving a better understanding of how people perceive, categorize, and assimilate speech sounds which don't occur in their native language.


Developmental patterns of ambiguity resolution
Collaborators: Jesse Snedeker, Amanda Worek

A summer internship during college brought me to the Harvard Lab for Developmental Studies, where I worked on a project designed to understand how children use real-world knowledge to help resolve syntactic ambiguities.

Home | Northwestern University | Department of Linguistics | SoundLab | Disclaimer