Perception of Indexical Cues in Adults and Children - XL

A cochlear implant (CI) is a prosthetic device that partially restores the
hearing of profoundly deaf individuals. However, the speech signal that is
delivered via the device is spectro-temporally degraded. As a consequence, CI
users have difficulties with recognizing and distinguishing the voices of
talkers, but the extent of this deficiency is currently not fully known. This
limitation may have an impact on speech communication in this population, but
the effects of limited voice perception on speech communication have not been
investigated. This problem is complex, requiring expertise from a range of
fields, such as engineering, linguistics, and medical sciences. We are now in
the position to be able to do so, with a systematic approach, thanks to our new
NWO VICI (awarded to D. Ba*kent) and VENI (awarded to T.N. Tamati) funding.

The main focus of this research is on the perception of talkers* voices and the
role of talkers* voices in speech communication across multiple listener
groups. The current study will target the perception of acoustic cues
characterizing talkers* voices, including but not limited to fundamental
frequency (F0), related to the glottal pulse rate of the vocal chords, and
vocal tract length (VTL), related to the size of the talker. These and other
voice characteristics are crucial for communication and speech perception. They
deliver essential information that can be used to identify a talker*s voice,
extract the actual meaning of words (emotions, sarcasm), adapt to a talker*s
voice to better understand his/her speech, and segregate speech from noise and
other speech streams in noisy environments. Since most speech communication in
everyday life involves dealing with multiple talkers in noisy environments, the
impaired perception of voice characteristics experienced by CI can considerably
and negatively affect their ability to successfully communicate with others.

Earlier research from our research group has provided critical evidence that
the perception of acoustic cues related to the perception of talkers* voices,
such as F0 and VTL, is impaired in adult CI users (Gaudrain & Ba*kent, 2015;
Gaudrain & Ba*kent, in press). This impairment has already been shown to lead
to abnormal patterns of vocal gender categorization and emotion recognition
compared to normal-hearing (NH) listeners (Fuller et. al, 2014; Gilbers et al.,
2015). A new study in our lab is currently investigating these limitations in
children (METc NL59930.042.16, Perception of Indexical Cues in Adults and
Children). The current larger study will extend these preliminary studies by
investigating the perception of talkers* voices in children and adults with
normal hearing, hearing impairment (HI), with or without CIs. In particular, we
will further investigate the perception of talkers* voices (and voice-related
acoustic cues), and its relation with the perception of other speech-related
indexical information (e.g., talker identity, gender, accents) and the
perception, understanding, and robust learning of linguistic information (e.g.,
sounds, words, sentences).

To investigate the role of talkers* voices in speech communication in NH, HI,
and CI children and adult listeners, and how the perception of talker voice
information relates to the processing of indexical and linguistic information
among these individuals.

The study consists of questionnaires concerning the language, hearing, and
demographic status of participants, and a cross-sectional observational study
that consists of behavioural experiments, eye-tracking, and EEG experiments on
the perception of voice cues, speech perception and linguistic processing.

There are no known risks or benefits associated with participation. All tests
are non-invasive and consist of simple tasks, such as listening to a recorded
sound or speech sample, and making a judgment about the voice or meaning of
this sample. The total duration will be broken into multiple sessions on
multiple days to minimize fatigue and maximize alertness. Session duration and
the number of sessions will be adjusted for each participant as needed. To
further reduce potential fatigue, adequate breaks will be built into the
experiment and will also be given on request of the participant.