You are viewing the SNL 2018 Archive Website. For the latest information, see the Current Website.

Poster C7, Friday, August 17, 10:30 am – 12:15 pm, Room 2000AB

Phonological rule application recruits left inferior frontal gyrus

Mathias Scharinger1,2,3, Johanna Steinberg3,4, Frank Zimmerer5, Jonas Obleser6;1Phonetics Research Group, Department of German Linguistics, Philipps-University Marburg, Germany, 2Center for Mind, Brain and Behavior, Philipps-University Marburg, Germany, 3Cognitive and Biological Psychology, University of Leipzig, Germany, 4AGAPLESION Diakonieklinikum Rotenburg GmbH, Rotenburg (Wuemme), Germany, 5Language Science & Technology, Saarland University, Germany, 6Institute for Psychology, University of Luebeck

Auditory processing in speech follows a trajectory emanating from primary auditory cortices towards anterior and posterior locations. With increasing abstraction from acoustic-phonetic stimulus details, processing areas are further away from primary auditory locations. Corroborating phonological theory positing such abstraction from acoustic-phonetic detail, recent research has established superior temporal sulcus and inferior frontal gyrus as possible phonological processing areas. However, most studies referred to “phonology” by rather simple definitions, therefore leaving some gaps between theoretical phonology and cognitive neuroscience. To this end, we re-visited a well-known and well-researched phonological process, namely, final devoicing in German. Final devoicing is a phonological rule that changes voiced consonants such as [d] into voiceless [t] if they occur in syllable-final position. We reasoned that native German speakers have tacit knowledge of this rule, such that ambiguous consonants in syllable-final position are automatically interpreted as voiceless. This process may require additional resources. Ambiguous consonants were constructed by parametrically changing acoustic cues to voicing in German, resulting in an 11-step continuum of consonants, reaching from transparently voiced to transparently voiceless members. Consonants embedded in syllable-initial and syllable-final position of pseudo-words were presented in an alternative forced-choice task during which participants had to identify the voiced or voiceless consonants. From this task, we selected the consonants with chance-level accuracy and used them as ambiguous consonants in a functional magnetic resonance imaging (fMRI) study. Therein, a 2 x 2 design of position (syllable-initial, syllable-final) and ambiguity (ambiguous, transparent) was employed. Results revealed a main effect of position in a typical auditory speech network, showing stronger BOLD signals in bilateral superior temporal gyri and sulci, left inferior frontal gyrus, left thalamus and left inferior parietal cortex for syllable-final compared to syllable-initial position. Next, we identified a main effect of ambiguity with stronger BOLD signals in left inferior frontal gyrus and left precentral cortex for ambiguous than for transparent consonants. Finally, there was an interaction of the effects of position and ambiguity in the pars triangularis of left inferior frontal gyrus (IFG). This interaction was driven by a stronger BOLD signal for ambiguous voicing in syllable-final position than in syllable-initial position. No such difference was seen for transparent voicing. We take this interaction to reflect the left IFG’s role in supporting the phonological rule application of final devoicing in German. This interpretation is corroborated by previous studies that established the IFG’s role for sequencing in phonological processes. We thus show that left IFG may indeed support proper phonological processes that go beyond mere syllable counting.

Topic Area: Phonology and Phonological Working Memory