Research reportSpatial representations of temporal and spectral sound cues in human auditory cortex
Introduction
Natural and behaviorally relevant sounds are often characterized by prominent amplitude modulations of their temporal envelope. Amplitude modulation in speech, for example, provides linguistic information and modulation rates up to 50 Hz often serve as cues to segment speech tokens (Giraud and Poeppel, 2012; Luo and Poeppel, 2007; Rosen, 1992). Slow envelope modulations are not only critical for speech understanding (Elliott and Theunissen, 2009) but speech recognition solely based on slow temporal envelope cues is surprising good (Drullman et al., 1994; Shannon et al., 1995). In music, individual notes (complex periodic sounds played by an instrument) occur mostly at rates of a few Hertz, while faster modulation rates (>40 Hz) characterize the pitch of individual instrumental sounds (Plack et al., 2005). Importantly, regular temporal envelope modulations are not confined to human communication signals, but are a characteristic feature of animal vocalizations and many other natural sounds (Chandrasekaran et al., 2009; Joris et al., 2004). In general, slow modulation rates of a few Hertz are perceived as individual events, with rates of a few tens of Hertz producing a gradually more blended percept known as acoustic flutter (Bendor and Wang, 2007; Miller and Taylor, 1948). Even faster modulation rates above approximately 40 Hz result in a continuous percept that is usually associated with a specific perceived pitch (Pressnitzer et al., 2001) and contribute to distinguishing speaker identities, emotional states or different environmental sounds (Latinus and Belin, 2011; Rosen, 1992; Singh and Theunissen, 2003). Despite the importance of these slow temporal envelope modulations in natural sounds for carrying behaviorally relevant information, it remains unresolved how auditory cortex represents these. Here, we emphasize on the cortical representation of sound envelope modulation in a range between 2 and 32 Hz that is crucial for communication but which does not induce a specific pitch percept.
In many auditory structures spectral sound features are represented as spatial (tonotopic) maps, whereby the neural organization of preferred sound frequency reflects the ordered frequency representation in the cochlea (Formisano et al., 2003; Petkov et al., 2006). These topographically ordered functional maps of spectral cues are observed consistently across species, and are used for the functional parcellation of auditory cortices into individual fields (Formisano et al., 2003; Merzenich and Brugge, 1973; Merzenich et al., 1975; Petkov et al., 2006; Recanzone et al., 2000). In general, maps representing features from the sensory environment are a common organizational principle across auditory, visual and sensory cortices (Formisano et al., 2003; Knudsen et al., 1987; Penfield and Boldrey, 1937; Sereno et al., 1995; Yu et al., 2005). These maps feature systematic spatial distributions of sensory information within a cortical area, providing computational advantages in sensory information processing (Knudsen et al., 1987). For the auditory system, various maps featuring different behaviorally relevant aspects of sound analysis have been observed [for review see Schreiner and Winer (2007)]. Given the behavioral importance of temporal sound cues several authors have argued for a similar spatially ordered representation of temporal modulation rates in auditory cortex (Moore, 2003; Schreiner and Winer, 2007). Indeed, animal studies have provided evidence for spatial encoding of temporal envelope cues in the auditory midbrain of the monkey (Baumann et al., 2011). However, other studies found no evidence for a consistent spatial representation of temporal sound cues in auditory cortex (Bendor and Wang, 2010; Nelken et al., 2008).
Despite clear evidence for selectivity to temporal sound features in individual neurons (Bendor and Wang, 2007, 2010; Imaizumi et al., 2011), it remains unclear whether in humans (i) auditory cortices provide spatially distinct representations of this neural selectivity of temporal sound cues, and (ii) how this presumed spatial representation of temporal sound features relates to auditory fields as defined either by anatomy or tonotopic maps. Here we combined an functional magnetic resonance imaging (fMRI) sequence optimized for imaging the auditory system (Seifritz et al., 2006) with a stimulation protocol commonly used to reveal topographical cortical sensory representations (Engel et al., 1994) to test for topographically ordered representations of temporal envelope modulation in human auditory cortex, and to compare these maps with those representing spectral sound features (i.e., tonotopic maps) within the same subjects.
Section snippets
Subjects
The experiments were approved by the joint ethics committee of the University Clinic and the Max Planck Institute, Tübingen, Germany, and all subjects provided written informed consent to participate in this study. Six adult subjects (three males, three females, ages 23–36; five right- and one left-handed) participated in the study. No subject had a history of neurological or hearing disorders. During data acquisition subjects were asked to keep eyes open and to listen to the sounds.
Data acquisition
fMRI data
Spatial representation of sound frequency
We first determined the representation of sound frequency to identify the tonotopic organization of auditory cortices within our subjects (Exp. 1). We presented sweeps of pulsed sine-wave tones with stepwise increasing spectral frequency (500–8000 Hz; Exp. 1, Figs. 1 and 2) and identified sites preferentially and significantly (FDR p < .01) activated by particular frequencies using a cortex-based alignment technique projected relative to the cytoarchitectonic organization around Heschl's gyrus
The representation of temporal sound cues in auditory cortex
Our results demonstrate a distinct spatial representation of temporal sound modulation rates in human PAC that was consistently revealed across subjects and hemispheres. This representation of temporal sound cues exhibits a topographical organization alongside HG that is orthogonal to the representation of spectral sound frequency (tonotopy). This finding propels speculations as to the use of a place coding or place preference for temporal sound features in auditory cortex, a hypothesis that is
Acknowledgments
This work was supported by the Swiss National Science Foundation and the ‘Schweizerische Stiftung für medizinisch-biologische Stipendien’ (Grant PASMP3-123222, M.H.) and the Max Planck Society.
References (71)
- et al.
Volumetric vs. surface-based alignment for localization of auditory cortex activation
NeuroImage
(2005) - et al.
A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data
NeuroImage
(2005) - et al.
Mirror-symmetric tonotopic maps in human primary auditory cortex
Neuron
(2003) - et al.
Endogenous cortical rhythms determine cerebral specialization for speech perception and production
Neuron
(2007) - et al.
Tonotopic organization of human auditory cortex
NeuroImage
(2010) - et al.
Spatial organization of repetition rate processing in cat anterior auditory field
Hearing Research
(2011) - et al.
The spectrotemporal filter mechanism of auditory selective attention
Neuron
(2013) - et al.
Human voice perception
Current Biology
(2011) - et al.
Phase patterns of neuronal responses reliably discriminate speech in human auditory cortex
Neuron
(2007) - et al.
Representation of the cochlear partition of the superior temporal plane of the macaque monkey
Brain Research
(1973)
Human primary auditory cortex: Cytoarchitectonic subdivisions and mapping into a spatial reference system
NeuroImage
Sensory neural codes using multiplexed temporal scales
Trends in Neurosciences
Optimizing the imaging of the monkey auditory cortex: Sparse vs. continuous fMRI
Magnetic Resonance Imaging
Probabilistic mapping and volume measurement of human primary auditory cortex
NeuroImage
Is it tonotopy after all?
NeuroImage
Representation of amplitude modulation in the auditory cortex of the cat. II. Comparison between cortical fields
Hearing Research
Auditory cortex mapmaking: Principles, projections, and plasticity
Neuron
Low-frequency neuronal oscillations as instruments of sensory selection
Trends in Neurosciences
Gamma band pitch responses in human auditory cortex measured with magnetoencephalography
NeuroImage
Enhancing BOLD response in the auditory system by neurophysiologically tuned fMRI sequence
NeuroImage
The coordinated mapping of visual space and response features in visual cortex
Neuron
Orthogonal acoustic dimensions define auditory field maps in human cortex
Proceedings of the National Academy of Sciences of the United States of America
Orthogonal representation of sound dimensions in the primate midbrain
Nature Neuroscience
Differential neural coding of acoustic flutter within primate auditory cortex
Nature Neuroscience
Neural coding of periodicity in marmoset auditory cortex
Journal of Neurophysiology
Hierarchical and asymmetric temporal sensitivity in human auditory cortices
Nature Neuroscience
Auditory Scene Analysis: The Perceptual Organization Of Sound
Evidence for pitch chroma mapping in human auditory cortex
Cerebral Cortex
The natural statistics of audiovisual speech
PLoS Computational Biology
Functional correlates of the anterolateral processing hierarchy in human auditory cortex
Journal of Neuroscience
Compartments within human primary auditory cortex: Evidence from cytochrome oxidase and acetylcholinesterase staining
European Journal of Neuroscience
Human primary auditory cortex follows the shape of Heschl's gyrus
Journal of Neuroscience
In vivo functional and myeloarchitectonic mapping of human primary auditory areas
Journal of Neuroscience
Effect of temporal envelope smearing on speech reception
Journal of the Acoustical Society of America
The modulation transfer function for speech intelligibility
PLoS Computational Biology
Cited by (45)
Using high spatial resolution fMRI to understand representation in the auditory network
2021, Progress in NeurobiologyFunctional characterization of human Heschl's gyrus in response to natural speech
2021, NeuroImageCitation Excerpt :Several previous studies have shown an encoding of temporal modulations in the human auditory system (Herdener et al., 2013; Leaver and Rauschecker, 2016; Overath et al., 2012; Schönwiesner and Zatorre, 2009; Wang et al., 2011). Studies that used ripple stimuli showed that the preferred temporal rate was highest in medial HG (Herdener et al., 2013; Leaver and Rauschecker, 2016; Overath et al., 2012; Schönwiesner and Zatorre, 2009). These studies, however, did not provide a precise spatial organization of temporal modulation in HG.
Cortical voice processing is grounded in elementary sound analyses for vocalization relevant sound patterns
2021, Progress in Neurobiology2.34 - Coding of Spectral Information
2020, The Senses: A Comprehensive Reference: Volume 1-7, Second Edition