Elsevier

Hearing Research

Volume 265, Issues 1–2, 14 June 2010, Pages 30-37
Hearing Research

Research paper
Active stream segregation specifically involves the left human auditory cortex

https://doi.org/10.1016/j.heares.2010.03.005Get rights and content

Abstract

An important aspect of auditory scene analysis is the sequential grouping of similar sounds into one “auditory stream” while keeping competing streams separate. In the present low-noise fMRI study we presented sequences of alternating high-pitch (A) and low-pitch (B) complex harmonic tones using acoustic parameters that allow the perception of either two separate streams or one alternating stream. However, the subjects were instructed to actively and continuously segregate the A from the B stream. This was controlled by the additional instruction to listen for rare level deviants only in the low-pitch stream.

Compared to the control condition in which only one non-separable stream was presented the active segregation of the A from the B stream led to a selective increase of activation in the left auditory cortex (AC). Together with a similar finding from a previous study using a different acoustic cue for streaming, namely timbre, this suggests that the left auditory cortex plays a dominant role in active sequential stream segregation. However, we found cue differences within the left AC: Whereas in the posterior areas, including the planum temporale, activation increased for both acoustic cues, the anterior areas, including Heschl’s gyrus, are only involved in stream segregation based on pitch.

Introduction

In recent years the search for the neural mechanisms underlying “auditory scene analysis” (Bregman, 1990) has become an important topic in neuroscience (for reviews see Carlyon, 2004, Denham and Winkler, 2006, Micheyl et al., 2007, Snyder and Alain, 2007). The basic question is how the auditory system segregates a mixture of competing acoustic sequences into distinct meaningful auditory objects or “auditory streams” referring to certain sound sources (Bregman, 1990). Experimentally, sequential stream segregation is mostly studied by using sequences of repeating tones presented in an ABA_ or ABAB design. In most studies, A and B pure tones with differing frequencies are used. Depending on the frequency difference (ΔF) and the presentation rate of A and B tones, three different perceptual domains can be distinguished by determining the fission and coherence boundary (van Noorden, 1975). Below the fission boundary, only a single alternating stream is perceived; beyond the coherence boundary, two segregated streams are perceived; and between these boundaries, both of these percepts are possible (ambiguous domain).

Results from electrophysiological recording studies in animals have led to a hypothesis on the neuronal mechanism that explains the percept of two separate streams (Bee and Klump, 2004, Bee and Klump, 2005, Fishman et al., 2001, Fishman et al., 2004, Kanwal et al., 2003, Micheyl et al., 2005). It is suggested that frequency selectivity of tonotopically organized neurons in primary auditory cortex fields in combination with physiological forward suppression leads to separate representations of A and B tones. The differential suppression of non-best frequency tones was suggested to be the neuronal basis for the percept of two separate A and B streams. However, recent studies have revealed that the subcortical auditory pathway also plays a role in auditory stream formation (Kondo and Kashino, 2009, Pressnitzer et al., 2008) and besides tonotopic neural response separation, temporal coherence may have an influence as well (Elhilali et al., 2009).

Within the ambiguous perceptual domain of streaming, the perceptual organization depends on the attentional set of the subject and the instructions given as first described by van Noorden (1975) (for review see Moore and Gockel, 2002). This strongly suggests additional top–down mechanisms involved in stream segregation in the ambiguous domain that are independent of the physical stimulus.

A number of human imaging studies investigated stream segregation, including perceptually ambiguous tone sequences: EEG and MEG studies using mismatch negativity (MMN) paradigms introduced deviants into sequences of ABAB tones which were easier to detect when the perceptual organization of two separate streams was present. However, MMN responses to the deviants were only observed when the subjects were cued to perceive two separate streams either by explicit instruction (Sussman et al., 1998), by initial priming for a segregated organization (Sussman and Steinschneider, 2006), or by visual stimuli synchronized to the tones of one separate stream (Rahne et al., 2007, Rahne et al., 2008).

Further studies systematically varied the frequency separation between the pure tones of ABA_ sequences and found increases of auditory evoked potentials/fields to the B tones corresponding to larger frequency separations between A and B pure tones (P1–N1–P2 and N1c in Snyder et al. (2006) and P1m–N1m in Gutschalk et al. (2005, first experiment)). This enhancement also correlated with behavioral reports of stream segregation. Similarly, Gutschalk et al. (2007) described an increase of the P1m component for the A tone with increasing separation in fundamental frequencies (f0) between harmonic complex tones in repeating ABBB_ sequences where A and B denote tones with different f0. In the second experiment of the earlier MEG study, Gutschalk et al. (2005) directly compared magnetic fields evoked by B tones of a certain perceptually ambiguous ABA_ sequence when the subjects perceived two separate streams and when the subjects perceived one integrated stream. They found a larger N1m amplitude during the perceptual organization of two separate streams both when the subjects followed the A or B tones and a larger P1 amplitude when the subjects followed the B tones. Gutschalk et al., 2005, Gutschalk et al., 2007 and Snyder et al. (2006) suggested that reduced forward suppression may be the neuronal mechanism underlying the enhanced long latency components/fields during the perception of two segregated streams. The dipoles of these components/fields were located in the non-primary auditory cortex.

This extends findings from electrophysiological recordings in animals and suggests additional stream segregation mechanisms beyond the primary auditory cortex.

An involvement of non-primary auditory cortex areas in stream segregation was also shown in two fMRI studies (Gutschalk et al., 2007, Wilson et al., 2007). Similar to the approach by Snyder et al. (2006) and Gutschalk et al., 2005, Gutschalk et al., 2007, first experiment) they used different frequency/f0 separations between A and B tones in repeating ABBB_ and ABAB sequences, respectively, and also found increasing fMRI activation in Heschl’s gyrus and the planum temporale with increasing frequency/f0 separation between the A and B tones.

In an fMRI study Cusack (2005) used ambiguous sequences of ABA_ tones and compared the activation when the subjects heard two separate streams or one alternating stream. Cusack did not find activation differences in the auditory cortex, but a difference in the right intraparietal sulcus. In a recent fMRI study (Rahne et al., 2008), also no differences in auditory cortex activation were observe when using an ambiguous sound sequence (ABAB) whose perception was differentially biased towards either an integrated or segregated percept by appropriate visual stimuli. A third fMRI study employing an ambiguous sound sequence, used timbre as acoustic parameter for segregation of ABAB sounds produced by an organ or a trumpet (Deike et al., 2004). They instructed the subjects to segregate the A- and B-timbre streams by continuously following the sound of a specific timbre. This led to an increase of fMRI activation in non-primary auditory cortex of the left hemisphere compared to a non-separable control stream.

In summary, a number of imaging studies suggest that additional mechanisms are involved beyond the one suggested for stream segregation based on pitch first put forward by Fishman (Fishman et al., 2001, Fishman et al., 2004). The first reason is that in contrast to the electrophysiological recordings in animals, that were restricted to the primary auditory cortex, the above mentioned human imaging studies showed an involvement of secondary auditory areas. Secondly, when using the same ambiguous tone sequence, additional mechanisms need to be involved to evoke one or the other perceptual organization and this may be reflected in the activation of auditory cortex. As this is independent of the physical stimulus properties we refer to such mechanisms as “top–down”. In previous streaming studies such top–down influences on the activation in auditory cortex were initialized by instructing the subjects to select one or the other perceptual organization either by priming (Sussman and Steinschneider, 2006), synchronized visual stimuli (Rahne et al., 2007) or explicit instructions (Deike et al., 2004, Sussman et al., 1998).

The aim of the current study was to investigate top–down mechanisms that are required to maintain a segregated percept of an ambiguous tone sequence instead of freely switching between the two possible perceptual organizations. Therefore, we instructed our subjects to actively and continuously segregate auditory streams on the basis of pitch differences. Our second goal was to localize and differentiate these task-specific effects within the auditory territories previously defined by Brechmann et al. (2002) and beyond that in more detail. The third aim was to clarify the specific role of the left hemisphere in stream segregation that was suggested in the study by Deike et al. (2004) using timbre as the acoustic parameter for segregation.

Section snippets

Subjects

Twenty right-handed (Edinburgh Handedness Inventory) normal hearing subjects (seven male, 13 female, from 20 to 35 years old) participated in the fMRI study, 11 in experiment I and nine in experiment II. All participants showed a language laterality toward the left hemisphere tested as described in Bethmann et al. (2006). Six participants from experiment II and 11 additional listeners took part in psychophysical measurements. The subjects gave written informed consent to the study which was

Task performance

The repeated-measure ANOVA of sensitivity index d′ revealed a significant main effect of the within-subject factor perceptual organization (segregation, control) with a lower d′ in the segregation condition compared to both control conditions [F(1, 16) = 9.5, (p = 0.007)]. Furthermore, the ANOVA revealed a significantly higher overall task performance in experiment II using the same physical presentation rate in the segregation and control condition than in experiment I using the same perceptual

Selective involvement of the left auditory cortex

The main result of this study is the selective involvement of the left auditory cortex when the subjects had to actively and continuously segregate the A from the B stream, using pitch differences as an acoustic cue. From a behavioral point of view, we first showed that ABAB tone sequences were perceptually ambiguous. Second, we made sure that the subjects constantly segregated the A from the B stream by introducing the task of detecting small changes in stimulus level.

A left auditory cortex

Funding

This work was supported by the “Deutsche Forschungsgemeinschaft” [SFB/TRR31].

Acknowledgments

We thank Antje Schasse for her statistical advice and Monika Dobrowolny for her technical assistance.

References (70)

  • C. Micheyl et al.

    The role of auditory cortex in the formation of auditory streams

    J. Res.

    (2007)
  • P. Morosan et al.

    Human primary auditory cortex: cytoarchitectonic subdivisions and mapping into a spatial reference system

    Neuroimage

    (2001)
  • R.D. Patterson et al.

    The processing of temporal pitch and melody information in auditory cortex

    Neuron

    (2002)
  • D. Pressnitzer et al.

    Perceptual organization of sound begins in the auditory periphery

    Curr. Biol.

    (2008)
  • T. Rahne et al.

    Visual cues can modulate integration and segregation of objects in auditory scene analysis

    Brain Res.

    (2007)
  • T. Rahne et al.

    A multilevel and cross-modal approach towards neuronal mechanisms of auditory streaming

    Brain Res.

    (2008)
  • F. Rivier et al.

    Cytochrome oxidase, acetylcholinesterase, and NADPH-diaphorase staining in human supratemporal and insular cortex: evidence for multiple auditory areas

    Neuroimage

    (1997)
  • E. Sussman et al.

    Neurophysiological evidence for context-dependent encoding of sensory input in human auditory cortex

    Brain Res.

    (2006)
  • E. Sussman et al.

    Attention affects the organization of auditory input associated with the mismatch negativity system

    Brain Res.

    (1998)
  • J.R. Tregellas et al.

    Effect of task difficulty on the functional anatomy of temporal processing

    Neuroimage

    (2006)
  • R.J. Zatorre et al.

    Structure and function of auditory cortex: music and speech

    Trends Cogn. Sci.

    (2002)
  • F. Baumgart et al.

    Electrodynamic headphones and woofers for application in magnetic resonance imaging scanners

    Med. Phys.

    (1998)
  • M.A. Bee et al.

    Primitive auditory stream segregation: a neurophysiological study in the songbird forebrain

    J. Neurophysiol.

    (2004)
  • M.A. Bee et al.

    Auditory stream segregation in the songbird forebrain: effects of time intervals on responses to interleaved tone sequences

    Brain Behav. Evol.

    (2005)
  • N. Behne et al.

    Contralateral white noise selectively changes right human auditory cortex activity caused by a FM-direction task

    J. Neurophysiol.

    (2005)
  • D. Bendor et al.

    The neuronal representation of pitch in primate auditory cortex

    Nature

    (2005)
  • A. Bethmann et al.

    Determining language laterality by fMRI and dichotic listening

    Brain Res.

    (2006)
  • J.R. Binder et al.

    Function of the left planum temporale in auditory and linguistic processing

    Brain

    (1996)
  • J.R. Binder et al.

    Neural correlates of sensory and decision processes in auditory object identification

    Nat. Neurosci.

    (2004)
  • A. Brechmann et al.

    Hemispheric shifts of sound representation in auditory cortex with conceptual listening

    Cereb. Cortex

    (2005)
  • A. Brechmann et al.

    Sound-level-dependent representation of frequency modulations in human auditory cortex: a low-noise fMRI study

    J. Neurophysiol.

    (2002)
  • A. Brechmann et al.

    Working memory specific activity in auditory cortex: potential correlates of sequential processing and maintenance

    Cereb. Cortex

    (2007)
  • A.S. Bregman

    Auditory Scene Analysis. The Perceptual Organization of Sound

    (1990)
  • R. Cusack

    The intraparietal sulcus and perceptual organization

    J. Cogn. Neurosci.

    (2005)
  • S. Deike et al.

    Auditory stream segregation relying on timbre involves left auditory cortex

    Neuroreport

    (2004)
  • Cited by (0)

    View full text