Elsevier

Brain Research

Volume 1139, 30 March 2007, Pages 153-162
Brain Research

Research Report
Prefrontal gamma-band activity distinguishes between sound durations

https://doi.org/10.1016/j.brainres.2006.12.085Get rights and content

Abstract

The present study used magnetoencephalography to assess the cortical representation of brief sound durations during a short-term memory task. Twelve subjects were instructed to memorize sounds S1 with durations of either 100 or 200 ms during an 800-ms delay phase. Subsequently, they had to judge whether the duration of a probe sound S2 matched the memorized stimulus. Statistical probability mapping of oscillatory signals revealed several components of gamma-band activity (GBA) over prefrontal cortex. A first component with a center frequency of 40 Hz responded more strongly to longer than shorter sounds during the encoding of S1. During the subsequent delay phase, shorter and longer durations were associated with topographically and spectrally distinct GBA enhancements at 71 and 80 Hz, respectively. S2 was again associated with stronger oscillatory activation for longer than shorter sounds at ∼ 72 Hz. Nonmatching compared with matching S1–S2 pairs elicited an additional ∼ 66 Hz GBA component peaking at ∼ 200 ms after the offset of S2. The analysis of magnetoencephalographic GBA thus served to identify prefrontal network components underlying the representation of different sound durations during the various phases of a delayed matching-to-sample task.

Introduction

Studies using intracortical recordings in animals have proposed oscillatory synchronization in the gamma-band range (∼ 30–100 Hz) as a mechanism underlying perceptual binding (Fries et al., 1997, Gray et al., 1989, Singer et al., 1997). Early electroencephalography (EEG) work in humans has supported this notion by showing increased gamma-band activity (GBA) during the perception of regularly moving bars (Lutzenberger et al., 1995, Müller et al., 1996) and gestalt-like (Kaiser et al., 2004, Tallon-Baudry et al., 1996) or meaningful objects (Keil et al., 1999, Pulvermüller et al., 1996, Rodriguez et al., 1999). Moreover, recent research has implicated fast oscillatory signals in a variety of cognitive processes including selective attention (Bauer et al., 2006, Fries et al., 2001, Gruber et al., 1999, Herrmann et al., 1999, Müller and Keil, 2004), memory (Axmacher et al., 2006, Gruber et al., 2004b, Herrmann et al., 2004c, Osipova et al., 2006), and decision-making (Jung-Beeman et al., 2004). These findings suggest a more general role of oscillatory synchronization for cognitive processes (Kaiser and Lutzenberger, 2005a, Kaiser and Lutzenberger, 2005b, Keil et al., 2001). Most of these functions have been related to induced, nonphase-locked GBA with typical peak latencies of ∼ 200–300 ms post stimulus onset (Tallon-Baudry and Bertrand, 1999), whereas evoked, phase-locked gamma responses may reflect both sensory processing (Busch et al., 2004, Karaka° and Ba°ar, 1998) and fast memory matches (Herrmann et al., 2004a). An overview of different types of gamma responses has been provided by Başar-Eroğlu et al. (1996).

At present, most researchers would agree that human GBA recorded with EEG or magnetoencephalography (MEG) may be interpreted as a correlate of the activation of cortical object representations. Evidence for GBA during the internal maintenance of an object representation in the absence of external stimulation has been yielded first in visual delayed matching-to-sample tasks where 25–45 Hz activity was increased at occipital and bilateral temporal electrodes compared with a nonmemory control condition (Tallon-Baudry et al., 1998, Tallon-Baudry et al., 1999). Furthermore, familiar pictures gave rise to stronger gamma responses than novel images, probably reflecting the activation of an existing memory trace (Gruber and Müller, 2005, Herrmann et al., 2004c). This gamma response is diminished to repeated presentations, suggesting a “sharpening” of the activated network (Gruber et al., 2004a, Gruber and Müller, 2002). Conversely, increased GBA with stimulus repetition of nonfamiliar objects may reflect the formation of new cortical representations (Fiebach et al., 2005, Gruber and Müller, 2005, Gruber and Müller, 2006). These EEG studies have usually yielded topographically widespread activations across a wide frequency range which may be attributable to the fact that averages across numerous stimuli were calculated, possibly resulting in the activation of a large number of different networks. However, so far there exists little evidence for GBA enhancements reflecting the activation of networks representing individual stimuli.

Here we used MEG with its high sensitivity to local activations to identify stimulus-specific oscillatory signals. Previous MEG studies using a statistical probability mapping have revealed highly local and spectrally narrow GBA enhancements in frequencies between ∼ 50 and 90 Hz exhibiting clear topographical differences, e.g., between auditory spatial versus nonspatial processing during passive change detection paradigms (Kaiser et al., 2000a, Kaiser et al., 2002a, Kaiser et al., 2002b). During delayed matching-to-sample tasks requiring the memorization of auditory spatial versus pattern-related information, enhanced GBA during the delay phase was found over regions of the putative auditory dorsal and ventral processing streams, respectively (Lutzenberger et al., 2002, Kaiser et al., 2003). This suggested that not the sounds as such were maintained, but their task-relevant acoustic features. GBA in response to the probe sound S2 was dependent on whether S1 and S2 were identical or not: a nonmatching S2 elicited an additional gamma response, possibly reflecting the activation of a further network in addition to the representation of S1 (Leiberg et al., 2006a). The present study used a short-term memory task where subjects had to memorize the duration of brief noise stimuli. Unlike previous studies where we averaged the responses across several different stimuli, here we employed only two S1 sounds to enable the direct comparison of oscillatory activity between two individual stimuli based on a sufficient number of trials. On the basis of previous work both in animals (Genovesio et al., 2006, Sakurai et al., 2004) and humans (Belin et al., 2002, Coull et al., 2004, Macar et al., 2002, Rao et al., 2001, Smith et al., 2003) we expected that representations of sound duration would involve premotor/prefrontal cortex. The present task required the maintenance of separate representations of the two durations of S1. We expected that different durations would be represented by different local networks which would be characterized by GBA enhancements at different frequencies and, possibly, with different topographies. The time courses of oscillatory activity in these local networks should reflect their involvement in the different stages of the working memory task, like stimulus encoding and maintenance and the comparison of sample and probe sounds.

Section snippets

Behavioral data

Subjects showed an overall performance of 76.5% hits (SD = 4.2) and 81.8% correct rejections (SD = 3.6). While correct response rates did not differ between short and long S2 stimuli, short S2 sounds elicited longer reaction times than longer stimuli (reaction time for short S2: 518.5 ms (SD = 36.3 ms) after the offset of S2; reaction time for long S2: 386.0 ms (SD = 39.6 ms)), F(1,11) = 5.7, p = 0.036. Hit rate and correct rejection rate for the matching trials were 82.5% (SD = 6.3) and 78.0% (SD = 5.2),

Discussion

The active memorization of brief sound durations and the subsequent comparison of sample and probe stimuli in a working memory task were associated with magnetoencephalographic GBA increases over frontal cortical areas. The comparison of shorter with longer sounds (100 versus 200 ms, respectively) yielded four GBA components with distinct spectral and temporal characteristics. During the encoding of S1, oscillatory activity at ∼ 40 Hz over the left fronto-temporal cortex differentiated longer

Subjects

Twelve healthy adults (seven females, five males, mean age 23.8 (SD = 0.4) years) gave their informed and written consent to participate in the study. The study was conducted in accordance with the Declaration of Helsinki.

Experimental procedure and stimulus materials

Subjects were seated upright in a magnetically shielded room (Vakuum-Schmelze, Hanau, Germany). They were instructed to sit still and keep their eyes open, looking at a fixation cross in the center of their visual field ∼2 m in front of them. Auditory stimuli were presented

Acknowledgment

This study was supported by Deutsche Forschungsgemeinschaft (DFG) grant SFB 550/C1.

References (74)

  • C.S. Herrmann et al.

    Gamma responses and ERPs in a visual classification task

    Clin. Neurophysiol.

    (1999)
  • C.S. Herrmann et al.

    Cognitive functions of gamma-band activity: memory match and utilization

    Trends Cogn. Sci.

    (2004)
  • R.B. Ivry et al.

    The neural representation of time

    Curr. Opin. Neurobiol.

    (2004)
  • L. Jäncke et al.

    Cortical activations during paced finger-tapping applying visual and auditory pacing stimuli

    Cogn. Brain Res.

    (2000)
  • J. Kaiser et al.

    Statistical probability mapping reveals high-frequency magnetoencephalographic activity in supplementary motor area during self-paced finger movements

    Neurosci. Lett.

    (2000)
  • J. Kaiser et al.

    Dynamics of gamma-band activity in human magnetoencephalogram during auditory pattern working memory

    NeuroImage

    (2003)
  • J. Kaiser et al.

    Magnetoencephalographic gamma-band responses to illusory triangles in humans

    NeuroImage

    (2004)
  • S. Karakaş et al.

    Early gamma response is sensory in origin: a conclusion based on cross-comparison of results from multiple experimental paradigms

    Int. J. Psychophysiol.

    (1998)
  • A. Keil et al.

    Functional correlates of macroscopic high-frequency brain activity in the human visual system

    Neurosci. Biobehav. Rev.

    (2001)
  • S. Leiberg et al.

    Gamma-band activity dissociates between matching and nonmatching stimulus pairs in an auditory delayed matching-to-sample task

    NeuroImage

    (2006)
  • S. Leiberg et al.

    Effects of memory load on cortical oscillatory activity during auditory pattern working memory

    Brain Res.

    (2006)
  • M.I. Leon et al.

    Representation of time by neurons in the posterior parietal cortex of the macaque

    Neuron

    (2003)
  • P.A. Lewis et al.

    Brain activation patterns during measurement of sub- and supra-second intervals

    Neuropsychologia

    (2003)
  • P.A. Lewis et al.

    Distinct systems for automatic and cognitively controlled time measurement: evidence from neuroimaging

    Curr. Opin. Neurobiol.

    (2003)
  • H. Luo et al.

    Discrimination and categorization of speech and non-speech sounds in an MEG delayed-match-to-sample study

    NeuroImage

    (2005)
  • W. Lutzenberger et al.

    Visual stimulation alters local 40-Hz responses in humans: an EEG-study

    Neurosci. Lett.

    (1995)
  • F. Pulvermüller et al.

    High-frequency cortical responses reflect lexical processing: an MEG study

    Electroencephalogr. Clin. Neurophysiol.

    (1996)
  • P. Sauseng et al.

    Theta coupling in the human electroencephalogram during a working memory task

    Neurosci. Lett.

    (2004)
  • C. Schmiedt et al.

    Event-related theta oscillations during working memory tasks in patients with schizophrenia and healthy controls

    Cogn. Brain Res.

    (2005)
  • R.I. Schubotz et al.

    Time perception and motor timing: a common cortical and subcortical basis revealed by fMRI

    NeuroImage

    (2000)
  • W. Singer et al.

    Neuronal assemblies: necessity, signature and detectability

    Trends Cogn. Sci.

    (1997)
  • A. Smith et al.

    A right hemispheric frontocerebellar network for time discrimination of several hundreds of milliseconds

    NeuroImage

    (2003)
  • C. Tallon-Baudry et al.

    Oscillatory gamma activity in humans and its role in object representation

    Trends Cogn. Sci.

    (1999)
  • M. Bauer et al.

    Tactile spatial attention enhances gamma-band activity in somatosensory cortex and reduces low-frequency activity in parieto-occipital areas

    J. Neurosci.

    (2006)
  • R.C. Blair et al.

    An alternative method for significance testing of waveform difference potentials

    Psychophysiology

    (1993)
  • A. Brand et al.

    Duration tuning in the mouse auditory midbrain

    J. Neurophysiol.

    (2000)
  • J.T. Coull et al.

    Functional anatomy of the attentional modulation of time estimation

    Science

    (2004)
  • Cited by (17)

    • Investigating the effects of different levels and types of construction noise on emotions using EEG data

      2022, Building and Environment
      Citation Excerpt :

      One possible factor which might have affected the results in this study, regarding the effects of different noise types and levels on the EEG data, is the exposure time to each noise condition. Duration of noise exposure is an important factor affecting emotions caused by the noise [45,109]. Since the duration for exposure to each noise condition was fixed at 3 min in this study, the participants might exhibit different reactions to different exposure times for the examined noise levels and types.

    • What works in auditory working memory? A neural oscillations perspective

      2016, Brain Research
      Citation Excerpt :

      Specifically, theta oscillations have been associated with the sequential aspects of memory content (Buzsáki, 2002; Hsieh et al., 2011; Lisman, 2010; Lisman and Jensen, 2013), much akin to the classic Sternberg paradigm (Sternberg, 1966). Gamma power, conceptualized to occur nested within slower cycles of theta, has been reported to increase during working memory maintenance of visual (Tallon-Baudry et al., 1998; Medendorp et al., 2007; Jokisch and Jensen, 2007; Roux and Uhlhaas, 2014), auditory (Lutzenberger et al., 2002; Kaiser et al., 2007b, 2009), and somatosensory (Haegens et al., 2010) stimuli, presumably reflecting active maintenance of working memory representations. We will now focus in more detail on recent evidence on how alpha power can reflect memory load in auditory working memory.

    • Frequency tuning for temporal perception and prediction

      2016, Current Opinion in Behavioral Sciences
    • Human gamma-band activity and behavior

      2011, International Journal of Psychophysiology
      Citation Excerpt :

      The amplitude difference between non-match and match trials was higher for good than poor task performers, suggesting that the decision about the match between the test sound and the memory representation of the sample stimulus involved the activation of networks coding differences between both events (Fig. 1c). This result was further corroborated by reanalyzing the data from two previous short-term memory studies (Kaiser et al., 2007; Leiberg et al., 2006). A recent MEG study has demonstrated that both resource allocation, reflected by an increase of induced GBA in prefrontal regions, and disengagement of task-irrelevant cortical regions involving an increase of posterior alpha activity, is related to task performance (Haegens et al., 2010).

    View all citing articles on Scopus
    View full text