Review
The role of alpha oscillations in temporal attention

https://doi.org/10.1016/j.brainresrev.2011.04.002Get rights and content

Abstract

Our brain does not process incoming sensory stimuli mechanistically. Instead the current brain state modulates our reaction to a stimulus. This modulation can be investigated by cognitive paradigms such as the attentional blink, which reveal that identical visual target stimuli are correctly reported only on about half the trials. Support for the notion that the fluctuating state of the brain determines stimulus detection comes from electrophysiological investigations of brain oscillations, which have shown that different parameters of ongoing oscillatory alpha activity (~ 10 Hz) can predict whether a visual stimulus will be perceived or not. The present article reviews recent findings on the role of prestimulus alpha oscillatory activity for visual perception and incorporates these results into a neurocognitive model that is able to account for various findings in temporal attention paradigms, specifically the attentional blink.

Research highlights

► Alpha oscillations indicate internally and externally oriented brain states. ► Different parameters of alpha oscillatory activity predict visual perception. ► A neuro-cognitive model is presented to explain the attentional blink.

Introduction

From time to time we all fail to perceive a salient sensory stimulus because the brain was occupied with some other cognitive process. For instance, we may not recognize an old friend passing by on our way to work, because we are thinking about a new friend who passed by just moments before. Such incidents show that our brain does not process incoming stimuli mechanistically; instead, whether or not a certain stimulus will be perceived partly depends on the brain's current state. Presumably, these brain states differ by whether the brain is biased towards processing stimuli from the external world (the old friend), or towards processing of internal representations (first friend passed). Identifying the neurophysiological signatures of such brain states, and predicting whether or not a sensory stimulus will be perceived, are two of the key goals in modern cognitive neuroscience. Brain oscillations have for a long time been recognized as indexing brain states. Several EEG/MEG studies point toward a fundamental role of ongoing oscillations for shaping perception and cognition. More specifically, a number of recent studies have revealed a significant correlation between the perception of a visual stimulus and different parameters of ongoing alpha activity prior to stimulus presentation1 (Ergenoglu et al., 2004, Hanslmayr et al., 2007, van Dijk et al., 2008, Busch et al., 2009, Busch and VanRullen, 2010, Mathewson et al., 2009). These findings suggest that brain oscillations indicate fluctuations between externally and internally oriented brain states, thus determining whether a visual stimulus will be perceived or not.

One of the key paradigms to study the effects of temporal attention is the attentional blink (AB) paradigm (Raymond et al., 1992). The basic finding of the AB is that the processing of a briefly presented stimulus (T1) leads to a significant decrease in the ability to correctly report a subsequent stimulus (T2) if it is presented ~ 250 ms after T1. Among others, a prominent hypothesis is that the AB arises because processing the first (external) stimulus (T1) forces the brain into an internal processing mode which renders the system unable to process the second stimulus (T2) to the level of conscious awareness (Shapiro et al., 1997; see Section 2.2 for details). The attentional blink paradigm is thus well suited for studying fluctuations between externally or internally oriented brain states. The goal of the current review is to integrate the findings of the aforementioned EEG/MEG studies into a neurocognitive framework that is able to explain temporal (i.e., sequentially occurring) attentional outcomes, such as revealed by the attentional blink paradigm.

We begin by providing a basic introduction into brain oscillations, with a specific focus on what different parameters of oscillatory activity tell us about the underlying neural processes. In section two, the attentional blink paradigm and its outcomes will be described, focussing on how current cognitive theories explain the AB. In section three, studies which examined the role of prestimulus alpha oscillatory activity for the perception of visual stimuli are reviewed. In the fourth section, the neurophysiological underpinnings of the different parameters of alpha oscillatory activity are considered. Finally, in section five a neurocognitive framework is presented that aims to integrate these findings into a coherent account of the AB and other phenomena of temporal attention.

Brain oscillations reflect rhythmic fluctuations of local field potentials between excitatory and inhibitory states of neural populations. Oscillations are generated by summated excitatory post-synaptic potentials (EPSPs) of several thousands of neurons (Lopes da Silva, 1991). The raw EEG/MEG signal contains various different brain oscillations, which can be analyzed by applying spectral analysis (e.g. Wavelet analysis; Fig. 1A). Oscillations are unambiguously characterized by three parameters: (i) frequency (Fig. 1B); (ii) amplitude (Fig. 1C); and (iii) phase (Fig. 1D).

Different brain networks are hypothesized to oscillate in different frequencies, with small networks oscillating in fast frequencies (> 40 Hz) and large networks oscillating in slower frequencies (< 20 Hz) (von Stein and Sarnthein, 2000, Buzsáki and Draguhn, 2004). Thereby, slower oscillations may represent brain networks of a higher hierarchy, encompassing multiple lower level networks, and thereby gating faster oscillations in a top-down manner (Lakatos et al., 2005, Lakatos et al., 2008). It is important to note that small and large networks do not refer to the Euclidian space covered by the neural network, but rather refer to the number of connections involved in a network. Thus, a fronto-parietal network with monosynaptic connections could be a small network, despite covering a distance of several centimeters.

The amplitude of MEG/EEG oscillations is determined by the total number of active neurons and their synchrony, and gives an estimate of how many EPSPs arrive in a given neural assembly at a certain time point (Varela et al., 2001). Therefore, stimulus-induced increases and decreases in oscillatory amplitude have been termed event-related synchronization and de-synchronization, respectively (Pfurtscheller and Aranibar, 1977; but see Hughes and Crunelli, 2007). Different frequencies, however, behave rather differently in response to sensory stimulation: alpha and beta (~ 15 Hz) oscillations typically decrease, whereas theta (~ 5 Hz) and gamma (~ 40 Hz) oscillations typically increase their amplitudes.

The phase of an oscillation specifies the current position in a given cycle and triggers the timing of neural signals by providing fine grained windows of excitation and inhibition. Evidence for this assumption comes from studies using intracranial recordings in primates and humans which show that the probability of a single neuron to fire is not random with regard to the local field potential, but is triggered by its phase (Lee et al., 2005, Jacobs et al., 2007). Using phase information, two different measures of neural activity can be calculated: (i) phase consistency across single trials, and (ii) phase-coupling between EEG/MEG sensors or brain regions. Although these two measures are mathematically closely related, they reflect totally different neural processes and should not be confused. Phase-consistency across single trials measures the variability in phase at a certain time point at one EEG/MEG sensor, e.g. the 10 Hz phase at stimulus onset.2 Low variability of phase across trials leads to high phase-consistency. Phase-coupling between EEG/MEG sensors or brain regions on the other hand, measures the variability of phase differences at a certain time point between two different recording sites. Phase-coupling has been interpreted as a measure of communication between distant neural assemblies (Fries, 2005, Varela et al., 2001), and evidence for this assumption has been reported by a recent intracranial study in primates (Canolty et al., 2010). The phase-locking value (PLV) is typically used to calculate phase-coupling (Lachaux et al., 1999).

Concerning the relationship between ongoing alpha activity and the perception of a sensory stimulus three different measures have been explored so far: (i) amplitude, (ii) phase-consistency at stimulus onset across single trials, and (iii) phase-coupling between distant brain areas. The results of these studies and their implications for understanding the attentional blink are reviewed following a description of this phenomenon.

Section snippets

The AB paradigm and its behavioral outcomes

The AB is one of the key approaches used to study the effects of temporal attention. The basic finding of the AB is that processing a briefly presented stimulus (T1) to the level of correct report leads to a dramatic decrease in the ability to correctly report a subsequent stimulus (T2), if T2 is presented between ~ 100 and 500 ms after T1 (Raymond et al., 1992, Shapiro et al., 1997).

Although there have been various approaches used to study the AB, the one taken by Raymond et al. (1992) depicts

Prestimulus alpha activity predicts visual perception

As mentioned above, three different parameters of prestimulus alpha oscillatory activity have been used successfully in several prior studies to predict the perception of a visual stimulus, (i) amplitude, (ii) phase at stimulus onset, and (iii) phase-coupling. The results of these studies are reviewed below.

Alpha amplitude and phase at stimulus onset are controlled by thalamo-cortical loops

One of the main stations of the visual pathway is the lateral geniculate nucleus in the thalamus, in which information from the retina is transferred to the visual cortex by thalamo-cortical neurons. Several electrophysiological studies in animals revealed that dynamics in thalamo-cortical loops generate alpha oscillations in posterior brain regions (Lopes da Silva et al., 1980, Hughes et al., 2004, Hughes and Crunelli, 2007). Moreover, thalamo-cortical alpha oscillations gate signal

Explaining the attentional blink by alpha oscillations

To provide an account of the attentional blink in terms of alpha oscillations we begin with the assumption that in the typical AB paradigm, the series of events preceding T2 presentation, induces a brain state which is quite unfavorable for visual perception. Note, however, that the first target is also preceded by some of these events but is still much better perceived than T2, a point to which we will return below. Specifically, we contend that the AB paradigm influences all three parameters

Conclusions

Integrating recent studies, which reveal a strong link between ongoing alpha activity and perceptual performance, the current review suggests that the AB outcome can well be predicted by alpha oscillations. In particular, the rapid serial visual presentation aspect of the AB paradigm as the brain struggles with the first target drives alpha amplitude, phase at stimulus onset, and phase-coupling toward an internally oriented state, which makes it difficult for briefly presented stimuli to be

Acknowledgments

The authors would like to thank Nathan Weisz and Maria Wimber for their very insightful comments on previous versions of the manuscript. Additionally, the authors thank Maria Wimber for contributing the data presented in Fig. 5.

References (83)

  • P. Jolicoeur et al.

    The demonstration of short-term consolidation

    Cogn. Psychol.

    (1998)
  • W. Klimesch et al.

    EEG alpha oscillations: the inhibition-timing hypothesis

    Brain Res. Rev.

    (2007)
  • C. Kranczioch et al.

    Temporal dynamics of access to consciousness in the attentional blink

    Neuroimage

    (2007)
  • H. Lee et al.

    Phase locking of single neuron activity to theta oscillations during working memory in monkey extrastriate visual cortex

    Neuron

    (2005)
  • F.H. Lopes da Silva et al.

    Relative contributions of intracortical and thalamo-cortical processes in the generation of alpha rhythms, revealed by partial coherence analysis

    Electroencephalogr. Clin. Neurophysiol.

    (1980)
  • F. Lopes da Silva

    Neural mechanisms underlying brain waves: from neural membranes to networks

    Electroencephalogr. Clin. Neurophysiol.

    (1991)
  • M.L. Lorincz et al.

    Temporal framing of thalamic relay-mode firing by phasic inhibition during the alpha rhythm

    Neuron

    (2009)
  • R. Marois et al.

    Neural correlates of the attentional blink

    Neuron

    (2000)
  • R. Marois et al.

    The neural fate of consciously perceived and missed events in the aftentional blink

    Neuron

    (2004)
  • K.E. Mathewson et al.

    Rescuing stimuli from invisibility: inducing a momentary release from visual masking with pre-target entrainment

    Cognition

    (2010)
  • M. Moosmann et al.

    Correlates of alpha rhythm in functional magnetic resonance imaging and near infrared spectroscopy

    Neuroimage

    (2003)
  • G. Pfurtscheller et al.

    Event-related cortical desynchronization detected by power measurements of scalp EEG

    Electroencephalogr. Clin. Neurophysiol.

    (1977)
  • K.L. Shapiro et al.

    The attentional blink

    Trends Cogn. Sci.

    (1997)
  • A. von Stein et al.

    Different frequencies for different scales of cortical integration: from local gamma to long range alpha/theta synchronization

    Int. J. Psychophysiol.

    (2000)
  • F.-B. Vialatte et al.

    Steady-state visually evoked potentials: focus on essential paradigms and future perspectives

    Prog. Neurobiol.

    (2010)
  • R. Ward et al.

    The slow time-course of visual attention

    Cogn. Psychol.

    (1996)
  • H. Bowman et al.

    The simultaneous type, serial token model of temporal attention and working memory

    Psychol. Rev.

    (2007)
  • N.A. Busch et al.

    The phase of ongoing EEG oscillations predicts visual perception

    J. Neurosci.

    (2009)
  • N.A. Busch et al.

    Spontaneous EEG oscillations reveal periodic sampling of visual attention

    Proc. Natl. Acad. Sci. U. S. A.

    (2010)
  • G. Buzsáki et al.

    Neuronal oscillations in cortical networks

    Science

    (2004)
  • E. Callaway et al.

    Relationship between reaction time and electroencephalographic alpha phase

    Science

    (1960)
  • R.T. Canolty et al.

    Oscillatory phase coupling coordinates anatomically dispersed functional cell assemblies

    Proc. Natl. Acad. Sci. U. S. A.

    (2010)
  • M.M. Chun et al.

    A two-stage model for multiple target detection in rapid serial visual presentation

    J. Exp. Psychol. Hum. Percept. Perform.

    (1995)
  • S. Dehaene et al.

    A neuronal network model linking subjective reports and objective physiological data during conscious perception

    Proc. Natl. Acad. Sci. U. S. A.

    (2003)
  • V. Di Lollo et al.

    The attentional blink: resource depletion or temporary loss of control?

    Psychol. Res.

    (2005)
  • J. Duncan et al.

    Direct measurement of attentional dwell time in human vision

    Nature

    (1994)
  • R.I. Goldman et al.

    Simultaneous EEG and fMRI of the alpha rhythm

    Neuroreport

    (2002)
  • W. Gruber et al.

    Alpha phase synchronization predicts P1 and N1 latency and amplitude size

    Cereb. Cortex

    (2005)
  • J. Gross et al.

    Dynamic imaging of coherent sources: studying neural interactions in the human brain

    Proc. Natl. Acad. Sci. U. S. A.

    (2001)
  • J. Gross et al.

    Modulation of long-range neural synchrony reflects temporal limitations of visual attention in humans

    Proc. Natl. Acad. Sci. U. S. A.

    (2004)
  • J. Gross et al.

    Anticipatory control of long-range phase synchronization

    Eur. J. Neurosci.

    (2006)
  • Cited by (268)

    View all citing articles on Scopus
    View full text