Elsevier

Progress in Neurobiology

Volume 55, Issue 4, 1 July 1998, Pages 343-361
Progress in Neurobiology

Neural correlates of attention and arousal: insights from electrophysiology, functional neuroimaging and psychopharmacology

https://doi.org/10.1016/S0301-0082(98)00011-2Get rights and content

Abstract

Attention and arousal are multi-dimensional psychological processes, which interact closely with one another. The neural substrates of attention, as well as the interaction between arousal and attention, are discussed in this review. After a brief discussion of psychological and neuropsychological theories of attention, event-related potential correlates of attention are discussed. Essentially, attention acts to modulate stimulus-induced electrical potentials (N100/P100, P300, N400), rather than generating any unique potentials of its own. Functional neuroimaging studies of attentional orienting, selective attention, divided attention and sustained attention (and its inter-dependance on underlying levels of arousal) are then reviewed.

A distinction is drawn between the brain areas which are crucially involved in the top-down modulation of attention (the ‘sources’ of attention) and those sensory-association areas whose activity is modulated by attention (the ‘sites’ of attentional expression). Frontal and parietal (usually right-lateralised) cortices and thalamus are most often associated with the source of attentional modulation. Also, the use of functional neuroimaging to test explicit hypotheses about psychological theories of attention is emphasised. These experimental paradigms form the basis for a ‘new generation’ of functional imaging studies which exploit the dynamic aspect of imaging and demonstrate how it can be used as more than just a ‘brain mapping’ device.

Finally, a review of psychopharmacological studies in healthy human volunteers outlines the contributions of the noradrenergic, cholinergic and dopaminergic neurotransmitter systems to the neurochemical modulation of human attention and arousal. While, noradrenergic and cholinergic systemsare involved in ‘low-level’ aspects of attention (e.g. attentional orienting), the dopaminergic system isassociated with more ‘executive’ aspects of attention such as attentional set-shifting or working memory.

Introduction

Although precise definitions of attention and arousal are somewhat lacking, it is generally accepted that both are heterogeneous processes. Attention, however may be thought of in the simplest terms as the appropriate allocation of processing resources to relevant stimuli. Over 100 years ago William James (1890)stated that “Everyone knows what attention is. It is the taking possession by the mind in clear and vivid form of one out of what seem several simultaneous objects or trains of thought”. This quote provides a prologue to many reviews on attention, most probably because it captures the very essence of attention without recourse to scientific jargon. However, attention is now thought to comprise several sub-processes, and one can ‘attend’ to a stimulus in several different ways. The term ‘stimulus’ here refers to objects, locations or even moments in time. Attention can be fractionated quite comprehensively into the following four sub-processes: (a) attentional orientation (the simple direction of attention to a particular stimulus); (b) selective (or focused) attention (giving attentional priority to one stimulus in favour of another); (c) divided attention (dividing attention between two or more different stimuli); and (d) sustained attention (attending to one stimulus over an increasing period of time). Many researchers would also cite spatial attention as one of the principal sub-processes. However, spatial attention can be thought of as a sub-category of each of the four mentioned above, since one can selectively attend to one spatial location, divide attention across two spatial locations etc. Another important point to remember here is that attention is distributed across both space and time. While the spatial allocation of attention is very-well researched, the temporal distribution of attention is less so.

Arousal can be defined very simply as the state of physiological reactivity of the subject (Broadbent, 1971; Kahneman, 1973; Eysenck, 1982; Cohen, 1993; Robbins and Everitt, 1995), ranging from sleep at one end to excitement or panic at the other. However, arousal should not be considered as an all-encompassing unitary construct. Both Broadbent (1971)and Eysenck (1982)have put forward similar theories of arousal which advocate a dual-level control mechanism. Essentially, a passive, low-level, physiological arousal system can be actively modulated by a higher-level, cognitive arousal system. The greater the fluctuation in arousal level of the first system (e.g. due to noise or lack of sleep), then the more the second system will try to compensate for these fluctuations, in an attempt to maintain an optimal level of arousal for the performance of a particular task. Robbins (1984)has discussed experimental evidence suggesting that these two mechanisms may be modulated by dopaminergic and noradrenergic mechanisms respectively. This will be discussed later in Section 4.

The most influential psychological theory of attention was Broadbent’s (1958) filter theory of early selection. He hypothesised that in order to cope with the surplus of information occurring in our everyday world, sensory stimuli were selectively filtered, based on their physical characteristics and before stimulus identification had occurred. Unattended stimuli were not subject to further processing. However, this position was criticised with evidence such as the famous ‘cocktail party effect’: although you may be attending to one conversation, you are normally aware of your name being mentioned in another (unattended) conversation. Clearly, such a phenomenon is not supported by Broadbent’s theory. In response to this and other, more empirical, lines of evidence Deutsch and Deutsch (1963)(and later Norman, 1968) advanced the ‘late selection’ position: all stimuli are non-selectively and involuntarily processed to the stage of object identification or semantics. However, this theory can be criticised since it takes no account of limitations in attentional capacity. It is clearly infeasible to process fully every single stimulus around us. Therefore, several ‘compromise’ models have since been advocated (Pashler, 1998). Triesman (1964)modified Broadbent’s early selection theory so that the selection mechanism was not an all-or-none filter, but rather a more flexible ‘attenuator’ which allows for the posibility that some pertinent but unattended stimuli can ‘re-enter’ the attentional mechanism for further processing. Another alternative theory is that of Kahneman (1973), who emphasised the interaction between attention and arousal or effort. Very simply, the extent to which a stimulus is processed depends on the amount of attentional capacity available, which is itself determined by levels of arousal and the degree of effort required to process the stimulus. This theory allows for the fact that we can selectively attend to one stimulus input, but may have residual capacity left over to process any other stimulus which may be relevant. These theories have been mentioned because they are historically and influentially among the most important. It is beyond the scope of this review to do adequate justice to the hordes of other attentional theories which are subtle variations on an early, late, or compromise model. The reader is referred to Pashler (1998)for a comprehensive review of this area.

This review is concerned specifically with the neural correlates of such attentional networks. Neural correlates refer to consistent and significant associations between particular cognitive functions and either (a) temporal patterns of neural firing, (b) levels of activity in discrete brain regions or (c) levels of activity of neuromodulatory neurotransmitters which can influence these other two systems.

Two of the most influential neuropsychological theories of attention came from Michael Posner and Marcel Mesulam in the 1980s. Both of these theorists advocated distinct neural bases for distinct parts of an attentional network, based on data from neuropsychological patients and single-cell recordings in monkeys. Posner and Petersen (1990)suggested that the attentional network can be separated into two sub-systems [Fig. 1(a)]. The posterior attentional system is involved in orienting to visual locations and is subserved by the posterior parietal lobe, the superior colliculus, and the thalamic lateral pulvinar nucleus. Conversely, the anterior attentional system is involved in detecting targets, for which efficient functioning of the anterior cingulate is important. Posner and Petersen (1990)also discuss an underlying arousal system located in the reticular system, which can influence both of these attentional systems. Evidence for the theory comes from studies of single-unit recording in monkeys, neuropsychological studies of brain-lesioned patients and functional brain imaging studies of normal volunteers. For example, Posner et al. (1984)reported that patients with lesions to the right parietal cortex were impaired on a test of spatial orientation of attention. More recently, using the same attentional orientation task in a Positron Emission Tomography (PET) experiment of healthy volunteers, Nobre et al. (1997a,b) localised the region of parietal cortex involved in spatial attention more accurately to right intraparietal sulcus. This task has become a ‘gold standard’ test of spatial orienting so I will take some time to describe it here, and will refer to it hereafter as the Posner covert orientation of attention task.

Subjects are shown a computer screen with a central fixation point and two peripheral boxes. Their task is to respond as quickly as possible to a visual target which appears briefly in one of the two peripheral boxes, while maintaing fixation in the centre of the computer screen. Their focus of attention, but not their gaze, is directed ‘covertly’ to the periphery. Immediately preceding the presentation of the target is a visual cue, which will either correctly (a ‘valid cue’) or incorrectly (an ‘invalid cue’) predict on which side of the computer screen the target will be located. A ‘neutral cue’ indicates that the target may appear on either side of the display. These cues can take the form of either (a) brightening of one of the peripheral boxes (an ‘exogenous’ cue which draws subjects attention in a ‘bottom-up’ manner) [Fig. 2(a)] or (b) a central arrow which points to one side of the screen or the other (an ‘endogenous’ cue which requires subjects to direct their focus of attention in a ‘top-down’ manner) [Fig. 2(b)]. In a typical behavioural experiment, the ratio of valid to invalid trials is 80:20, and so it is possible to calculate the ‘cost’ (in reaction time) of being cued to the wrong side of the display or the ‘benefit’ of being cued to the correct side. Unsurprisingly, subjects are significantly slower to respond to targets which have been invalidly cued, and faster to those which have been validly cued, compared to those preceded by a neutral cue (Posner, 1980). Fig. 1(c) shows this effect clearly with some of our own data collected from seven healthy subjects (Coull et al., 1997a).

Mesulam’s (1981) model of attentional orienting [Fig. 1(b)] suggests parietal regions are implicated in a sensory (usually visual) representation of space, lateral frontal regions (around the frontal eye-fields) in motor responses to spatial stimuli, and anterior cingulate in determining whether the stimulus is motivationally salient or not. An underlying reticular arousal system is also proposed. There are many important ideas shared by the models of Posner and Mesulam. Both highlight the importance of parietal and cingulate brain areas, particularly in the right hemisphere, in spatial attentional processes. Furthermore, the involvement of the cingulate in target detection, as suggested by Posner, can be easily reconciled with Mesulam’s idea that the anterior cingulate is involved in motivational salience since targets are, by their very definition, salient to the subject. The main difference between the two theories in this regard is that Mesulam would segregate the process of target detection into two constituent processes, motivational salience and motor response, being subserved by two discrete neuroanatomical foci (the cingulate cortex and frontal cortex respectively). Also, Posner’s notion that the parietal cortex is especially implicated in the disengagement of attention in a distributed spatial array can be reconciled with Mesulam’s idea that the parietal cortex is critical for forming a multimodal sensory representation of extrapersonal space. Finally, both models note the important influence of an underlying reticular arousal mechanism on the attentional system. This mechanism can alter the level of vigilance or arousal of the subject, which then has ‘knock-on’ effects on the higher-level cortical attentional systems.

In a review of this vast topic, it is necessary to constrain the discussion in some way so as not to cloud important issues. Therefore, I will focus on studies using healthy human volunteers and mention only briefly the relevant literature from studies using animals or neuropsychological patients. The reader is referred to Robbins and Everitt (1995)and Desimone and Duncan (1995)for reviews of the psychopharmacological and neuroanatomical bases of attention and arousal in animals, and to Cohen (1993)for a review of the neuropsychology of human attention.

Section snippets

EEG and arousal

One of the first windows on in vivo human brain function came from electroencephalographic (EEG) studies of human arousal in terms of the sleep/wake cycle. Activity in the low and high frequency bands of the EEG are considered to be an index of cortical arousal, such that power in the high frequency (alpha and beta) bands declines with the early stages of sleep onset while activity in the low frequency (delta and theta) bands increase. Similar changes in power of these high and low frequency

Functional neuroimaging

While ERPs have excellent temporal resolution, their ability to inform as to which areas of the brain are implicated in the processes being measured are drastically limited. Even with the more sophisticated multi-channel approaches now being adopted, the signals measured are only measured on the scalp. We have no way of telling what is happening sub-cortically. Patients with focal lesions to the frontal or temporal cortices, for example, are more informative but the lesions are often not

Psychopharmacology

While there have been many cognitive and neuropsychological theories of attention (Broadbent, 1958; Treisman 1960; Kahneman, 1973; Posner and Petersen, 1990; Mesulam, 1981), the neurochemical bases of these fundamental psychological processes remain to be elucidated. Furthermore, the combination of psychopharmacological methods with either PET, fMRI or ERPs has been attempted only rarely. With the reassessment of traditional localisationist models of brain function from the connectionist

Summary and conclusions

In this review of the neural correlates of attention and arousal, I have briefly examined the psychological relationship between arousal and attention. Then, drawing from three major neuroscientific methodologies, I have delineated the psychophysiological, neuroanatomical, and psychopharmacological basis of both of these constructs. Essentially, selective attention may act via an ‘early’ selection filter by amplifying signals in sensory association areas associated with stimulus processing.

Acknowledgements

JTC is supported by the Medical Research Council. I would like to thank Professor Chris D. Frith for his kindness in reading an earlier version of this manuscript and for his helpful comments.

References (123)

  • D.A. McCormick

    Cholinergic and noradrenergic modulation of thalamocortical processing

    TINS

    (1989)
  • J.L. Muir et al.

    Excitotoxic lesions of basal forebrain cholinergic neurons: effects on learning, memory and attention

    Behav. Brain Res.

    (1993)
  • R. Näätänen et al.

    Early selective attention effect on evoked potential reinterpreted

    Acta Psychol.

    (1978)
  • T.W. Robbins et al.

    Functions of dopamine in the dorsal and ventral striatum

    Sem. Neurosci.

    (1992)
  • A.F.T. Arnsten et al.

    Alpha-2 adrenergic agonists decrease distractability in aged monkeys performing the delayed response task

    Psychopharmacology

    (1992)
  • A.F.T. Arnsten et al.

    Alpha-2 adrenergic mechanisms in prefrontal cortex associated with cognitive decline in aged nonhuman primates

    Science

    (1985)
  • E. Awh et al.

    Attentional modulation of visual responses due to rehearsal in spatial working memory

    Soc. Neurosci. Abstr.

    (1997)
  • Broadbent, D. E. (1958) Perception and Communication. Pergamon Press, New...
  • Broadbent, D. E. (1971) Decision and Stress. Academic Press, New...
  • T.J. Brozoski et al.

    Cognitive deficit caused by regional depletion of dopamine in prefrontal cortex of rhesus monkeys

    Science

    (1979)
  • C. Büchel et al.

    Modulation of connectivity in visual pathways by attention: cortical interactions evaluated with structural equation modelling and fMRI

    Cereb. Cortex

    (1997)
  • M.C. Bushnell et al.

    Behavioural enhancement of visual responses in monkey cerebral cortex. I. Modulation in posterior parietal cortex related to selective attention

    J. Neurophys.

    (1981)
  • Cohen, R. A. (1993) The Neuropsychology of Attention. Plenum Press, New...
  • R.M. Cohen et al.

    Functional localization of sustained attention

    Neuropsychiat. Neuropsychol. Behav. Neurol.

    (1988)
  • Coles, M. G. H. and Rugg, M. D. (1995) Event-related brain potentials: an introduction. In: Electrophysiology of Mind:...
  • Coles, M. G. H., Smid, H. G. O. M., Scheffers, M. K. and Otten, L. J. (1995) Mental chronometry and the study of human...
  • M. Corbetta et al.

    Attentional modulation of neural processing of shape, colour and velocity in humans

    Science

    (1990)
  • M. Corbetta et al.

    Selective and divided attention during visual discriminations of shape, colour and speed: functional anatomy by Positron Emission Tomography

    J. Neurosci.

    (1991)
  • M. Corbetta et al.

    A PET study of visuospatial attention

    J. Neurosci.

    (1993)
  • M. Corbetta et al.

    Superior parietal cortex activation during spatial attention shifts and visual feature conjunction

    Science

    (1995)
  • J.T. Coull

    α2-adrenoceptors in the treatment of dementia: an attentional mechanism?

    J. Psychopharm.

    (1996)
  • Coull, J. T., Frackowiak R. S. J. and Frith C. D. (1998). Monitoring for target objects: activation of right frontal...
  • J.T. Coull et al.

    Clonidine and diazepam have differential effects on tests of attention and learning

    Psychopharmacology

    (1995)
  • J.T. Coull et al.

    Differential effects of clonidine, haloperidol, diazepam and tryptophan depletion on focused attention and attentional search

    Psychopharmacology

    (1995)
  • J.T. Coull et al.

    The α2 antagonist idazoxan remediates certain attentional and executive dysfunction in patients with dementia of frontal type

    Psychopharmacology

    (1996)
  • J.T. Coull et al.

    fMRI reveals distinct neuronal networks for attention to time or space

    Soc. Neurosci. Abstr.

    (1997)
  • J.T. Coull et al.

    The neural correlates of the noradrenergic modulation of human attention, arousal and learning

    Eur. J. Neurosci.

    (1997)
  • DeRenzi, E. (1982) Disorders of Space Exploration and Cognition. Wiley, New...
  • R. Desimone et al.

    Neural mechanisms of selective visual attention

    A. Rev. Neurosci.

    (1995)
  • J.A. Deutsch et al.

    Attention: some theoretical considerations

    Psychol. Rev.

    (1963)
  • E. Donchin et al.

    Is the P300 component a manifestation of context updating?

    Behav. Brain Sci.

    (1988)
  • Drachman, D. A. and Sahakian, B. J. (1979) The effects of cholinergic agents on human learning and memory. In:...
  • R. Elliott et al.

    Effects of methylphenidate on spatial working memory and planning in healthy young adults

    Psychopharmacology

    (1997)
  • Everitt, B. J. and Robbins, T. W. (1996) Central cholinergic systems and cognition. A. Rev. Psychol 48,...
  • Eysenck, M. W. (1982) Attention and Arousal: Cognition and Performance. Springer,...
  • P.C. Fletcher et al.

    Brain systems for encoding and retrieval of auditory-verbal memory: an in vivo study in humans

    Brain

    (1995)
  • C.D. Frith et al.

    Brain mechanisms associated with top-down processes in perception

    Phil. Trans. R. Soc. Lond. B

    (1997)
  • P.M. Grasby et al.

    Functional mapping of brain areas implicated in auditory-verbal memory function

    Brain

    (1993)
  • Hart, S. and Semple, J. M. (1990) Neuropsychology and the Dementias, pp. 66–74. Taylor and Francis,...
  • J.V. Haxby et al.

    The funcitonal organization of human extrastriate cortex: a PER-rCBF study of selective attention to faces and locations

    J. Neurosci.

    (1994)
  • Cited by (0)

    1

    Tel.: +44 (0) 171 833 7484; Fax: +44 (0) 171 813 1420; e-mail: [email protected].

    View full text