Elsevier

Brain Research Reviews

Volume 35, Issue 2, April 2001, Pages 146-160
Brain Research Reviews

Review
The cognitive neuroscience of sustained attention: where top-down meets bottom-up

https://doi.org/10.1016/S0165-0173(01)00044-3Get rights and content

Abstract

The psychological construct ‘sustained attention’ describes a fundamental component of attention characterized by the subject’s readiness to detect rarely and unpredictably occurring signals over prolonged periods of time. Human imaging studies have demonstrated that activation of frontal and parietal cortical areas, mostly in the right hemisphere, are associated with sustained attention performance. Animal neuroscientific research has focused on cortical afferent systems, particularly on the cholinergic inputs originating in the basal forebrain, as crucial components of the neuronal network mediating sustained attentional performance. Sustained attention performance-associated activation of the basal forebrain corticopetal cholinergic system is conceptualized as a component of the ‘top-down’ processes initiated by activation of the ‘anterior attention system’ and designed to mediate knowledge-driven detection and selection of target stimuli. Activated cortical cholinergic inputs facilitate these processes, particularly under taxing attentional conditions, by enhancing cortical sensory and sensory-associational information processing, including the filtering of noise and distractors. Collectively, the findings from human and animal studies provide the basis for a relatively precise description of the neuronal circuits mediating sustained attention, and the dissociation between these circuits and those mediating the ‘arousal’ components of attention.

Introduction

Beginning with Mackworth’s experiments in the 1950s, the assessment of sustained attention (or vigilance) performance typically has utilized situations in which an observer is required to keep watch for inconspicuous signals over prolonged periods of time. The state of readiness to respond to rarely and unpredictably occurring signals is characterized by an overall ability to detect signals (termed ‘vigilance level’) and, importantly, a decrement in performance over time (termed ‘vigilance decrement’). The psychological construct of ‘vigilance’, or ‘sustained attention’, has been greatly advanced in recent decades, allowing the development and validation of diverse tasks for the test of sustained attention in humans and animals and thereby fostering research on the neuronal circuits mediating sustained attention performance in humans and laboratory animals.

Sustained attention represents a basic attentional function that determines the efficacy of the ‘higher’ aspects of attention (selective attention, divided attention) and of cognitive capacity in general. Although impairments in the ability to detect and select relevant stimuli or associations are intuitively understood to impact modern living skills (e.g., driving a car), cognitive abilities (e.g., acquiring novel operating contingencies of teller machines, or detecting social cues important to communicate effectively), and possibly even consciousness [74], psychological research on sustained attention has largely focused on parametric, construct-specific issues and only rarely addressed the essential significance of sustained attention for higher cognitive functions like learning and memory [17]. In fact, the evidence in support of the fundamental importance of sustained attention for general cognitive abilities has largely been derived from studies in neuropsychiatric populations (see below). Thus, determining the brain networks mediating sustained attention not only represents a crucial step toward understanding the neuronal mechanisms underlying this critical cognitive function, but also toward the development of cognitive neuroscience-inspired theories of neuropsychiatric disorders characterized by impairments in fundamental attentional functions (see below).

Cognitive neuroscience research has consistently documented activation in right hemispheric prefrontal and parietal regions during sustained attention performance. The recent overview by Cabeza and Nyberg [8] illustrates the sobering degree to which such imaging data remain isolated if they are not embedded in a theory describing the neuronal mediation of the cognitive performance of interest. As we have argued earlier [85], the development of such a theory requires converging evidence from studies manipulating the cognitive function of interest (i.e., sustained attention) and measuring brain correlates (typically human imaging studies) with results from experiments on the consequences of manipulations in the integrity, excitability, or integrative capacity of defined neuronal circuits on the cognitive function of interest (typically animal behavioral–neuroscientific studies). Such converging evidence ensures that research aimed at explaining complex high-level cognitive functions by successively lower neural levels of description benefits from neuroscientific research approaches, and that efforts to determine low-level neuronal mechanisms of cognitive functions benefit from cognitive construct-driven research in humans [25], [63], [76]. Furthermore, it is crucial that evidence in support of neuronal circuits acting top-down to modulate attentional information processing (e.g., Ref. [75]) will be integrated with evidence on the role of ascending cortical input systems to arrive at a comprehensive theory of sustained (or any other form of) attention. In other words, the title phrase ‘where-top down meets bottom-up’ reflects two interrelated issues, that is (1) the convergence of information generated by cognitive-neuroscience research in humans and by the more fundamental neurosciences that is required to attribute sustained attention to defined neuronal circuits [85], and (2) the convergence of the functions mediated via top-down cognitive processes and bottom-up sensory input processing that is crucial for sustained attentional performance. These convergences are the focus of this review, indicating that the different levels of analysis employed by cognitive neuroscience research on sustained attention have sufficiently developed, albeit largely separately, to permit now an integration of evidence and therefore the development of a theory about the neuronal mediation of sustained attention.

The terms ‘top-down’ and ‘bottom-up’ processes, and the special focus of this review, require further clarification. ‘Top-down’ or ‘bottom-up’ regulation of attentional processes represent conceptual principles rather than referring to anatomical systems, such as ascending and descending projections. ‘Top-down’ processes describe knowledge-driven mechanisms designed to enhance the neuronal processing of relevant sensory input, to facilitate the discrimination between signal and ‘noise’ or distractors, and to bias the subject toward particular locations in which signals may appear [41]. For example, in sustained attention performance, the subject knows where to expect what type or modality of signal, how to respond in accordance with previously acquired response rules, and so forth. Furthermore, the subject develops expectations concerning the probability for signals and strategies for reporting signals versus false alarms (see below). All these variables influence performance, based on mechanisms that range from changes in sensory signal processing to the enhanced filtering of distractors and the modification of decisional criteria.

Such a ‘top-down’ biasing of attentional performance contrasts with ‘bottom-up’ perspectives that describe attentional functions as driven mainly by the characteristics of the target stimulus and its sensory context [97]. ‘Bottom-up’ perspectives attempt to explain a subject’s ability to detect targets and target-triggered attentional processing largely by the sensory salience of the targets, and their ability to trigger attentional processing by recruiting ‘higher’ cortical areas in a bottom-up manner (e.g., from the processing of a visual target in the primary visual cortex to temporal regions for object identification and to parietal regions for location). Importantly, ‘top-down’ and ‘bottom-up’ processes represent overlapping organizational principles rather than dichotomous constructs, and in most situations, top-down and bottom-up processes interact to optimize attentional performance [22].

Activation of top-down processes are traditionally considered a component of the frontal cortical mediation of executive functions. Such processes were previously conceptualized in the context of attention by Posner and Petersen’s [75] anterior and posterior attention systems that function to detect targets and bias the subjects’ orientation to target sources, respectively. Data from human imaging and primate single unit recording studies have substantiated the notion of top-down processes by demonstrating sequential activation of frontal-parietal–sensory regions, including decreases in activity in task-irrelevant sensory regions, and the modulation of neuronal activity in sensory and sensory-associational areas reflecting the top-down functions described above [19], [39], [41], [93]. This review focuses on the functions of the basal forebrain cholinergic projections to the cortex as a major component of the activation of top-down processes in the mediation of sustained attention. To avoid confusion, it should be reiterated that therefore, the anatomically ascending basal forebrain system is proposed to contribute to the functionally top-down processes in sustained attention.

As will be discussed below, the activation of cortical cholinergic inputs as a major component of the top-down processes in sustained attention performance acts to bias the processing of sensory inputs at all levels of cortical sensory information processing, thereby facilitating and maintaining sustained attention performance. Fig. 1 captures this hypothesis by illustrating the anterior attention system and its top-down regulation of posterior and sensory cortical information processing ( [75]; see also legend for Fig. 1). The basal forebrain cholinergic corticopetal projection system is conceptualized as a major and necessary component of these top-down processes. In sustained attention, this projection system is activated, via direct connections from the prefrontal cortex to the basal forebrain. Increased activity in cortical cholinergic inputs, that terminate in all cortical regions and layers, facilitate all aspects of the top-down regulation of sustained attention performance, ranging from the enhanced sensory processing of targets to the filtering of distractors and the optimization of decisional strategies. Evidence in support of this hypothesis will be discussed in detail, following a description of the construct ‘sustained attention’.

Section snippets

‘Sustained attention’ and ‘arousal’: conceptual overlaps and differences

As the terms ‘vigilance/sustained attention’ and ‘arousal’ have been used interchangeably, particularly in clinical contexts and in interpreting electroencephalographic (EEG) data, the specific meaning of vigilance/sustained attention needs to be defined and dissociated from the more global classification of brain states that include ‘arousal’. Obviously, the ability to perform monitoring tasks requires an activated forebrain and thus depends on ‘arousal’. Likewise, the ‘arousing’ consequences

General interpretational issues

Macroanatomical correlates of sustained attention performance have been determined mainly by two lines of research. First, findings from neuropsychological studies have identified areas of brain damage or degeneration which are correlated with impairments in sustained attention performance. Second, and more recently, functional imaging studies have located areas in the intact brain in which changes in metabolic activity correlate with sustained attention performance.

Complexities in the

Neuronal circuits mediating sustained attention performance

Evidence from studies designed to manipulate the functioning of defined neuronal circuits and assess the consequences for attentional performance, or to record from neurons constituting such circuits in animals performing in attentional tasks, is expected to explain the prominent role attributed to frontal cortical and parietal regions in human neuropsychological and functional imaging studies. Such reductionist explanations are also expected to specify the nature of the processing mediated via

Ascending noradrenergic projections: the ‘arousal’ link?

The basis for the conceptual dissociation between sustained attention and arousal is stressed above. In support of this dissociation, the findings from functional neuroimaging studies have suggested that the neuronal circuits mediating the necessary arousal for proper sustained attention performance differ from those correlated with sustained attention. Specifically, the arousal components may be mediated via activation of the thalamus. Attentional performance-associated increases in thalamic

Relationships with other forms of attention

Given the close theoretical relationships between different aspects of attention (sustained, selective, divided), it should not come as a surprise that the available data from neuropsychological, functional imaging, and animal experimental studies already suggest extensive overlaps in the circuits mediating different aspects of attention (e.g., Refs. [1], [6], [8], [10], [13], [42], [59], [99], [101]). For research purposes, it is important to maintain clear definitions of the aspects of

Major unresolved issues

The hypotheses that the activation of the fronto-parietal areas observed in humans performing sustained attention tasks reflects in part the consequences of cholinergic stimulation of these regions is attractive but remains unsubstantiated. The present model predicts more widespread areas of cortical activation than reported in most of the available studies, although the relatively selective activity changes reported in those studies may in part be due to the methods used to isolate regions of

Conclusions

Sustained attention represents a fundamental component of the cognitive capacities of humans. Aberrations in the ability to monitor significant sources of information rapidly develop into major cognitive impairments. Human neuropsychological and functional imaging studies have pointed to fronto-parietal areas, particularly in the right hemisphere, as being prominently involved in the mediation of sustained attention, or vigilance. Animal experimental evidence strongly supports the basal

Acknowledgements

The authors’ research was supported by NIH grants NS32938, MH57436, NS37026, and AG10173.

References (107)

  • A.M. Himmelheber et al.

    Increases in cortical acetylcholine release during sustained attention performance in rats

    Cogn. Brain Res.

    (2000)
  • R.D. Lane et al.

    Common effects of emotional valence, arousal and attention on neural activation during visual processing of pictures

    Neuropsychology

    (1999)
  • H. Moore et al.

    Bidirectional modulation of cortical acetylcholine efflux by infusion of benzodiazepine receptor ligands into the basal forebrain

    Neurosci. Lett.

    (1995)
  • P.C. Murphy et al.

    Cholinergic enhancement of direction selectivity in the visual cortex of the cat

    Neuroscience

    (1991)
  • T.W. Robbins

    Arousal systems and attentional processes

    Biol. Psychol.

    (1997)
  • T.W. Robbins et al.

    Comparative effects of ibotenic acid- and quisqualic acid-induced lesions of the substantia innominata on attentional function in the rat: further implications for the role of the cholinergic neurons in the nucleus basalis in cognitive processes

    Behav. Brain Res.

    (1989)
  • I.H. Robertson et al.

    Auditory sustained attention is a marker of unilateral neglect

    Neuropsychologia

    (1997)
  • L. Rueckert et al.

    Sustained attention deficits in patients with right frontal lesions

    Neuropsychology

    (1996)
  • L. Rueckert et al.

    Sustained attention deficits in patients with lesions of posterior cortex

    Neuropsychology

    (1998)
  • M. Sarter et al.

    Cognitive functions of cortical ACh (acetylcholine): lessons from studies on the trans-synaptic modulation of activated efflux

    Trends Neurosci.

    (1994)
  • M. Sarter et al.

    Cognitive functions of cortical acetylcholine: toward a unifying hypothesis

    Brain Res. Rev.

    (1997)
  • M. Sarter et al.

    Abnormal regulation of corticopetal cholinergic neurons and impaired information processing in neuropsychiatric disorders

    Trends Neurosci.

    (1999)
  • A.M. Treisman et al.

    A feature-integration theory of attention

    Cogn. Psychol.

    (1980)
  • J. Turchi et al.

    Cortical acetylcholine and processing capacity: effects of cortical cholinergic deafferentation on crossmodal divided attention in rats

    Cogn. Brain Res.

    (1997)
  • R.A. Adcock et al.

    Functional neuroanatomy of executive processes involved in dual-task performance

    Proc. Natl. Acad. Sci.

    (2000)
  • T. Audet et al.

    Attention deficits: is there a right hemisphere specialization for simple reaction time, sustained attention, and phasic alertness?

    Brain Cogn.

    (2000)
  • D.J. Bucci et al.

    Removal or cholinergic input to rat posterior parietal cortex disrupts incremental processing of conditioned stimuli

    J. Neurosci.

    (1998)
  • R. Cabeza et al.

    Imaging cognition II: an empirical review of 275 PET and fMRI studies

    J. Cogn. Neurosci.

    (2000)
  • A.A. Chiba et al.

    Basal forebrain cholinergic lesions disrupt increments but not decrements in conditioned stimulus processing

    J. Neurosci.

    (1995)
  • R.M. Cohen et al.

    Metabolic brain pattern of sustained auditory attention

    Exp. Brain Res.

    (1992)
  • M. Corbetta et al.

    Attentional modulation of neuronal processing of shape, color, and velocity in humans

    Science

    (1990)
  • J.T. Coull et al.

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

    Eur. J. Neurosci.

    (1997)
  • J.T. Coull et al.

    A fronto-parietal network for rapid visual information processing: a PET study of sustained attention and working memory

    Neuropsychology

    (1998)
  • N. Cowan

    Attention and Memory: An Integrated Framework

    (1995)
  • R. Desimone

    Neural mechanisms for visual memory and their role in attention

    PNAS

    (1996)
  • R. Desimone et al.

    Neural mechanisms of selective visual attention

    Annu. Rev. Neurosci.

    (1995)
  • R. Dias et al.

    Dissociation in prefrontal cortex of affective and attentional shifts

    Nature

    (1996)
  • J. Duncan et al.

    A neural basis for general intelligence

    Science

    (2000)
  • H.E. Egeth et al.

    Visual attention: control, representation, and time course

    Annu. Rev. Psychol.

    (1997)
  • J. Fadel, M. Sarter, J.P. Bruno, Basal forebrain glutamatergic modulation of cortical acetylcholine release, Synapse...
  • M.J. Farah

    Neuropsychological inference with an interactive brain: a critique of the ‘locality’ assumption

    Behav. Brain Sci.

    (1994)
  • G.R. Fink et al.

    Neural mechanisms involved in the processing of global and local aspects of hierarchically organized visual stimuli

    Brain

    (1997)
  • A.D. Fisk et al.

    Automatic and controlled processing approach to interpreting vigilance performance: a review and reevaluation

    Hum. Factors

    (1987)
  • S.L. Foote et al.

    Impulse activity of locus coeruleus neurons in awake rats and monkeys is a function of sensory stimulation and arousal

    PNAS

    (1980)
  • T.M. Gill et al.

    The role of acetylcholine within the medial prefrontal and posterior parietal cortices during sustained visual attention in the rat

    Soc. Neurosci. Abstr.

    (1999)
  • T.M. Gill et al.

    Sustained visual attentional performance-associated prefrontal neuronal activity: evidence for cholinergic modulation

    J. Neurosci.

    (2000)
  • T.H. Harding et al.

    A cholinergic-sensitive channel in the cat visual system tuned to low spatial frequencies

    Science

    (1983)
  • M.E. Hasselmo et al.

    Cholinergic modulation of cortical associative memory formation

    J. Neurosci.

    (1992)
  • T. Hatfield et al.

    Neurotoxic lesions of basolateral, but not central, amygdala interfere with Pavlovian second-order conditioning and reinforcer devaluation effects

    J. Neurosci.

    (1996)
  • L.A. Holley et al.

    Functions of cholinergic inputs to visual cortical areas: effects of visual cortical cholinergic deafferentation on visual attention in rats

    Soc. Neurosci. Abstr.

    (1995)
  • Cited by (860)

    • The basal forebrain serves social information processing

      2024, Current Opinion in Behavioral Sciences
    View all citing articles on Scopus
    View full text