Functional anatomy of intrinsic alertness: evidencefor a fronto-parietal-thalamic-brainstem network in theright hemisphere
Introduction
In accordance with van Zomeren and Brouwer [48]attention can be subdivided into twosubsystems, one representing the intensity aspects alertness and sustained attention and the other onethe selectivity aspects of focused and divided attention. Since intensity aspects are probably aprerequisite of the more complex and capacity demanding domains of attention selectivity [43]we were especially interested in the assessment of neural networks involved in the most basicintensity aspect, namely alertness. A typical task for the assessment of alertness in thesense of a general level of response readiness [27]is a simple reaction time (RT) measurement.Phasic alertness—i.e., the ability to increase response readiness by external cues—is called for inreaction time tasks in which a warning stimulus precedes the target stimulus [28]. If, however,one is interested in the examination of an optimal activation of intrinsic attentional resources, themeasurement of RT without an external warning is most appropriate. The distinctionbetween intrinsic, non-phasic activation of alertness and sustained attention, however, is not clearcutin the PET-literature although it has been given in taxonomies of attention 31, 42, 48. Some authors have defined even short periods of endogeneously maintaining vigilantresponding as sustained attention ([34]; other references given there). In our view, however, the distinction between this notion of sustained attention and alerting without warning stimuli lies in thefact that sustained attention tasks usually do not focus on speed of response as alerting tasks typicallydo. Rather, subjects have to monitor (frequent) occurrences of a certain stimulus [36]. The totalnumber of hits and misses for these stimuli is taken as the primary dependent variable. Thus,sustained attention tasks in contrast to alerting RT measurements probably do not necessarilyprovoke an optimal level of brainstem activation which was, however, shown in a study byKinomura et al. [14]in which subjects had to respond rapidly and in which response times wereregistered. Lesion studies in stroke patients revealed an important role of the righthemisphere for alertness; e.g., Howes and Boller [13]as well as Posner et al. [29]andLadavas [15]have reported a dramatic increase in simple visual and auditory RT afterRH-lesions. The observation that RH patients nevertheless can profit from a warning stimulus [29]shows that it is the intrinsic and not the phasic aspect of alertness which is impaired after lesionsof the right hemisphere. According to Posner and Petersen [30], the role ofnoradrenergic (NA) pathways in alerting may lie, in particular, in providing an adequate level ofarousal. From lesion studies in rats 37, 38, 39there is evidence for a righthemisphere (RH) bias in the NA system, originating in the locus coeruleus and projecting moststrongly to frontal areas. These studies lend support to the hypothesis that there also exists top-downregulation of this NA activation by the right frontal cortex since lesions in this area led to asignificant decrease of NA in both hemispheres and in the locus coeruleus. In a PET study lookingfor the effects of clonidine, an α2-adrenoceptor agonist and noradrenaline antagonist [8]on the activation during a rapid visual information processing task, the authors found a decrease ofactivation in the right thalamus and bilaterally in the superior frontal and inferior parietal cortex butan increase in the right anterior cingulate. It seems that under the influence of clonidine the subjectsneeded more effort to cope with the demands of the cognitive tasks, which might call for a higherinternal, cognitive control of arousal. According to Fernandez-Duque and Posner [10]the alerting network seems to co-activate, either directly or via the brainstem, the posterior attentionsystem in the parietal cortex involved in spatial orienting of attention [30]. PET- 3, 24and fMRI- studies [18]of sustained attention in normal subjects using tasks withoutRT measurements and lasting for short periods (40 s) have revealed a fronto-parietal RH networkfor the auditory, visual and somatosensory stimulus modality. Coull et al. [6]in another PETstudy used a rapid visual information processing task, which they describe as a test of sustainedattention, also requiring working memory and selective attention for its successful execution.Subjects had to monitor sequences of digits (100 or 200 per min) and were asked to press a mousekey in response to target sequences, but no RTs were reported. Compared with a rest condition, thetask revealed bilateral activation in the inferior frontal gyri, parietal cortex and fusiform gyrus aswell as in the right frontal superior gyrus rostrally. In comparison with a simple sustained attentioncontrol condition, the right frontal activation was no longer present. These results, too, corroboratethe assumption of a right fronto-parietal network for simple sustained attention tasks, whereas formore complex tasks also comprising aspects of working memory and selective attention anadditional involvement of left hemisphere frontal and parietal structures could be demonstrated (seealso [4]). Paus et al. [25]studied PET-activation for an auditory vigilance taskwhich lasted for 60 min with very infrequent target stimuli, thus representing a classical vigilanceparadigm in the sense of Mackworth [19]. In contrast to most other studies on sustainedattention, subjects had to respond as fast as possible to the very infrequent auditory target stimuliwhich appeared at a rate of about 1 per min. The authors found an increase in reaction time and ofθ-activity in the EEG over time which correlated with activation decreases of the thalamus,the right ventrolateral and dorsolateral frontal cortex, the parietal and the temporal cortex. Activationof the thalamus co-varied with activation of the right ponto-mesencephalic tegmentum, the anteriorcingulate and the substantia innominata. Coull [6]summarized similar results from anotherPET-study on sustained attention for the cortical and thalamic structures. Again, the specific role ofthe right frontal and parietal cortex in sustained attention was pointed out. The fact that activationin these two areas only decreased over time for a non-selective attention task, but not for a selectiveone was interpreted as a functional modulation of selective by sustained attention. In thePET-study by Kinomura et al. [14]which explicitly addresses activation of alertness, aninvolvement of thalamic and brainstem (mesencephalic tegmentum) structures in simple visual aswell as somatosensory RT tasks was revealed. The authors, however, only reported a region ofinterest analysis for thalamus and brainstem and did not provide information on cortical activation. Up to now in the visual modality, only subcortical or right hemisphere corticalbrain regions subserving intrinsic regulation of alertness or sustained attention have beenrevealed in functional imaging and animal studies although a more extended cortico-subcorticalnetwork for intensity aspects of attention might be hypothesized, as has been demonstrated for theauditory modality in a vigilance task by means of long duration reaction time measurements [25]. Therefore, a simple centrally presented visual reaction time paradigm without warning stimuluswas chosen for our study, just like in the PET-study by Kinomura et al. [14]. With such a simplereaction time task, aspects both of short term visual sustained attention (detection of target stimulias proposed in Robertson et al. [34]) and optimal intrinsic alerting by asking for fast responses27, 48are combined, at the same time avoiding involvement of selective attention orworking memory. We chose a long interstimulus interval of 3–5 s, making it very unlikely that apreceding reaction stimulus can act as a warning for the subsequent stimulus, which would changethe task into a phasic alertness task. Lansing et al. [17]in an EEG study have shown that awarning stimulus leads to an optimal preactivation if presented 0.6–1 s before the stimulus whichhas to be responded to.
Section snippets
Subjects
A total of 15 healthy, right-handed (according to a German translation of the Oldfield [23]handedness inventory) male volunteers (median age = 27 years, range = 23–34 years) with nohistory of neurological or psychiatric illness took part in the study which was approved by the ethicalcommittee and federal authorities. All were students of the Technical University of Aachen. Allsubjects gave informed, written consent and were paid for participation in the study.
Task
There were two different tasks,
Changes in rCBF
Results of the SPM96 comparison between the alertness and the combined control conditionare depicted in the three SPM{Z} two-dimensional projections of Fig. 1. Brain areas showing significant changes in rCBF between both conditions arelisted in Table 1, showing all regions comprising at least 40 voxels and having P-values for individual voxelsof at least 0.005. Alertness led to a significantly increased rCBF in the right anterior cingulate(Brodmann area BA 32), in the right middle frontal gyrus
Discussion
The results of this simple visual reaction time, intrinsic alertness PET-activation study showinvolvement of a number of right hemisphere (RH) structures comprising the anterior cingulate, themiddle frontal gyrus, the inferior parietal lobulus, the pulvinar and possibly the reticular nucleus ofthe thalamus, the middle and superior temporal gyrus, and the brain stem reticular formation. Thesite of the brain stem activation superimposed on the average structural MRI is identical to the
Acknowledgements
The study was funded by grant 01 KO 9507-8⧸11 from the German Ministry for Education andResearch (BMBF). We thank both anonymous reviewers for their detailed and valuable commentsand S. Fellows for language editing.
References (49)
- Aston-Jones G, Foote SL, Bloom FE. Anatomy and physiology of locus coeruleus neurons:Functional implications. In:...
- Cabeza R, McIntosh AR, Tulving E, Nyberg L, Grady ChL. Age-related differences in effectiveneural connectivity during...
- Cohen RM, Semple WE, Gross M, Holcomb HJ, Dowling S, Nordahl TE. Functional localizationof sustained attention....
- Corbetta M, Miezin FM, Dobmeyer S, Shulman GL, Petersen SE. Selective and divided attentionduring visual...
- Corbetta M, Shulman GL, Miezin FM, Petersen SE. Superior parietal cortex activation duringattention shifts and visual...
- Coull JT. Neural correlates of attention and arousal: insights from electrophysiology, functionalneuroimaging and...
- Coull JT, Frith CD, Frackowiak RSJ, Grasby PM. A fronto-parietal network for rapid visualinformation processing: a PET...
- Coull JT, Frith CD, Dolan RJ, Frackowiak RSJ, Grasby PM. The neural correlates of thenoradrenergic modulation of human...
- Dupont P, Orban GA, Vogels R, Bormanns G, Nuyts J, Schiepers C, De Roo M, Mortelmans L.Different perceptual tasks...
- Fernandez-Duque D, Posner MI. Relating the mechanisms of orienting and alerting.Neuropsychologia...
Cited by (382)
What is autonomous sensory meridian response (ASMR)? A narrative review and comparative analysis of related phenomena
2023, Consciousness and CognitionAnterior insula as a gatekeeper of executive control
2022, Neuroscience and Biobehavioral ReviewsBrainstem auditory evoked potential in cognitive impairment in patients with type 2 diabetes mellitus
2022, Experimental GerontologyA comparison of stages of attention for social and nonsocial stimuli in schizophrenia: An ERP study
2021, Schizophrenia Research