It has long been thought that there is a close relationship between brain activity and cerebral blood flow [reviewed by Raichle (Raichle, 1998)]. While in 1890 Roy and Sherrington proposed the concept of an ‘intrinsic mechanisms’ responsible for coupling neural activity to blood flow (Roy and Sherrington, 1890), more than a decade earlier the Italian physiologist Angelo Mosso was already undertaking a detailed study of the changes in cerebral hemodynamics associated with various psychological states in humans (Mosso, 1881). Following up on these pioneering efforts, our understanding of the localization of brain function has advanced considerably thanks to techniques that detect the changes in brain blood flow associated with brain activity (Table 1). One of the most recent approaches is based on monitoring activity-induced increases in cerebral oxygenation. During functional activation cerebral blood flow increases more than the oxygen demands of the brain (Fox and Raichle, 1986), and such mismatch results in an increase in oxyhemoglobin and a decrease in deoxyhemoglobin (Raichle, 1998). activity-induced increases in brain oxygenation can be detected either by magnetic resonance imaging (mri), as blood oxygenation level dependent (bold) contrast (Ogawa et al., 1992), or, by optical imaging, as ‘intrinsic signals’ (Grinvald et al., 1986). Because of the importance and widespread use of these approaches in functional brain mapping, there is a great interest in elucidating the precise mechanisms by which these signals are generated and how accurately they reflect brain activity (Raichle, 1998).

In the present issue of Cerebral Cortex, Harrison and colleagues (Harrison et al., 2002) provide a much-needed piece of information concerning the relationship between intrinsic signals and cerebral vasculature. These authors used optical imaging to investigate the intrinsic signals evoked by auditory stimuli in the auditory cortex of the chinchilla. The spatial distribution of the signals was then correlated with the distribution of cerebral blood vessels, visualized in the same animal post mortem by corrosion casts. Through this careful analysis they were able to determine that the area in which intrinsic signals are detected overlaps remarkably well with the distribution of cerebral blood vessels supplying the activated region. Thus, signals emanated only from areas endowed with a rich vascular network, and no signals were obtained from adjacent regions in which the vasculature was less dense. Although neural activity was not mapped in the same animals in which the optical recordings and vascular casts were obtained, a map correlating unit activity with intrinsic signals in separate animals revealed that there was a general correspondence between active sites and location of the intrinsic signals. However, in some cases unit activity was not detected in areas in which intrinsic signals were generated. They also found that the cerebral vasculature is endowed with sphincter-like structures, often strategically placed at arteriolar branch points. The authors argue that these structures are involved in controlling the local distribution of flow within the vascular network.

These findings have notable implications for functional brain mapping using hemodynamic changes as a ‘proxy’ for neural activity. On the one hand, the finding that intrinsic signals identify reasonably well the area of activation, assessed by electrophysiological recordings, supports the validity of using vascular-based methods to localize brain function. On the other hand, the observation that the topography of the intrinsic signals overlaps with the distribution of cerebral blood vessels within the activated region suggests that the vascular geometry of a brain region dictates the spatial distribution of these signals. Therefore, depending on the spatial relationships between vascular network and active neurons, there might be instances in which the area of activation will not match the area from which the signals are generated. This point is demonstrated by the data of Harrison et al. (Fig. 2) showing that intrinsic signals can be detected from areas in which there is no evoked neural activity. Although in some areas vascular and neural architecture are well matched, e.g. in the rodent whisker–barrel cortex (Woolsey et al., 1996), this correspondence cannot be assumed to occur in all brain regions and for all activation paradigms. Another factor that may prevent complete overlap between vascular and activity maps is that the vascular dilatation responsible for the increase in blood flow evoked by neural activity is propagated in a retrograde fashion to upstream arterioles located outside the activated area (Iadecola et al., 1997). This phenomenon, termed retrograde vasodilation, serves the purpose of dilating upstream arterioles that are critical for regional flow control. Therefore, during functional activation, an increase in flow is observed also in remote regions in which no neural activity can be detected.

Another implication of the data of Harrison et al. is that the intensity of the intrinsic signals generated from a brain region depends on local vascular density. Therefore, given an equal degree of neural activation, the area in which vascular density is more sparse will generate a less robust intrinsic signal. Consequently, the ‘hot spots’ on a functional map may not necessarily indicate the areas that are most active in the execution of a specific task. While Harrison et al. focused on the relationship between intrinsic signals and vascular networks, Logothetis and colleagues investigated the specific components of the neural activity linked to the vascular signal (Logothetis et al., 2001). These authors recorded simultaneously electrophysiological parameters and BOLD fMRI in the monkey visual cortex and found that field potentials, rather than spiking activity, best predicted the vascular signal. This finding suggests that neural input and local processing, rather than neural output, are the most important determinants of the vascular change, a conclusion reached also by others on the basis of direct flow measurements in cerebellum (Mathiesen et al., 1998). Logothetis and colleagues also point out that, due to its higher signal-to-noise ratio, neural activity can be present in areas in which no BOLD contrast is detected, resulting in underestimation of activated areas assessed by fMRI. This conclusion also suggests caution in the functional interpretation of activity maps based exclusively on the presence or absence of the BOLD signal.

The finding that there might be sphincter-like structures capable of controlling the distribution of flow within the cerebral microvascular network is also of interest, and its significance has to be discussed in the context of a large body of work suggesting that cerebral blood vessels receive neural projections from intrinsic neurons (Eckenstein and Baughman, 1984; Vaucher and Hamel, 1995; Krimer et al., 1998). It is tempting to speculate that these sphincter-like structures are controlled by neural projections originating from neurons dedicated to flow regulation. Such neurovascular units would be well suited to control the timing and spatial distribution of the increases in flow evoked by neural activity, and could regulate flow for purposes other than to support the changing energetic needs of the tissue. For example, these neural networks could mediate anticipatory increases in blood flow in preparation for, rather than in response to, an increase in energy demands. Such anticipatory neurovascular control could, perhaps, explain why during activation the increase in cerebral blood flow greatly exceeds the oxygen needs of the tissue. This hypothesis, however, awaits experimental confirmation.

Table 1

Histor ical overview of activity-induced cerebrovascular changes in the human brain

Authors Finding Technique 
(Mosso,1881Increased brain volume during mental processing Monitoring of intracranial pressure through a breach in the skull 
(Fulton,1928Increased flow turbulence in occipital cortex during visual stimulation Recording of bruit from vascular malformation in occipital cortex 
(Kety,1950Increase in flow and cerebral metabolic rate for oxygen during anxiety Cerebral artero-venous difference of inhaled nitrous oxide (Kety and Schmidt method) 
(Ingvar and Risberg,1967Increased cerebral blood flow in speech-related areas during word repetition Regional brain clearance of 133Xe injected into the carotid artery 
(Fox and Raichle,1986Sensory stimulation increases blood flow in somatosensory cortex out of proportion to oxygen use Positron emission tomography using 15O water as a tracer 
(Belliveau et al.,1991Increased blood volume in visual cortex during photic stimulation MRI using a paramagnetic contrast agent injected intravenously 
(Ogawa et al.,1992Increased blood oxygenation level dependent (BOLD) contrast in visual cortex during photic stimulation MRI using intrinsic signals deriving from deoxyhemoglobin 
Authors Finding Technique 
(Mosso,1881Increased brain volume during mental processing Monitoring of intracranial pressure through a breach in the skull 
(Fulton,1928Increased flow turbulence in occipital cortex during visual stimulation Recording of bruit from vascular malformation in occipital cortex 
(Kety,1950Increase in flow and cerebral metabolic rate for oxygen during anxiety Cerebral artero-venous difference of inhaled nitrous oxide (Kety and Schmidt method) 
(Ingvar and Risberg,1967Increased cerebral blood flow in speech-related areas during word repetition Regional brain clearance of 133Xe injected into the carotid artery 
(Fox and Raichle,1986Sensory stimulation increases blood flow in somatosensory cortex out of proportion to oxygen use Positron emission tomography using 15O water as a tracer 
(Belliveau et al.,1991Increased blood volume in visual cortex during photic stimulation MRI using a paramagnetic contrast agent injected intravenously 
(Ogawa et al.,1992Increased blood oxygenation level dependent (BOLD) contrast in visual cortex during photic stimulation MRI using intrinsic signals deriving from deoxyhemoglobin 

References

Belliveau JW, Kennedy DN, McKinstry RC, Buchbinder BR, Weisskoff RM, Cohen MS, Vevea JM, Brady TJ, Rosen BR (
1991
) Functional mapping of the human visual cortex by magnetic resonance imaging.
Science
 
254
:
716
–719.
Eckenstein F, Baughman R (
1984
) Two types of cholinergic innervation in the cortex, one colocalized with vasoactive intestinal polypeptide.
Nature
 
309
:
153
–155.
Fox PT, Raichle ME (
1986
) Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects.
Proc Natl Acad Sci USA
 
83
:
1140
–1144.
Fulton JF (
1928
) observations upon the vascularity of the human occipital lobe during visual activity.
Brain
 
LI
:
310
–320.
Grinvald A, Lieke E, Frostig RP, Gilbert C, Weisel TM (
1986
) Functional architecture of cortex revealed by optical imaging of intrinsic signals.
Nature
 
324
:
361
–364.
Harrison RV, Harel N, Panesar J, Mount RJ (
2002
) Blood capillary distribution correlates with hemodynamic-based functional imaging in cerebral cortex.
Cereb Cortex
 
12
:
225
–233.
Iadecola C, Yang G, Ebner T, Cheng G (
1997
) Local and propagated vascular responses evoked by focal synaptic activity in cerebellar cortex.
J Neurophysiol
 
78
:
651
–659.
Ingvar DH, Risberg J (
1967
) Increase of regional cerebral blood flow during mental effort in normals and in patients with focal brain disorders.
Exp Brain Res
 
3
:
195
–211.
Kety SS (
1950
) Circulation and metabolism of the human brain in health and disease.
Am J Med
 
8
:
205
–217.
Krimer LS, Muly EC, 3rd, Williams GV, Goldman–Rakic PS (
1998
) Dopaminergic regulation of cerebral cortical microcirculation.
Nature Neurosci
 
1
:
286
–289.
Logothetis NK, Pauls J, Augath M, Trinath T, oeltermann A (
2001
) Neurophysiological investigation of the basis of the fMRI signal.
Nature
 
412
:
150
–157.
Mathiesen C, Caesar K, Akgoren N, Lauritzen M (
1998
) Modification of activity-dependent increases of cerebral blood flow by excitatory synaptic activity and spikes in rat cerebellar cortex.
J Physiol
 
512
:
555
–566.
Mosso A (1881) Ueber den Kreislauf des Blutes im menschlichen Gehirn. Leipzig: Viet.
Ogawa S, Tank DW, Menon R, Ellerman JM, Kim S-G, Merkle H, Ugurbil. K (
1992
) Intrinsic signal changes accompanying sensory stimulation: Functional brain mapping with magnetic resonance imaging.
Proc Natl Acad Sci USA
 
89
:
5951
–5955.
Raichle ME (
1998
) Behind the scenes of functional brain imaging: a historical and physiological perspective.
Proc Natl Acad Sci USA
 
95
:
765
–772.
Roy C, Sherrington C (
1890
) on the regulation of the blood-supply of the brain.
J Physiol
 
11
:
85
–108.
Vaucher E, Hamel E (
1995
) Cholinergic basal forebrain neurons project to cortical microvessels in the rat: electron microscopic study with anterogradely transported Phaseolus vulgaris leucoagglutinin and choline acetyltransferase immunocytochemistry.
J Neurosci
 
15
:
7427
–7441.
Woolsey TA, Rovainen CM, Cox SB, Henegar MH, Liang GE, Liu D, Moskalenko YE, Sui J, Wei L (
1996
) Neuronal units linked to microvascular modules in cerebral cortex: response elements for imaging the brain.
Cereb Cortex
 
6
:
647
–660.