Article summary with Resting state functional connectivity: its physiological basis and application in neuropharmacology by Lu & Stein - 2014

Brain structures do not operate independently. Instead, they work in connection with each other. Resting state magnetic resonance imaging (rsMRI) is a technique that enables researchers to study activity at the level of functional networks. Accumulating evidence shows that rsMRI can be used to identify large-scale networks and their spatiotemporal characteristics. Although the method is widely applied, knowledge on the neurophysiological basis is incomplete. This article aims to review the fysiological studies on electical, metabolic and hemodynamic fluctuations that are implicated in rsMRI, as well as clinical applications of the technique.

What is rsMRI?

Contrary to functional magnetic resonance imaging (fMRI), rsMRI does not require participants to engage in a specific task. Since the discovery of the default mode network (DMN), rsMRI is considered of fundamental importance for understanding brain functioning.

During the past decade, several large-scale cortical and subcortical networks have been identified. These networks have been reported by different research institutes. Evidence indicates fluctuations across the life span and deviations in these networks in neurodegenerative and psychiatric disorders. Furthermore, similar networks have been observed in animals.

What are fysiological fluctuations?

Since the 50s, a large body of research has been conducted into spontaneous fluctuations in brain activity. Two parameters to indicate these fluctuations are oxygen availability (O₂a) and cerebral bloof flow (CBF). Older techniques, in which gold electrodes were placed in the brain, revealed results similar to those that can be measured with the BOLD-signal. Blood and oxygen availability converge in time, which suggests the two parameters share a common underlying mechanism. The most compelling finding was that O2a flows in slow waves of less than 0.1 Hz.

In addition, fluctuations in metabolic activity have been studies using redox reactions. Again, spontaneous activity at a very low frequency were found.

What neuronal correlates of spontaneous activity are there?

Electrical brain activity can be categorized based on speed. The fastest activity consists of separate action potentials, which last a few milliseconds, whereas post-synaptic potentials last about ten milliseconds. The slower potentials can be categorized into frequency bands: delta (1-4 Hz), theta (5-8 Hz), alpha (9-14 Hz), beta (15-24 Hz) and gamma (> 24 Hz). The cortical potentials of < 0.1 Hz are labelled ‘infraslow potentials’. In addition, potentials between 0.5 and 1.5 Hz are associated with hyperpolarisation (deactivation) and depolarisation (activation). These electrical currents in the brain can be measured with EEG, ECoG of LFP.

In fMRI research, validation of the BOLD-signal was crucial. LFP was the most suitable technique for this purpose, because the BOLD-signal is found to correlate strongly with pre- and post-synaptic fluctuations in local field signals.

For rsMRI, it remains unclear which neuronal processes cause the slow spontaneous fluctuations in brain activity.

Applications in addiction research

Despite the limited understanding of the neuronal basis of rsMRI, the technique is applied widely in studies on neuropsychiatric disorders. In addiction research, for instance, differences in networks after administration of cocaine, regional specificity of connectivity, and genetic biomarkers in nicotine addiction. These studies have been explorative in nature and have yielded conflicting results. Nevertheless, rsMRI can potentially offer new insights in preclinical research.

What should future research investigate?

Since rsMRI is a relatively new technique, several questions have to be clarified:

• What is the underlying fysiological mechanism of spontaneous slow fluctuations in brain activity?

• To what extent is functional connectivity related to anatomical structures?

• Is rsMRI possible at the small scale of synchronisation between neurons, instead of large-scale networks?

• How is functional connectivity related to behaviour?