Multifractal Dynamics in Functional Connectivity Analysis

Multifractal dynamics in functional connectivity analysis examine how interactions among brain regions fluctuate across multiple temporal scales with variability that cannot be captured by a single scaling exponent. Unlike static approaches, multifractal methods characterise the full spectrum of scaling behaviours, revealing both short- and long-range dependencies in neural signals. These techniques have illuminated how complexity in brain connectivity adapts to cognitive demands, ageing and pathology, by quantifying measures such as the variation of Hurst exponents and the width of singularity spectra. Practical applications range from understanding cognitive decline and task engagement to identifying biomarkers in neuropsychiatric disorders. By integrating waveletleader and detrended fluctuation algorithms, researchers delineate the multifractal signature of dynamic functional connectivity, offering a richer depiction of neural coordination than monofractal metrics alone. This framework underscores the brain’s capacity to operate near criticality, optimising information transfer and resilience across scales.

Research from Nature Portfolio

Recent studies have demonstrated that fractal scaling of resting and task-related fMRI signals is systematically modulated by cognitive effort, task novelty and ageing. One foundational investigation showed that the Hurst exponent of BOLD time series decreases under high cognitive load, novel tasks and in older adults, indicating suppressed scale-free dynamics during engagement. These changes align with concurrent reductions in global signal variability and functional connectivity strength, suggesting a global brain phenomenon sensitive to effort. A complementary electroencephalography study further revealed that local dynamic functional connectivity expresses true multifractal scaling across cortical regions, with frontal and occipital sites exhibiting stronger long-range correlations and higher degrees of multifractality compared with central and temporal areas. This spatial topology mirrors intrinsic network organisation and supports multifractal metrics as potential biomarkers for regional connectivity dynamics.

Research from all publishers

Recent work in healthy ageing has shown that dynamic fractal connectivity, assessed via sliding-window detrended cross-correlation analysis, not only diminishes in older adults but also correlates with reduced cognitive performance, offering over 90 per cent accuracy in age-group classification. In schizophrenia research, multifractal and entropy analyses of delta-band EEG connectivity revealed that patients exhibit stronger long-range autocorrelations and increased multifractality alongside reduced temporal complexity, supporting multifractal features as discriminative biomarkers. Earlier functional near-infrared spectroscopy and EEG investigations captured multifractal fluctuations of global and local network topologies during rest, demonstrating bimodal multifractal behaviour across frequency bands and identifying distinct patterns of autocorrelation and multifractality in short- and long-distance connections.

Multifractal Dynamics in Functional Connectivity Analysis publication trend

The graph below shows the total number of articles in multifractal dynamics in functional connectivity analysis across all publications each year (not limited to Nature Index journals).

Technical terms

Dynamic Functional Connectivity: Time-varying patterns of synchronisation among brain regions reflecting transient network states.

Multifractal Dynamics: A characterisation of signals by a spectrum of scaling exponents capturing heterogeneous temporal correlations.

Hurst Exponent: A measure of long-range dependence quantifying the tendency of a time series to persist in its direction over time.

Detrended Cross-Correlation Analysis: A method for evaluating power-law cross-correlations between two non-stationary time series across scales.

Singularity Spectrum: A function describing the distribution of local scaling exponents, with its width indicating strength of multifractality.

References

1. Fingerprints of decreased cognitive performance on fractal connectivity dynamics in healthy aging. GeroScience (2023).
2. The suppression of scale-free fMRI brain dynamics across three different sources of effort: aging, task novelty and task difficulty. Scientific Reports (2016).
3. Multifractal and Entropy-Based Analysis of Delta Band Neural Activity Reveals Altered Functional Connectivity Dynamics in Schizophrenia. Frontiers in Systems Neuroscience (2020).
4. Multifractal Dynamic Functional Connectivity in the Resting-State Brain. Frontiers in Physiology (2018).
5. Scale-Free and Multifractal Time Dynamics of fMRI Signals during Rest and Task. Frontiers in Physiology (2012).
6. Multifractal and entropy analysis of resting-state electroencephalography reveals spatial organization in local dynamic functional connectivity. Scientific Reports (2019).