On the Spectrum

From Chronnectivity To Chronnectopathy: Connectivity Dynamics of Typical Development
A number of developmental disorders, including autism spectrum disorder (ASD), have
demonstrated abnormal functional hypoconnectivity and hyperconnectivity within the
connectome (Uddin, Supekar and Menon 2013). Atypical development of neural interactions
has been considered as a major basis in theoretical models of neuropsychiatric disorders
(Geschwind and Levitt 2007). Evidence suggests that short-range or intra-domain connectivity
is more dominant during infancy (Fransson, Skiold, Horsch, Nordell, Blennow et al. 2007, Gao,
Gilmore, Giovanello, Smith, Shen et al. 2011) and decreases with age during childhood and
adolescence, while long-range or inter-domain connectivity becomes more dominant in early
adulthood (Fair, Cohen, Power, Dosenbach, Church et al. 2009, Dosenbach, Nardos, Cohen,
Fair, Power et al. 2010). To our knowledge, no study has provided a baseline to understand
how the brain’s dynamic functional connectivity (i.e. chronnectivity) matures with age
during childhood, and compared this baseline with the dysfunctional chronnectopathy of
emerging mental illness. Likewise, the majority of existing models have made the assumption
that the brain’s functional connectivity is static over a period of multiple minutes. This has
been shown to be a major limitation as important dynamic patterns of connectivity could be
overlooked (Calhoun, Miller, Pearlson and Adali 2014). In the context of this paper, the term
chronnectome (and therefore chronnectivity) refers to dynamics in the functional network
connectivity among multiple brain regions. Also, in this paper, the term connectome refers to
the functional network connectivity at a macro-scale, with the assumption that the functional
connectivity over this period of time is relatively static. In this work, we address both of these
limitations by performing a chronnectomic analysis of typical development and autistic traits.
5
Over the past decade, various in vivo techniques, including functional magnetic resonance
imaging (fMRI), have been increasingly used to study neuronal connectivity in the developing
brain, particularly during rest (rs-fMRI). A wide array of methods has been used to categorize
the brain into functionally interconnected parcels, or intrinsic connectivity networks (ICNs),
such as the default-mode network. Importantly, the majority of existing rs-fMRI analysis
strategies operate under the assumption that connectivity remains stationary or static
throughout the entire measurement period (static functional network connectivity (sFNC)),
potentially obscuring transient patterns of connectivity at different time instances. Recent
advances in analysis methods have relaxed this stationarity assumption to yield indices
of dynamic functional network connectivity (dFNC) that offer unique chronnectomic
information (Hutchison, Womelsdorf, Allen, Bandettini, Calhoun et al. 2013, Allen, Damaraju,
Plis, Erhardt, Eichele et al. 2014) and are sensitive to neurobiological features of normal
brain development (Hutchison and Morton 2015) and psychopathology (Rashid, Damaraju,
Pearlson and Calhoun 2014).
99