These conclusions must be tempered by several caveats. First, for the most part, the graph structures of functional networks in any one study have only been described at limited ages, and only within limited subsets of brain regions. Future networks should extend networks to include other brain regions to examine developmental trends more comprehensively. Second, only limited comparisons of network properties have been made thus far. Statistical comparisons were not made in small-world properties by either Fair et al. (Fair et al., 2009) or Fransson et al. (Fransson et al., 2010), and Supekar et al. (Supekar et al., 2009) made comparisons at only a single edge density. Thus, rigorous examinations of differences in network properties such as the small-world measures have not yet been performed. Confirmation and extension of these results awaits studies that examine influences of both threshold and edge density while examining larger, brain-wide functional networks over development from infancy through adulthood.
