membrane layers will result in higher diffusivity, whereas smaller spaces will lead to lower diffusivities (Alexander et al., 2007). Research has demonstrated that the parallel organization of WM fiber bundles and axonal membranes is the primary determinant of diffusion anisotropy in both peripheral nerves and the central nervous system WM, while myelin integrity appears to modulate the degree of anisotropy (Beaulieu, 2002, Concha et al., 2010). For example, rodent dysmyelination models showed that FA values still indicate anisotropy and reduce only by ~15% in the complete absence of myelin (cited from Mills and Tamnes, 2014). Almost all studies of myelination with normal brain development (Neil et al., 1998) or demyelination with disease-related processes (Sun et al., 2006) have found less diffusion anisotropy, where axons are less myelinated. Studies of dysmyelination and demyelination have also confirmed increased RD i.e., diffusion away from the axis (Harsan et al., 2006, Song et al., 2005, Tyszka et al., 2006), whereas AD, i.e., diffusion along the axis, has been suggested to be a more specific marker of axonal damage, loss, and density (Harsan et al., 2006). Precise mechanisms aside, AD and RD provide more specific information about tissue diffusion properties than FA alone.
