The challenges involved in identifying the neuropathological substrates of the clinical syndrome recognized as schizophrenia are well known. Stereological sampling provides a means to obtain
accurate and precise quantitative estimates of components of neural circuits, and thus offers promise of
an enhanced capacity to detect subtle alterations in brain structure associated with schizophrenia. For
example, fractionator sampling is used to estimate cell number. A) The region of interest is split into a
number of blocks and a fraction of these is sampled. As every second block is sampled in the example,
the block samplings fraction *bsf* is ˝ in the figure. B) Sections are sampled from each block with a
constant section sampling fraction *ssf*. Sections could be sampled by exhaustive sectioning of all blocks
and sampling. Alternatively, if the blocks are cut as slabs with a known thickness *T, ssf* can be calculated
as the microtome block advance *BA* divided by *T*. C) The sections are subsampled in the area with a
constant area sampling fraction *asf*—typically by a uniformly random grid of unbiased counting frames.
*asf* is then calculated as the ratio between the counting frame area *a* and the area *A* of the basic tile of
the grid. D) When using optical disector probes, the sections are subsampled in the thickness. Often the
final section thickness varies due to an uneven shrinkage of the mounted section. This results in a
varying local sampling fraction from frame to frame as illustrated here. Therefore, the height sampling
fraction *hsf* should be calculated based upon the number-weighted mean section thickness ¯¯¯*t*_{Q¯} as
indicated, where *t*_{i} is the local section thickness and *q¯*_{i} the corresponding disector cell count of the *i*’th
frame. |