Stereological methods for estimating the total number of neurons and synapses: issues of precision and bias

The emergence of a new generation of stereological techniques for counting objects in histological sections has prompted a debate about whether or not these methods are better than previously available techniques when they are used to make estimates of the total numbers of neurons and synapses in a neural structure. During this debate, the concepts of an unbiased estimate and that of a precise estimate have often been confused. A full understanding of the distinction between these two separate aspects of an estimate is required in order to be able to appreciate the virtues of these new counting methods and to apply them correctly. This review intends to make the fundamental issues of this debate more clear, and describes (I) the fundamental differences between the newer design-based counting techniques and previously available assumption-based techniques, and (2) the distinction between an unbiased estimate and a precise estimate.