Any way you cut it: A new journal policy for the use of unbiased counting methods
✍ Scribed by Saper, Clifford B.
- Publisher
- John Wiley and Sons
- Year
- 1996
- Tongue
- English
- Weight
- 137 KB
- Volume
- 364
- Category
- Article
- ISSN
- 0021-9967
No coin nor oath required. For personal study only.
✦ Synopsis
A major problem that has confronted the editors of the Journal of Comparative Neurology over the last few years has been the assimilation of the revolutionary stereological methods for performing counts of neurons, synapses, and other structures. As noted by Coggeshall and Lekan in their Commentary article in this issue, the stereological methods for providing unbiased estimates of the numbers or densities of structures in a tissue section were introduced about a decade ago, beginning with the seminal work of Sterio in 1984. Stereological methods have since been validated and extended by many other workers. Coggeshall and Lekan provide a cogent and thoughtful description of why classical methods often produce counts that are not only incorrect, but may differ from the true numbers of structures by large and variable amounts that are difficult to predict. They provide a strong argument for requiring stereological methods for virtually all studies estimating numbers of cells, synapses, or other structures, from profiles in tissue sections. Given that more than half of the articles published in recent issues of the journal have used some estimation of the numbers or densities of cells, fibers, synapses, etc., the importance of this issue is clear.
At the same time, information about the stereological methods has penetrated only slowly and incompletely into the community that both submits papers to the journal and reviews those manuscripts. We still receive papers from authors who believe they have avoided double-counting errors by counting profiles in every fourth section, and multiplying by four (if you do not see the illogic of this position, please read section IV.B.3 of the Coggeshall and Lekan commentary; you will not be alone). Hence, we have had some papers accepted by expert referees with no mention or justification of the use of biased counting methods, either by the authors or reviewers. In other cases, knowledgeable reviewers have (justifiably) held authors to very high standards, requiring unbiased estimates of populations or densities. Because the editors are neither experts in stereology, nor are we able to provide a third review of every manuscript for adherence to these principles, we have not previously been able to establish a uniform standard for such papers in the journal.
Beginning with this issue, we would like to establish a policy that we will attempt to apply to all papers that are henceforth submitted to the journal. We will expect both authors and referees to adhere to this policy. To our knowledge, we are the first journal to take such a stand, and we hope by doing so to raise the standards both for the journal and for the entire field.
First, the editors agree with Coggeshall and Lekan that stereologically-based unbiased estimates are always preferable for establishing absolute counts or densities of structures in tissue sections. We expect that any papers that use simple profile counts, or assumption-based correction fac-