This epistemological paper criticizes the lack of rigor in the use of the words "necessary and sufficient" in biology, in particular in neuroscience. These are words borrowed from formal logic, where they have a specific meaning. To say that something is necessary and sufficient for a biological function conveys a sense of importance and completeness, as if the biological question under scrutiny were then totally deciphered. Not surprisingly, such uses appear to be favored by the editorial policies of high-profile journals. Unfortunately, as the authors show, there is no instance where these are justified.

A typical example is when optogenetic activation of a neuron triggers a particular behavior. In this case, a particular type of manipulation is sufficient to trigger a behavior. The semantic drift occurs when one says that the neuron is sufficient for that behavior. The error is that it is the manipulation, and not the neuron itself, that is sufficient. It may well be that other neurons and various conditions play a key role in the expression of the behavior. The same issue occurs in genetics, when expressing a gene (the result depends on other genes and the organism, not just on that gene). The issue becomes particularly critical when one inhibits the neuron (or do a lesion) and the behavior does not occur anymore, which is described as the neuron being « necessary » for that behavior. The neuron then becomes "necessary and sufficient", even though those two words refer to two different types of manipulations, and not matched logical propositions. This causes serious issues in the interpretation of experimental data.

Related points are made in a chapter book by Alex Gomez-Marin, "Causal circuit explanations of behavior: Are necessity and sufficiency necessary and sufficient? " (2016).