The reptilian brain does not exist

The neuroscientist Paul MacLean proposed in the mid-twentieth century a model called the triune brain: a reptilian core responsible for survival automatisms, a limbic brain responsible for emotions, a neocortex responsible for reasoning, superimposed like three successive layers of evolution.

The model enjoyed immense success, because it is visual, memorable, and provides an immediate explanation for the banal experience of inner conflict. It has passed from textbooks into corporate training, where it still circulates.

Comparative anatomy contradicts it on its central point. Reptiles and birds possess structures homologous to the mammalian cortex; brain evolution does not proceed by adding layers onto a preserved core, but by reorganising and rescaling structures present for a long time.

The very notion of the limbic system, whose name MacLean popularised, is judged poorly delimited: the list of structures placed in it has never been stable and no anatomical or functional criterion allows it to be fixed. Several researchers, including Joseph LeDoux, have proposed abandoning the term.

The idea of an emotional storey would presuppose that emotions are localised. A meta-analysis published in 2012 by Kristen Lindquist, Lisa Feldman Barrett and colleagues gathered the available imaging studies and looked for a brain signature specific to each basic emotion. It found none.

The regions involved are largely common from one emotion to another, and also take part in functions that are not emotional at all. Our language categories do not carve up the brain, which is a strong constraint on any theory that would treat them as separate programmes.

The field's most cited structure illustrates the same gap between research and its popularisation. Joseph LeDoux's work on fear conditioning in the rat established its role in threat detection, and he has spent twenty years correcting the reading made of it.

His current position distinguishes survival circuits, non-conscious and shared with other species, from the feeling of fear, which presupposes cortical circuits and a subjective experience. He goes so far as to write that speaking of a fear circuit was a vocabulary error with lasting consequences.

A false but simple model holds up better than a correct and complicated one, especially when it serves to justify a practice: explaining a reaction by an archaic brain saves one from examining the situation that produced it.

The minimum requirement, for anyone speaking about the brain in a training document, fits in one question: is the source cited a neuroscience publication from the last twenty years, or a retelling of a retelling of a diagram from the 1960s?