Julian Jaynes coined a useful term in 1976 which I wish had gained currency before Jerry Fodor came out with The Modularity of Mind in 1983, because I think it captures all the useful aspects of the notion of "
mental modules" without any of the misleading connotations. He defines it briefly in a footnote from The Origins of Consciousness (p. 31):
"Aptic structures are the neurological basis of aptitudes that are composed of an innate evolved aptic paradigm plus the results of experience in development. The term . . . is meant to replace such problematic words as instincts. They are organizations of the brain, always partially inate, that make the organism apt to behave in a certain way under certain conditions."
I like this better than "mental module" because the notion of modularity implies something functionally detachable from the broader system it's embedded in. This is likely unhelpful when we're dealing with a biological system whose diverse parts evolved by a process of descent with modification from earlier parts, and particularly not when it's an organ whose functioning is defined by the ways in which its components
interconnect. A "structure" on the other hand carries no such connotation of functional (or geographical) closure. Jaynes' term also has the advantages of being both more inclusive in two useful senses: it can be applied to the neurological basis of any kind of behavior (not just what we'd narrowly call "cognition"), and it incorporates learning as well as developmental
canalization.
"Aptic" is a particularly, well, apt term because its layers of meaning all fit quite well with what we're expressing: it can mean both "prone/inclined" and also "appropriate", the latter meaning deriving from the Latin aptus ("fitted/suited"). Aptus is the past participle of apere, meaning "to affix, fasten, attach" (etc.), which is itself derived from the Greek haptein, meaning "to fasten, to hold, to touch". When this is prefixed with syn- ("together with"), we get synaptien, whence the word "synapse" for a biological process that functionally joins two neurons together. It's also the same linguistic branch from which we get the Latin adaptere ("to fit/adjust") and thence "adapt" and its derivatives, which is also fitting (heh) since these things are adaptations to the organism's environment.
I shall henceforth abandon "mental module" in preference for "aptic structure". It seems a small matter, but for reasons I'll soon elaborate I've come to agree with
Twain and Orwell that recruiting the right words for the job is crucial to achieving clearer thinking about unclear things.
(In another footnote in the Afterword, Jaynes also casually mentions "my friend W.V. Quine". Small world.)