If we consider spoken language as a complex linear system, we will be able to use it as a pretty good standard for understanding individual psychology as well as interpersonal communication.
All words have words associated with them. Though we all share many of the same word-associations (coffee/beverage; booze/drunk; etc.), we also all have an abundance of word associations that belong only to us. I suppose this is fairly obvious, though I am not so sure it is well enough appreciated.
For example, we all know that coffee is a beverage and that booze can make people drunk, but beyond that each one of us has many other associations connected with these words, unique associations that have been gathered through years of experience. You may have pleasant associations with coffee and unpleasant ones with booze, or it could be the other way around. You may visualize the Caribbean when you think of either of these words, or Alaska. As the associations become richer and get further from the word that generated them, the psycholinguistic network they create will become increasingly complex.
If we could put diagrams of these associative networks on paper–including all of the images and feelings that go with them–I am sure that each person would be uniquely identifiable from just a few of them, in much the same way that we can be identified from our fingerprints. No two of us are alike in how we use and understand language.
The ways in which words, phrases, word-associations, gestures, tones of voice, expressions, dramatic poses, and so on strike each one of us are unique. This point is more than touched upon by an Emory University study, Metaphorically feeling: Comprehending textural metaphors activates somatosensory cortex, that demonstrates that “texture-selective somatosensory cortex in the parietal operculum is activated when processing sentences containing textural metaphors.”
What this means is that when people hear a tactile metaphor (soft as silk), the brain responds, at least in part, as if the person is feeling silk. I would contend that this and similar sorts of extended responses within the brain (and body) are a huge part of virtually all interpersonal communication. In this context, what FIML does is allow partners to access these deep associations and sort them out rationally without becoming lost in different associative versions of the “same” linguistic event.
FIML does not have to always depend on language, but it helps to bring it back to the actual words spoken as much as possible because the other sorts of associations and emotions that are generated during speech events are simply too complex to sort out without a stable reference point most of the time. Actual short bits of speech provide partners with the best data that both can readily agree upon. The many associations that are connected to that short segment of speech are often a big part of the material of a FIML discussion.