I am, apparently, someone well on the linguistic side of the spectrum, yet I too was somewhat taken aback by this statement from the abstract: "Here we bring re

I am, apparently, someone well on the linguistic side of the spectrum, yet I too was somewhat taken aback by this statement from the abstract: "Here we bring recent evidence from neuroscience and allied disciplines to argue that in modern humans, language is a tool for communication, contrary to a prominent view that we use language for thinking" [my emphasis]. My own thinking on this has been to consider the origins of language. In a community of hominins on the verge of developing a language, would they not need, as a prerequisite, some non-linguistic ability to consciously grasp that at least some of the vocalizations of their peers represented feelings that they recognized in themselves? But that, by itself, would not be enough for them to develop a language; in addition, I feel, it would take both a desire to communicate their thoughts and feelings to others (motivated in part, I would guess, by recognizing that it would be useful to do so) and the recognition that this might be accomplished through artificial sounds and/or gestures with arbitrarily-allocated yet specific meanings. I feel that there's a bootstrap problem in regarding language as primarily a means of thinking that was found to be useful in communicating. Like you (and, apparently, the survey compilers) I am surprised to learn that a sizable minority of people report frequently visualizing words as text. On reflection, however, that sounds somewhat like a form of synesthesia, and I occasionally get enough of a hint of that to put it just barely in range of my personal experience - but then, I don't regard people whose experiences differ markedly from mine as wierdos.