When we read something silently we are, essentially, saying it to ourselves in our internal monologue. Psychology researchers at Britain's University of Nottingham wanted to know whether the voice that reads in our heads matches the voice that we read aloud in. In other words, does your internal monologue have an accent?
It's an interesting question. Although you might think it's a given, previous studies have suggested that the voice you speak with and the voice you think with might not be pronouncing words quite the same. This newer study, published in PLOS last fall, found the opposite—that there is at least some level of match between audible and silent pronunciation.
What I really like, though, is how they constructed the study. After all, you can't just ask people how they pronounce words in their heads. Like the question of whether you say "soda" or "pop" or "coke", once you start thinking about it hard enough to answer, you suddenly lose all ability to know what you do when you aren't paying attention. (Note: That soda/pop thing hasn't actually been scientifically demonstrated. It's just a bit of personal anecdata that I thought was relevant here.) In order to get around that problem, the Nottingham researchers had subjects read limericks while carefully monitoring their eye movements. The subjects were chosen based on their accents—one group pronounced their "a" sounds so that "path" would rhyme with "Kath". To the other group, that rhyme wouldn't rhyme at all. Instead, for them, "path" rhymed with "Garth".
The subjects read the limericks silently to themselves. But when they got to rhymes that didn't make sense with their spoken accent, there was a distinct disruption in eye movement. Basically, the physiological equivalent of the subjects having to stop and think, "Wait. That doesn't rhyme."
The other really cool thing I found in this paper: The fact that what we know about he author of the piece can influence how we read it.
… some previous studies have presented evidence to suggest that 'person-particular' knowledge of the author of a piece of text can influence reading of that piece of text. For example, it has been demonstrated that knowledge of the presumed author's speaking speed can influence how quickly people read aloud a passage of text [32]. This finding has also been replicated, and extended to silent reading [33]. Findings from other studies examining auditory imagery during reading have suggested that readers simulate aspects of the voices of the characters featured in the text (see [34], and also [35], for related findings). The current research supports, and extends these findings, by demonstrating that in the absence of information about the writer's voice, or that of characters involved in the text, inner speech during silent reading resembles the reader's own voice.
Henceforth, I shall refer to this as "The Just-Read-Trainspotting Effect", in honor of the three weeks during college when I couldn't get my inner monologue to stop drifting into an approximation of a heavy Scottish accent.
Via Stan Carey
Image: Eight-Minute Mouth Move, a Creative Commons Attribution Share-Alike (2.0) image from bruce-asher's photostream