If you're reading this sentence, chances are you're reading it silently (if you're reading it out loud, hey, that's cool too). Your lips aren't moving, you're not making any sound that other people can hear. But are you making "sound" in your head? Many people who read silently do so by imagining a voice speaking the words they are reading (and often, it's your own voice, so there's even a specific "tone". I wonder if this is what makes people react so strongly to some blog posts). This could be because when we learn to read, we associate symbols with verbal sounds until the association is effortless (as for reading learning in the deaf, it may occur another way).*
This is particularly interesting because it means that reading silently is producing "cross-talk" between different sensory systems, with written words producing an auditory experience for the reader. But is it really an auditory experience?
Perrone-Bertolotti et al. "How Silent Is Silent Reading? Intracerebral Evidence for Top-Down Activation of Temporal Voice Areas during Reading" Journal of Neuroscience, 2012.
It's a relatively easy hypothesis to assume that if we are "reading aloud" when we read silently, we should see increases in activity in the auditory-related areas of our brains, particularly things like the temporal voice area (which is particularly sensitive to voices as opposed to sounds in general). There are some fMRI studies that have indeed shown activity in this area during silent reading. But when does this occur? Is it part of the processing of silent reading? Do we have to read "aloud" to ourselves to read silently? Or is it something that happens later on, where we insert the voice reading "aloud" in our heads to aid us in comprehension?
This isn't something that fMRI can answer. But it is something you can answer if you have electrodes implanted in the right places. While most people don't walk around with electrodes in their heads and are unlikely to volunteer to do it for science, there is a small population of people who DO. Some of these people have severe intractable temporal lobe epilepsy. One of the last-ditch treatments for this is often the resection (taking out) of the temporal lobes. But before this is done, you have to determine if the seizures really are the result of temporal lobe activity, and where the seizures start (you really don't want to have to take out more than you absolutely need to). So patients get implanted with electroencephalographic electrodes that are underneath the skull and over the temporal lobes to monitor their activity.
And of course, if you've got the electrodes anyway, you might as well participate in a reading study.
So the authors of this study had four people previously implanted with EEG electrodes near the temporal lobes read a story silently and listen to a voice giving them instructions. While they read and listened the authors were taking recordings.
You can see above recordings from the four auditory areas, one from each patient (sadly, there were only four patients, it's a rare condition, and those who need surgical treatment for it are even more rare). You can see that these areas in the temporal lobes respond significantly to speech (French, Suomi, and reversed French) as compared to other random sounds like coughs, music, or animal noises.
And this area also responds to the written word.
You can see the blue lines (when the patients were asked to pay attention) showed increases in electrical activity in this area when the patients were presented with written words. This is an auditory cortex that usually responds to speech, and apparently, to our brains, the written word counts as speech.
What's particularly new about this study is that it not only shows that silent reading causes high-frequency electrical activity in auditory areas, but it shows that these areas as specific to voices speaking a language. This activity was only present when the person was paying attention to the task. The authors believe that these results back up the hypothesis that we all produce an "inner voice" when reading silently. And it is enhanced by attention, suggesting that it's probably not an automatic process, but something that occurs when we attentively process what we are reading. And the next time you read silently, remember that it's not quite to silent to your brain.
Perrone-Bertolotti M, Kujala J, Vidal JR, Hamame CM, Ossandon T, Bertrand O, Minotti L, Kahane P, Jerbi K, & Lachaux JP (2012). How Silent Is Silent Reading? Intracerebral Evidence for Top-Down Activation of Temporal Voice Areas during Reading. The Journal of neuroscience : the official journal of the Society for Neuroscience, 32 (49), 17554-17562 PMID: 23223279
*Side note: the authors also comment that "few would contest that most of our waking time is spent talking to ourselves covertly". This amuses me greatly. Do we? I mean, no citation for that, but do we all spend a lot of time talking to ourselves in our heads? Is this one of those things that everyone is slightly too embarrassed to talk about?