In a recent piece for the Guardian’s Inner Voices series, Psychology PhD student Peter Moseley writes: 

Most of us will be familiar with the experience of silently talking to ourselves in our head. Perhaps you’re at the supermarket and realise that you’ve forgotten to pick up something you needed. “Milk!” you might say to yourself. Or maybe you’ve got an important meeting with your boss later in the day, and you’re simulating – silently in your head – how you think the conversation might go, possibly hearing both your own voice and your boss’s voice responding.

This is the phenomenon that psychologists call “inner speech”, and they’ve been trying to study it pretty much since the dawn of psychology as a scientific discipline. In the 1930s, the Russian psychologist Lev Vygotsky argued that inner speech developed through the internalisation of “external”, out-loud speech. If this is true, does inner speech use the same mechanisms in the brain as when we speak out loud?

We have known for about a century that inner speech is accompanied by tiny muscular movements in the larynx, detectable by a technique known as electromyography. In the 1990s, neuroscientists used functional neuroimaging to demonstrate that areas such as the left inferior frontal gyrus (Broca’s area), which are active when we speak out loud, are also active during inner speech. Furthermore, disrupting the activity of this region using brain stimulation techniques can interrupt both “outer” and inner speech.

So the evidence that inner speech and speaking out loud share similar brain mechanisms seems pretty convincing. One worry, though, is whether the inner speech we get people to do in experiments is the same as our everyday experience of inner speech. As you might imagine, it’s quite hard to study inner speech in a controlled, scientific manner, because it is an inherently private act.

Typically, studies have required participants to repeat sentences to themselves in their heads, or, sometimes, count the syllables in words presented on a computer screen. These lack both the spontaneity of typical inner experiences and the conversational quality (think of the conversation with your boss) and motivational purposes (“Milk!”) of inner speech. Although the experience is undoubtedly different for everyone (not everyone reports having “conversations” in their head, for example), what does seem clear is that inner speech is a complex and multifaceted phenomenon.

Why does it matter whether we have an accurate understanding of what’s going on in our brains when we use inner speech?

One reason is that understanding typical inner experience may be the key to understanding more unusual inner experiences. For example, psychologists have argued that hearing voices (“auditory verbal hallucinations”) might simply be a form of inner speech that has not been recognised as self-produced (although there are also important competing theories). Neuroscientists have found some evidence in favour of this theory. When they scanned the brains of people who reported hearing voices, they discovered that many of the same areas of the brain are active during both auditory hallucinations and inner speech. Broca’s area, for example, is often active in people when they’re hearing voices. Great promotions, high odds and great games from the leading bookmaker and casino on the link

But if we really want to know what the difference between what happens in the brain during inner speech and voice hearing – and how inner speech might become hearing voices – then first we need to understand what our internal talk is usually like. A recent study by researchers in Finland attempted to address flaws in previous brain-imaging studies of inner speech. Using functional magnetic resonance imaging (fMRI), they studied the difference between activity in the brain when participants experienced an auditory verbal hallucination, and when they deliberately imagined hearing the same voice. In this way, they controlled for aspects of the experience such as the sound and the content of the voice.

They found the main difference between the two conditions was the level of activation in a cortical region known as the supplementary motor area (SMA), which contributes to the control of movement. When participants heard voices, there was significantly less activation in the SMA, which fits with previous hypotheses suggesting that recognising actions as one’s own might rely on signals from motor cortical areas reaching sensory areas of the brain.

Of course, none of this is to say that understanding what happens in the brain is the only, or the most important, aspect of research into hearing voices. We also need to understand what the experience is like, how we can help people who are distressed by it, and when there’s a need for psychiatric care. But to do any of this, we first need to know what typical inner speech is like, and the underlying neuroscience is part of that understanding.

This article was originally published here in the Guardian’s ‘Notes and Theories’ science blog on Thursday 21 August 2014.