For most people, hearing your own voice ‘inside your head’ is a defining part of what being alive is like. It was long assumed that this was a universal experience, but today there are reports of people with no inner voice. Most of us spend a lot of our cognitive resources on verbalising our inner world. Some estimate that 25% of our waking time is spent on it. Why do we have the ability to speak silently to ourselves? And what sets it apart from the spoken word?

​

We are born with the capacity to learn language. Ancestors who had greater ability to flag threats, coordinate activities and share information probably had higher chances of survival. This is why language is thought to have evolved as a function of natural selection. Speaking out loud to guide our own thinking is referred to as self-talk, and is a typical feature of children’s cognitive development. Even though it is directed to oneself, the content and structure are similar to social speech, as it is thought to develop through interactions with others. The volume of self-talk tends to gradually decrease as children mature, it first turns into whispers and later becomes silent. This silence is suggested to indicate that the overt self-talk has transformed into an inner voice, which usually happens by the age of 5-6 years. However, some teenagers and adults still use self-talk, especially during the initial phase of encountering a new problem.

​

What is the purpose of inner speech? Several hypotheses exist. Some suggest that it evolved to facilitate speech planning by predicting reactions to what we are about to say. Essentially a way for us to manage social interactions. Another suggestion is to facilitate individual problem-solving. That inner speech is an internalised version of how we solve problems with others (by exchanging information and perspectives). One explanation has focused on inner speech as being the primary mediator for complex cognition. However, the latter has received criticism due to the highly complex conceptual thought found in animals, pre-linguistic as well as language-impaired humans.

​

Inner speech is usually directed towards our own experiences. It has been found to use a compromised syntax, meaning that the sentences used are not grammatically complete and consist mostly of key words. It has been found to operate on a more abstract level compared to the spoken word, despite this we always know what the inner voice means. The speed of speaking out loud is estimated to be 120-180 words per minute, while the inner speech is processed 3-4 times faster.

​

There are current debates on the exact number of components and types of inner speech. However, researchers tend to differentiate between two main categories: monologic and dialogic inner speech. Monologic refers to using a single voice, often with the purpose of regulating or instructing oneself, as well as supporting one’s memory by rehearsal. Dialogic inner speech, on the other hand, is relational in nature and focuses on simulating social dynamics.

​

It is proposed that the processes responsible for generating spoken language are involved during inner speech. Neurological findings show that language production areas are active when we use our inner voice, and some language perception areas. During dialogic inner speech areas supporting theory of mind (the ability to take other people’s perspectives) also light up, showing activity in centres related to social cognition. Explanations based on forward models suggest that inner speech can be understood as the efference copy of the motor signal for speech action. This would mean that the inner voice is a sensory representation of what the speech action would have sounded like, a prediction that can run ‘offline’. However, others consider inner speech as the action of its own forward model.​ 

 

Voice and dialogue are essential to us as social beings, and tend to influence how we perceive our inner worlds. Words audible only to us help us comprehend not only ourselves but the world around us.