Consciousness can mean different things to different people. To me, it describes a capacity for subjective awareness. This is a bit like selective attention, meaning that some things get perceived while others get ignored, in an ongoing experience-dependent stream. It is important to get definitions clear before even trying to explain of something like how consciousness emerges.

My definition is admittedly narrow. It doesn’t make sense to explain how consciousness emerges without first trying to understand how consciousness evolved, or why it might be adaptive. I think its evolution is linked to prediction.

Simple animals propelling themselves through the world (‘motile’ creatures) need to predict what might happen next (like, what they might bump into), and this is often based on prior experience or knowledge. They build internal models of their world and operate by constantly updating predictions based on these models. When they make a wrong prediction or encounter something new, this triggers oscillatory error signals in their brain which enables a re-evaluation of their models so that they don’t make this same mistake in the future. We call this learning.

This error-correcting mechanism is adaptive and leads to more efficient brain functions. Brains are adaptive prediction machines, and dopamine-driven learning mechanisms associated with error detection help them become even better prediction machines. So far, this doesn’t say anything about consciousness, but the pieces are in place.

The first predictions all brains need to figure out are about the body that they are attached to, to be able to map the relationship between actions and their consequences, and to be able to disambiguate self from non-self.

This applies to simple animals such as flies and fish as well as to a developing human foetus. So a rudimentary level of subjective awareness develops in an animal by building models of the world, starting first with a model of a body plan, and then with other models that often exploit the original body plan architecture. 

Consciousness does not emerge from complete silence, like switching on a computer; it is a process. It emerges because of a history of embodiment and a consequent need to optimize predictions about the world.

Sign up for the Daily Dose Newsletter and get every morning’s best science news from around the web delivered straight to your inbox? It’s easy like Sunday morning.

There are many theories of how brains produce conscious experiences. Some involve calculations about the complexity within a system. Others discuss necessary feedback loops that integrate different circuits and modalities in the brain. Predictive coding theory presents a complementary explanation that makes a lot of sense regarding the evolution of consciousness in animals. I think that consciousness emerges in any developing brain that is making models of its body plan, separating self from non-self. But I also think that conscious experience is less about making good predictions, and more about making mistakes. That’s why consciousness is adaptive.  A perfect prediction machine in an entirely predictable environment would not need to be conscious, because it would never need to detect anything new. I think we (and most animals) are conscious because it is inherently adaptive to have an ongoing subjective awareness of the mistakes we are constantly making, large and small, so that we can keep on learning and adapting. The real world is never entirely predictable. So consciousness emerges (or evolves) as we build models of the world, but consciousness is maintained through life because we are never getting it quite right.