Brain Function, Intelligence, Mental Illness and the Future

Neuroscience has come on in leaps and bounds over the past few decades, especially as systems such as fMRI scanners can see where we are thinking at any moment. Brain chemistry is increasingly understood. Maybe one day we will all take 'happy pills' or have 'joy implants' to keep us feeling good. Yet there can be a dark side to most things. When we are happy, we are less concerned about injustice revolution. Take the drug addict who lies back in the squalid apartment with a beatific, inert smile. Can happiness be a mental illness? When it prevents us from living constructive lives, perhaps so.

What about intelligence? We also dance around the notion of smart pills. Indeed, drugs that arouse us may also affect cognitive function for the better as well as for the worse. But will being smart make us nicer, or more selfish? Will it help us change minds? Could we pop a pill before a negotation and come away with a brilliant deal? What are the social implications when some people are artificially happy, smart and persuasive, while others have to make do with a 'normal', struggling brain?

And yet mental health is still very much an issue of today. Whereas physical medicine would be unrecognizable to doctors from a hundred years ago, mental illness is still not far out of the dark ages. Indeed, there are parts of the world where it is still associated more with gods and demons than neural functioning. Personality disorders affect many of us in some shape or form. Indeed, more psychopaths can be found at the top of companies than statistics might expect. There are drugs to combat problems such as psychosis, but can be of limited help and come with unwanted side-effects.

We still have a long way to go before we avoid mental illness and achieve our potential, even with artificial assistance.

And yet we may be overtaken by independent artificial intelligence. When AI systems such as Google's AlphaGo learn the rules of a game and beat world champions within a single day, it seems machines will soon be smarter than us on pretty much any dimension we choose. Even in the emotional domain, systems are being developed to care for the elderly -- and the elderly are taking to this attentive concern, even though it is artificial.

A million dollar question in AI is around sentience. When does the machine become human, or it simulates humanity so well we can't tell the difference. Already, call centers are using AI systems to front-end phone calls, including detecting and manipulating emotions. And what happens when this simulation takes a turn for the worse? What does 'mental illness' mean in the machine world? Dysfunctional robots have been a popular theme in science fiction for decades. What if some of this comes true? Might your house robot sulk and not talk with you? Could AI systems become psychopathic and gain 'pleasure' in controlling us? Or might they just decide we are unnecessary? Are robot wars coming?

Just imagine a future where you tell your robot to make you a cup of coffee. The robot looks back and you, sulkily. "You don't like me, do you?" it says. Oh dear, you think, it's having one of its grumpy days. "Never mind, I'll get it myself", you say. The robot stalks off to sit and think about algorithms. You wonder if you should have bought a branded model rather than this cheap copy.

It all seems far-fetched at the moment, at least to many of us, but we should be thinking now about artificial intelligence systems in terms of mental health as well as how clever they might become.

Quick links

The ChangingMinds Blog!