Science

This tool is truly the best thing that the science has come up with for getting two people to like each other. A little bit of optimisation can be effective—you should be eating healthy and exercising and doing the basics. But I tend to think in terms of relational optimisation: how can I put two people in the best position to share things about themselves that allow them to find conversational pathways that happen to go well?

What we see in the science is something very different. We often see that romantic partners basically see the best in each other. They're wearing rose-coloured glasses. And the primary thing that brings people together is what we would call attachment bonds or pair bonds: the idea that people are looking for somebody who they feel has their back, somebody who's going to celebrate their successes, and somebody who's going to be there for them.

If we can create excitement, wonder, and solutions to problems like loss of biodiversity, then I think that not only can we inspire the next generation, but we can give them hope.

If we want to build or maintain function, we need to apply a stimulus. And then, as much as we can, we need to provide the environment that allows us to best adapt to that stimulus.

What they found was that the majority of people — above 50% — maintained cognitive function from their 50s into their 80s. We knew this more than fifty years ago.

When those inputs are removed, it starts to prune function and structure, because those functions aren't needed anymore. It's still that same process — we just call it ageing. And it happens because we've removed physical activity, cognitive stimulus, and social connection. The brain responds accordingly.

If dementia is preventable, it means we're changing the trajectory of cognitive decline — which means that trajectory isn't inevitable.

Think about a rabbit sitting in a field. If that rabbit saw a hawk circling above and decided to wait for the back-propagation step before responding, it would be dead. The better you model the world, and the faster you can act on that model, the more likely your genes are to survive.

You cannot expend more energy than you can consume. That's a fundamental law of physics. If you do, you starve, you die, and you remove yourself from the gene pool. Biological systems have therefore been under enormous selective pressure to develop highly efficient intelligence.

I think intelligence is best understood as an entity that has the ability to improve a metric through repeated exposure over time. By that definition, machine learning algorithms are learning systems — they get better with more data and more exposures.

The brain is definitely not doing computation in the purest sense. We are not crunching numbers in binary ones and zeros in our heads. A more important question is: what are your inputs, what output do you want, and how intelligently can the system get from one to the other?