This has been a long week. I want to thank all the people who expressed their support. It's one thing to have an honest scientific disagreement. It is another to insult people, question their political and moral intentions, and gang up to silence them.

It puzzles me that an academic would do that to another person.

I find this all the more puzzling that this project is deeply rooted in the historical literature. In college, I majored in history. That's where I discovered the Annales School, the 'longue durée' and the 'histoire des mentalités'.

One social scientist who inspired me in particular was Norbert Elias and his work 'On the Process of Civilisation' in which he describes how violence, sexual behaviour, bodily functions, table manners and forms of speech were gradually transformed over the modern period.

(This was in 1939: as you can see, our claim about the rise of trust is hardly a new one).

One of the reasons I moved to cognitive sciences after my studies in history is that I wanted to understand what were exactly the 'psychic processes' Elias was describing.

Years later, I asked myself: How can we measure these psychological changes? Can we quantify emotions and preferences in history?

Coralie Chevallier, with whom I shared these questions, introduced me to the literature on first impressions. She reasoned that since humans have all theses biases, painters should be aware of these biases, and would exploit them in order to display specific qualities.

At the same time, I was reading 'The Smile Revolution' by cultural historian Colin Jones.

In this book, Colin Jones describes the increasing value placed on smile in 17th and 18th France.

Actually, this change was not confined to painting.

But it is in painting that this was most visible. This is particularly striking in the paintings of Elisabeth Louise Vigée-Lebrun (see for instance her portrait of Madame Molé-Reymond as well as the cover of Colin's book).

Vigée-Lebrun's success tells us that her new style, which emphasizes naturalness and sensibility, was popular in late 18th France.

We teamed with social neuroscientist Julie Grèzes and cognitive scientist Lou Safra to further develop this idea.

Lou Safra leveraged the experimental literature on first impressions to develop the algorithm, an algorithm that detects PERCEIVED trustworthiness, not trustworthiness itself.

We are interested in the former, not the latter: what do portraits display for others to perceive, be it real or not.

The last piece of historical knowledge that informed this article is the work of economic historians. The dominant view has long been that there was no economic growth before the Industrial Revolution.

However, in the last 30 years or so, using wages, prices, economic historians have demonstrated and quantified the rise of living standard during the late medieval and early modern period, well before the Industrial Revolution.

(similar work has also been done for Late Imperial China, Mughal India, the Edo period in Japan and the Abbasid period in Iraq and Egypt).

Such an increase in living standard is also visible in urbanization, life expectancy, education.

That's all for now! I am going to be off Twitter for a while. Thanks again for your support.