It is widely believed that smiling means a person is happy, and it usually occurs when they're meeting another person or a group of people. However, a new study led by the body language expert Dr. Harry Witchel shows this isn't always the case.

In his research, he asked 44 participants aged 18-35 to play a geography quiz game consisting of nine difficult questions so that they often got the answer wrong. Participants seated interacted with a computer alone in a room while their faces were video recorded.

After the quiz, the participants were asked to rate their experience using a range of 12 emotions including “bored”, “interested” and “frustrated”. Meanwhile, their facial expressions were then computer analysed frame by frame in order to judge how much they were smiling based on a scale of between 0 to 1.

Dr Witchel said: “According to some researchers, a real smile reflects the inner state of cheerfulness or amusement. However, behavioral ecology theory suggests that all smiles are tools used in social interactions, meaning cheerfulness is neither necessary nor rich for smiling. Our study showed that in these human-computer interaction experiments, smiling isn't driven by happiness; it is associated with subjective involvement(主观参与),which acts like a social fuel for smiling, even when socializing with a computer on your own.”

Surprisingly, participants didn't tend to smile during the period when they were trying to figure out the answers. However, they did smile right after the computer game informed them if their answer was correct or wrong. Participants smiled more often when they got the answer wrong. Dr. Witchel added: “During these computerized quizzes, smiling was greatly increased just after answering questions incorrectly. This behavior could be explained by self-ratings of engagement, rather than by ratings of happiness or frustration.”

In recent years, a lot of “business thinking” books have been published, all of which search for new answers on how to run organizations effectively. Obliquity tells us that the most profitable companies are not the most aggressive in chasing profits. Wikinomics demonstrates new models of production based on community and collaboration. Peter Miller’s new book, Smart Swarm, however, challenge leaders to think differently. He studied creatures, like bees and ants and he found their habits, actions and instincts can be applied to business. Miller believes his book is the first time anyone has explained the science behind management theory. “The examples of how ant colonies (蚁群) or beehives (蜂窝) work are appealing models for organizations and systems that can be applied in a business context,” he says.

So how exactly can bees help run organizations? “By the way they work independently before they work together,” Miller says. “Picture a huge beehive hanging on a tree, with about 5,000 bees competing for space. They know their population is getting too big, leaving them unsafe. They must all agree to find a new home. In today’s business environment, managers need to be able to make the right decisions under intense pressure. Yet, it is clear that some of the best-paid leaders in some of the biggest organizations can get it dramatically wrong. How is it that they can fail to make efficient business decisions when a large group of bees can make a critical decision about their hive in just a few seconds?”

According to Miller, “swarm theory” can help managers in three simple steps: discover, test and evaluate. The bees first realize they have a problem. They then fly into the neighborhood to find potential new sites. They come back and perform a “dance” to get other bees to follow them. Eventually, the bees with the best dance attract the most votes – and a decision is made. Miller says. “The bee example tells you that you need to seek out diversity in your team. You need to have a way of gathering up quite different ideas so you can make sure you pick the right one.”

Ants, in addition, can help businesses organize workflow and people. In an ant colony, there is no leader. Ants are self-organized, and respond to their environment and each other. One ant on its own could not raid (袭击) a kitchen cupboard, but one ant telling the next one that it’s worth following him to find food ends up creating a food chain. “In an ant colony, you get the right number going in and out searching for food, you get the right number taking care of the babies,” Miller says. “As a manager, this can tell you that your hierarchy (等级制度) is getting in the way of getting the work done.”

In looking through your social media, there’s a good chance you’ve noticed pictures accompanied by texts. The pictures are likely made possible by a text-to-image program called DALL-E. For example, Twitter user posted a tweet with the text, “To be or not to be, rabbi holding avocado, marble sculpture.” Then a matching picture appears below.

The AI models come from Google’s Imagen software as well as OpenAI. a start-up backed by Microsoft. On its website, OpenAI calls DALL-E “a new Al system that can create realistic images and art from a description in natural language.” But most of what’s happening in this area is coming from a relatively small group of people sharing their pictures. That’s because Google and OpenAI have not made the technology broadly available to the public.

The text-to-picture services identify the most important parts of a user’s text and then guess the best way to picture those terms. There’s generally a text box, a button to start the generation process and an area below to display images. To indicate the source. Google and OpenAI add watermarks in the bottom right comer of images from DALL-E and Imagen.

Engineers trained the models on various collections of words and pictures from the web. OpenAI recognizes the potential for harm that could come from a model. To avoid the risk, employees removed violent content from training data, and there are filters（筛选）stopping DALL-E from producing images if users submit（提交）violent or illegal content.

Boris Dayma, a developer from Texas spelled out the problems in an explanation of their software. Despite the risks, Dhariwal, a research scientist at OpenAI, said it could open up creative opportunities for individuals and could help with commercial applications for dressing up websites. Results should continue to improve over time.