Every year, my April 1st article is dedicated to humor. Humor is very basic to what makes us humans. Babies laugh well before they can walk and talk. Some animals, like rats and apes, are also capable of laughter. Some scientists claim this is a side-effect of the mirror neurons which can be responsible for empathy and many other things. You can read further here, here, here, here [pdf], here, here, here, here, here, here and here. Today’s links are more complex and varied than usual, since writing the article made me ask myself many questions.
The science of humor is a strange thing. If you write down “humor formula” in the search engine, you will see how different scientists try to explain humor and manage to cover only a small aspect of what is considered “fun”. Engineers build strange AI setups that generate funny things, not as a primary goal but when something goes seriously wrong. Humor has been frustrating scientists for as long as I can tell.
Why is humor important? Funny connections are rare, as rare as gold particles in the send or numbers with high “difficulty” in a large hash formula. Being rare, the funny stuff is memorable. Additionally, funny thoughts tend to generate highly creative thinking, because they generate rare associations. When we share these gems with people, both parties enjoy the exchange and fill happier. Some say humor even prolongs life as it generates the “feel good” hormones, laughter increases the flow of the oxygen and smile strengthens our social bonds.
Yet, many scientists think that the origins of humor have very little to do with joy. Some point out that laughter appears to be a very spasmatic facial expression, a convulsive jerking that can be found in people suffering from hysteria. Others link humor with male aggression hormone that found more benign role in our life as a part of socialization adaptation. Rats and apes probably also can laugh just like we do. Both rats and apes are highly social creatures, with strongly developed mirror neurons. The mirror neurons are also responsible for some empathic behavior. The animals sometimes laugh when they are tickled and when they are playing with each other. Some antidepressants that make people happy can be derived from the rat humor hormone (GLYX13).
A totally different way to generate humor is used in artificial intelligence. A neural network, and maybe some day a quantum computer, can analyze millions of funny lines, anecdotes, and situations, comparing them to millions of lines and situations which are not funny and maybe even tragic. Then it may generate different rules of what funny is and why it sounds funny. These rules can be tested, by asking the artificial intelligence to generate the funny lines of its own. Personally, I speculate such a machine will have three levels of logic: the lower level will try to understand the situation, the middle level will try to map the situation with respect to other known situations, and the top level will produce the quality score of how funny the situation is. Artificial intelligence is still very far from replacing comedians, but we are on the verge of training it to understand knock-knock jokes or lines like “Why was 6 afraid of 7? Because of 789.” .
While even a rat or a baby will laugh when the situation is funny, the job of a master comedian is anything but easy. The comedian has to sift through myriads of possible situations, introducing minor changes that make the situation each time slightly funnier. Each time, the comedian performs a mental experiment, asking on simple question: “Will they laugh?”. I quote from Wikipedia:
The O’Shannon model of humor (OMOH) was introduced by Dan O’Shannon in “What Are You Laughing At? A Comprehensive Guide to the Comedic Event”, published in 2012. The model integrates all the general branches of comedy into a unified framework. This framework consists of four main sections: context, information, aspects of awareness, and enhancers/inhibitors. Elements of context are in play as reception factors prior to the encounter with comedic information. This information will require a level of cognitive process to interpret and contain a degree of incongruity (based on predictive likelihood). That degree may be high or go as low as to be negligible. The information will be seen simultaneously through several aspects of awareness (the comedy’s internal reality, its external role as humor, its effect on its context, the effect on other receivers, etc.). Any element from any of these sections may trigger enhancers/inhibitors (feelings of superiority, relief, aggression, identification, shock, etc.) which will have an impact on the receiver’s ultimate response. The various interactions of the model allow for a wide range of comedy; for example, a joke needn’t rely on high levels of incongruity if it triggers feelings of superiority, aggression, relief, or identification. Also, high incongruity humor may trigger a visceral response, while well-constructed word-play with low incongruity might trigger a more appreciative response. Also included in the book: evolutionary theories that account for visceral and social laughter and the phenomenon of comedic entropy.
Being interested in learning, memory, and creativity, the element of well-constructed comedic entropy is especially important. Quite often we instruct our students to generate funny visual associations because they are easy to remember. For truly memorable associations we try to generate something which we rarely use, e.g. we optimize comedic entropy. The more time we invest in making the visualization well-formed and rare, the more stable the visualization and the related memory becomes. We will also probably enjoy more reviewing well-formed associations later on. However, we do pay a heavy price in terms of time. The initial improvement may cause us next to nothing, as we simply add details we wanted to remember anyway. The next level of improvement involves making the visualization more lucid and believable, and this is a serious effort. Much more effort is involved in the gradual improvement of the details, removing the unnecessary and adding some touches to enhance the effect. We will probably spend our time better developing the context and associating our new visualizations with other visualizations we have, which is similar to working on triggers in comedy. What will trigger me remember or forget something? Is it a good trigger? Can I generate further triggers?
The surprise is at the heart of some good humor and creativity. When working with visualizations there are two approaches. We can take the first thing that comes to our head, and modify it slightly by adding details, till it becomes surprising and hard to recognize. Alternatively, we can throw away the most obvious visualizations, and then the brain will be forced to come up with a surprising one. These two approaches roughly correspond to the two major forms of creative thinking. The first approach works better when we read stuff, the second approach is better if we want to make sure to remember something without any time limitations. In comedy, we can see both approaches at work.
What about people who do not have the sense of humor? Babies five-month-old can laugh, but adults with autism often cannot distinguish humor. This is associated with social issues. However, it does not make them less creative or good at remembering stuff. Humor is not the only mechanism to make things memorable. Anything we find interesting and are obsessed with can become memorable. And if we cannot generate this personal connection, we can still build a logical structure, like a mind map or a flowchart to support the memory.
We find strange things funny. Animals tend to be funny, and scientists are quite often are funny. By trying to combine animals and scientists we can generate quite a few jokes. Here is a short selection from
- Pavlov walks into a bar. The phone rings, and he says, “Damn, I forgot to feed the dog.”
- Did you know that dolphins are so smart that within a few weeks of captivity, they can train people to stand on the very edge of the pool and throw them fish?
- My email password has been hacked. That’s the third time I’ve had to rename the cat.
- A dog has an owner. A cat has a staff.
- Dr.’s are saying not to worry about the bird flu because it’s tweetable.