Jaak Panksepp (1943-2017) was a pioneering scientist in the neuroscience behind emotions in humans and animals. In his book Affective Neuroscience, he explained the science behind what exactly is going on in our brains when we feel various emotions. The book is cold, hard science written by an expert practicing scientist in the field. That means the book is not an easy read and its main purpose is to be used as a university textbook. This is not a science book written for popular audiences. I was mainly interested in one chapter within this book, as I had heard it referred to at various times by other people discussing the phenomena of mammalian play. Panksepp is the scientist who conclusively demonstrated that mammal brains have a separate PLAY circuit in them, and I wanted to get the information on that research straight from the source.
Panksepp lays out his evidence for a ludic brain circuit in chapter 15 of his book. He used rats to study play. By doing various brain experiments on them where he damaged various parts of their brain (I hope this is not allowed currently under modern ethical research standards), or administered various chemicals, he was able to figure out where in the brain play urges come from in mammals. Rats, and all mammals, love to play, particularly when they are younger. The only type of play rats show is, of course, rough-and-tumble play (abbreviated RAT play in the chapter). RAT play is easily observable and quantifiable in rat interactions and thus was a perfect subject for Panksepp’s investigations. A side note is that Panksepp found that rats also laugh, and that they do it a lot while they are playing. Humans obviously have more forms of play than just RAT play, due to our more advanced brains. But Panksepp is careful not to overstep the bounds of his actual experimental results with rats when discussing how play probably works in the brains of humans. He admits that he can not experimentally show where these alternate forms of human play come from in the human brain itself, but he does say his own best guess would be that humans also have the one PLAY circuit in their brains, just like rats and the other mammals do, and that those ludic impulses that it produces are being modified by the reasoning centers of the brain. He does not think that human brains have separate play circuits in the brain for different types of play since such a configuration would seem to be unnecessary. Regardless of his conjecture in this area, there are some interesting things he managed to determine experimentally about play.
First, he found that rats quickly establish a relationship with each other during their play where one rat is clearly the superior player. This superior player will usually end up pinning the other player on their back at the end of their RAT play. This signals that one rat has won and that it’s time for a breather perhaps. The interesting thing is that superior player rats will intentionally lose from time to time in the play to keep the other playmate interested. Rats will only continue to play with specific playmates if they win roughly 30% of the time at minimum. Thus, the superior player can win 70% of the time without risking losing a playmate. If a rat is successful in their play 30% of the time, they will rush through a maze to get to a play area to play with the other rat that is beating them the majority of the time. The rats also know that the players that win most of the time are worthy of respect, assuming those rats don’t cross the 70% line. Thus, play does show, to the rats in a community, a competency hierarchy. I find that fascinating in its implications for playing with young children. Adults often wonder where to draw the line in how hard they should honestly try to win when playing with children.
Based on these experimental results it would not be too far of a hypothetical leap to surmise that human children brains are probably fine with losing somewhere around three quarters of the time (assuming their other emotions are being engaged in a positive way during the experience). So I personally think the most scientifically sound approach for adults that play with children (particularly young children) is to look for games with a mix of play and randomness that results in a child winning somewhere between 30% and 50% of the time. A purely random game, such as Candy Land, will result in a roughly 50%-win rate for two players. The problem with such games is that the child has no room to improve. There is no superior player in evidence for the child to imitate. On the other hand, a game of pure strategy like chess, in which the adult is going to win 100% of the time (assuming they are a competent player), is probably going to frustrate the child’s mammalian PLAY circuit. The way around this issue is for the adult to purposefully throw the game to the child 30% to 50% of the time. A problem with this approach is that the child will probably eventually figure out that the adult is purposefully losing, and that is guaranteed to suck some of the fun out of the play for a human child as they age (and their ability to understand other’s viewpoints grows). I would assume rats don't have this problem of potentially getting upset by an opponent purposefully throwing a game. The solution I believe is to look for games which have just enough strategic choices that the child has a chance of mastering them and subsequently pushing the win rate to 50% (in a two-player game), but that also contain enough randomness that they might still win at a reduced rate even if they make the wrong choices. The younger the child, the fewer strategic choices the game should contain, but the choices should never drop to zero (as in say Candy Land).
Thankfully, there are now tons of tabletop games and video games being produced that require just a little bit of strategic choice-making to produce a truly random game with a 50%-win rate. For instance, a worthwhile replacement for Candy Land is the board game Monza. In that game, players need to figure out how to optimally order the colors they get when they roll dice with colored faces. Doing so makes their little race car move the farthest down the track of colored spaces that it can with that roll. Thus, young children are learning to sequence a few variables in an optimal pattern while playing the game. They can see themselves getting better at the game as they learn to do this, until eventually they master it and the game becomes a purely random one. The game allows the adult to play to their best ability and provide an example to the child of how to figure out optimal arrangements. Even if the adult always makes the optimal arrangement, and the child does not, there is still a significant chance (but not as much as a 50% chance) that the adult is going to lose anyway due to the randomness in the dice rolls. There is no need for the adult to throw the game, assuming the game’s color arrangements can be eventually figured out by the child in question.
An example familiar to everyone of limited choice-making, but without randomness, is tic-tac-toe. There are a couple of move choices that need to be figured out in that game. Once both players know them, the game will always end in a tie. If the child is old enough to learn those move sequences at a quick pace, through game-play, there is no need for an adult to throw the game to them at any point. They will start winning soon enough that they don’t trigger the PLAY circuit shutdown that occurs with constant losing. The problem with this style of game, however, is the fact that a young child who does not know the optimal move pattern is guaranteed to never win against an adult. Thus, there is a need in a non-random game for the optimal pattern to be able to be figured out comparatively quickly, as opposed to a game with randomness, such as Monza. As long as the inherent randomness of the game is allowing the child to win roughly 30% of the time without figuring out the optimal moves, they can hopefully keep working at the game indefinitely while they learn those rules and won’t get frustrated in the process.
Once again, this assumes all mammals have a play circuit tipping point of somewhere around 30%. Perhaps humans on average can tolerate winning even less than 30%. Personally I don't care about winning or losing in play, but I know others who do. This also assumes an emotionally healthy child playing with supportive adults of course. If the young child has been emotionally damaged in their play experiences, say through an adult’s introduction of teasing or trash-talking into play, then their tolerance for losing is probably going to be far less.
In education circles there is a popular concept laid out by Vygotsky called the Zone of Proximal Development. The concept is that students should be pushed just hard enough in their studies that they are forced to improve with the teacher’s help, but do not cross their frustration threshold while doing so. Vygotsky based this on arm chair philosophizing without any real scientific evidence, as often happens in the social sciences. However, it seems that Panksepp may have provided an amount of oblique scientific support for Vygotsky’s theory by showing that mammalian PLAY circuits, at least, do have a frustration tipping point in them. Granted the PLAY circuit is not attached to learning in any way that can be emphatically demonstrated so far (as is discussed later), but at least it is a similar brain concept.
Another thing Panksepp demonstrated is that play is not related to aggressive activity in the brain. The PLAY circuit itself is not part of the brain circuitry for aggression. Someone untrained in rat observation would probably see rats playing and think they were fighting, when such was not the case. Rats play in a very different manner than they fight. Rats who are actually fighting stand up on their hind legs a lot and box each other with their front paws. They also prance sideways and assume aggressive postures. Rats who are playing do not do that. Playing rats frantically hop around, pounce at each other, and dart. Sometimes rats who are playing go too far and start to fight. The instant the play turns into aggression, all the play behaviors cease, and the rats start performing their fight behaviors instead. There is no overlap. Once aggression enters, the play is done. Not only that, but the experimental introduction of increased levels of testosterone into the rats shows some play reduction over time, as play bouts quickly degenerate into aggression. Conversely, the introduction of antiagressive drugs (which reduce fighting) into the rats, such as fluprazine and eltoprazine, increases play activity. Finally, lesioning the VMH area of the brain, which produces massive aggression increases in the subjects, massively reduces play activity. That is a fascinating experimental result.
I think this shows that play is not related to aggression in humans at the biological level. If it is not in other mammals there is no reason it should be in us. This puts the lie to many of the behaviors some people evidence in play and try to excuse. It is often most evident in contact sports, but it happens often enough in all types of playing. Players losing their cool, and demonstrating aggressive displays and postures, are no longer playing. They should be called out on it and eliminated from the play activity for the time being. The “I was only playing” excuse for aggressive behavior does not seem to hold water on an actual neurological or hormonal level. No, they weren’t using their PLAY circuit in their brains, they were using their fight circuits, and should be treated appropriately. The rats are smart enough to recognize the difference. We should be too. Unfortunately, with our advanced brains comes the ability to make great sounding excuses for bad behavior. Rats don’t have that ability. On the flip side, this also stops the accusation some people point at rough-and-tumble play as somehow being undesirable, as they suppose it is a show of aggression. We have the actual scientific biological evidence here that RAT play and aggression are not at all the same thing on a neurological and hormonal level. The problem is that too many people get aggressive during play and then try to pass it off as still playing. This then taints the idea of play itself for many people on the receiving end of the aggression. If these individuals displaying aggression were summarily called out on it and eliminated from the play experience as a matter of course, it would serve to decouple those two different behaviors in the minds of everyone. It would also start to quell much of the aggressive behavior that occurs in play currently.
Another thing Panksepp discusses is that hunger and illness reduced playfulness in the rats. Play impulses are contingent upon feeling good. Going along with that is the discovery that the introduction of small amounts of opioids into the rats increases their play drive and that alternatively introducing opioid antagonists into their systems can eliminate the play drive. During play the brain releases opioids naturally. The better you feel, the more you want to play, and conversely, the more you play the better you feel.
Next Panksepp gets into a discussion on the actual function of play. Here he discusses the fact that the function of play is largely a mystery scientifically. There are many theories as to its function, but Panksepp points out that none of them have valid scientific experimental proof behind them. They are all simply hypotheses, no matter how persuasively they are argued for or against. In terms of play being helpful to learning, Panksepp emphatically lays out that there is no such valid evidence. He surmises that all attempts to connect play with learning by social scientists so far may well be, in actuality, testing something entirely different from the PLAY circuit in the brain. They may simply be testing the power of positive social interactions to promote learning, while thinking they are testing play. He states that while it may be very attractive to want to link play and learning there is absolutely no empirical evidence that such is the case or not the case. The only actual experimental evidence so far for a function, is that rats who are deprived of play throughout their lives do poorly against other rats who have had lots of play, in terms of problem solving. But unfortunately, the observed increase in ability is a mild one at best.
However, Panksepp has found that play appears related to REM sleep in that it seems to exercise the potentials for organizing affective information in emotional circuits. In other words, play might be targeting various emotional circuits within the brain for later growth during REM sleep. But this is still at the hypothesis stage. Another promising hypothesis is that play may have anti-stress and positive health effects. Once again, these still need to be shown experimentally.
Panksepp then spends time distinguishing the SEEKING circuit from the PLAY circuit in the brain. Exploration and play are often combined in the literature of various social sciences as somehow being related, and Panksepp points out that they are not. They are two different brain circuits and the best evidence so far seems to indicate that they are antagonistic to each other. For instance, administering psychostimulants (which increase exploratory urges) to mammals, including humans, reduces play urges in them.
Panksepp ends his discussion on the science of play by discussing play disorders. Mania, impulse control disorders, hyperkinesis, and attention deficit disorders show what one would expect of an overstimulation of the PLAY circuit. However, there is no direct evidence of this at present. Indirectly, he notes that parents of hyperactive children who are given stimulants (which increase cortical functions) often report a decrease in the children’s play activity. It is believed that heightened cortical activity probably suppresses the play circuit. Also, other mammals given these drugs do decrease their play frequency. Dopamine blocking agents used to suppress Tourette’s syndrome also reduce playfulness in animals. Thus, these disorders may be related to an overactive PLAY circuit in the brain. Finally, he mentions the fact that children with pronounced autism still like RAT play. This means that the PLAY circuit is not connected to those interpersonal relationship circuits affected by autism. One thing he does note is that subjects with autism appear to not include a pinning activity in their play. Remember that pinning seems to function as a breather and summation to a bout of play. Thus, the act of pinning may be a function not of the PLAY circuit itself, but of another brain circuit, which is affected by autism. He ends the section with an aside that the main use of play in an educational setting may be as a powerful reinforcer for good behavior. One thing that is not controversial, is that play is fun for mammals and this makes it a great reward in an educational setting.
In his afterthought section, Panksepp closes with the idea that play’s main function may well be the generation of a strongly positive emotional state. It may also stimulate muscular development and the generation of new ideas. If these hypotheses are true, then play could function as somewhat of a fountain of youth. Panksepp says that play seems to be an index of youthful health. He also says that once adults have children they tend to become more playful again. Parents are generally more playful than nonparents. This makes sense as the new parents now must provide play experiences for their children. However, all attempts so far to increase play impulses in humans pharmacologically have not been successful. The search for a “ludic cocktail” may never work he says. This would be because it is possible that the PLAY circuit in the brain degrades throughout life. If that happens, then no amount of chemicals is going to be able to offset that.
I learned a great deal from Panksepp and I highly recommend reading his book. There is so much pseudoscience going on in the education and psychology fields that it is refreshing to read the work of an actual scientist rigorously defining what the current state of verifiable knowledge is in the realm of play. He is exceedingly scrupulous in pointing out exactly where the evidence ends, and the conjectures begin. It is rare to find that level of rigorous thought within education, psychology, or sociology professional literature.
Ted Talk by Panksepp: The Science of Emotions