Larking About: Play Science

What actually is play? And why is it so important?
26 August 2020
Presented by Katie Haylor
Production by Katie Haylor.


a photo of children smiling by a water fountain


This month and the next, we're having a laugh! We'll be getting stuck in to the science of play. What exactly is play? Why bother? And are we playing enough? Plus some of the latest from the world of neuroscience news with our local experts...

In this episode

Brain schematic

01:04 - The neuroscience of mediation

Our experts pick apart some of the latest neuroscience news...

The neuroscience of mediation
Helen Keyes, ARU; Duncan Astle, Cambridge University

It’s time to crack open some naked neuroscience news, with friends of the show Helen Keyes and Duncan Astle. And perceptual psychologist Helen is on a bit of a couples theme at the moment. Last month, it was brain synchronisation among couples who sleep together, and this month it’s about what’s going on at a neural level when we get into an argument with a significant other. Helen explained that there’s quite a lot known about good and less good ways of arguing with a partner in terms of conflict resolution, and that there’s also a fair bit known about the brain activity observed when we look at a partner, so this new study set out to marry up the two...

Helen - They recruited 36 heterosexual couples. So the couples were going to be involved in a conflict or an argument and half would have a mediator there. So someone who is trained in couples counseling to kind of structure the discussion and the argument, manage negative emotions if they're are getting out of hand, redirect people towards more positive and focused aspects of the argument. All couples would, first of all, complete questionnaires about the quality of their relationship and the level of disagreement within their relationship in general, the level of closeness and things like how mindful they were as people and their current mood. After that, one member of the couple - so this is before any argument - would go into the fMRI machine and be shown images of their partner's face. They would also be shown images of a stranger's face. Now the stranger would be the same gender as their partner. Their brain activity would be monitored in the fMRI machine while they looked at their romantic partner and the stranger, and they would also be asked to rate, using smiley and unhappy faces, how they felt after seeing an image of their partner or the stranger.

So following this is when the conflict happens. The couples engaged in basically an argument around some common point of conflict in their relationship, of which I'm sure there were many to choose from for long established couples! And this was either mediated by a professional for half of these couples, or it was an unmediated discussion. Now for the un-mediated discussion, a third person was present in the room, but remained silent and didn't structure or manage the conversation.

Katie - Is that to control for, say, if you wouldn't actually say something if someone else was there, you might be a bit embarrassed to, I don't know, shout at them or something like that.

Helen - That's exactly what it was for. Absolutely. The couples then completed questionnaires about how satisfied they were with the session. So did they feel the conflict resolution worked, was the conflict resolved, and how much disagreement was remaining? Then the person who initially went into the fMRI machine went back in and their brain activity was again monitored while they again saw images of their romantic partner and images of a stranger.

And what they found was behaviourally the mediation by the third party lead, not surprisingly, to a greater number of agreements between couples and in a reduced level of disagreement at the end of the conflict session. That's not that surprising.

Looking at the fMRI, what was interesting was that in general, across both groups, whether you'd had a mediated or unmediated conflict, there was decreased activation in the areas associated with romantic love, when you saw your partner's face. However for the mediated group, they had greater activation in one particular area of the brain that's really associated with our reward circuitry, compared to those who hadn't been in a mediated discussion.

Katie - Is this basically showing that mediation is very effective, actually?

Helen - It is showing that. Well it's showing a few things. It's showing that conflict with your partner is likely to be damaging to your representation of your partner or how you view them. But conflict isn't obviously always avoidable. And this is showing that mediation does seem to have a genuinely beneficial effect. That's not just an effect that we observe in terms of couple satisfaction, but we can also see on a neural level that people are still maintaining that reward response when they see their partner following mediated discussions, but not un-mediated discussions.

So if we're thinking about what's going to help couples in resolving conflicts, it's not just about somebody else being there, or having a shoulder to cry on ,or somebody listening, or a third party. It's about a trained professional able to structure that argument and that discussion and manage those negative emotions. That seems to be what's having the positive effect.

Katie - Which bit of the brain was it that was particularly distinctive in this?

Helen - It was the nucleus accumbens. And that's part of the basal ganglia near your hypothalamus. And it's largely associated with reward circuitry in your brain.

Katie - You mentioned all the couples were heterosexual. Were there any differences, say if it was the female person in the fMRI compared to the male person?

Helen - They didn't record any differences between genders, which was quite interesting in and of itself.

Katie - Because I was just wondering whether there would be any reason to suspect that you might see anything different if the couples were of other orientations?

Helen - I don't know. And I think it's a big pity in psychological research that we tend to use what is deemed to be the majority of the population. So for example, in this study even, they only use right-handed participants. And they only used heterosexual participants, and they only used people with normal vision or corrected-to -normal vision. It's an odd feature of psychological research that they think they will do the most benefit by testing people who are what they would consider to be the most normal representation of society. But it has obviously damaging effects and can really miss out on key areas of the population.

Katie - Duncan. What do you make of this study as Helen's explained it?

Duncan - I wonder about individual differences. I know they haven't got that many subjects in this study, but you can imagine that this is the kind of thing that would be really variable across individuals and across couples.

Helen - Yes, there is really well established variability in this, depending on individual differences, but also quality, length, strength, depth of your relationship. So, not in this study here, but in other studies on what areas of the brain are activated and how strongly when you view pictures of your romantic partner, these links are already really well established.


Cognitive neuroscientist Duncan Astle has been mulling over memory for us this month…

Duncan - Well, Katie, we're all getting older. Some days it feels more rapid than others, I'm afraid to say! But as we do age, one thing that changes is our memory. And one particular type of memory that changes is episodic memory. It's partly supported by a structure in the brain called the hippocampus. You have two, one in each hemisphere. And in this study, they were interested in how episodic memory changes as people get older and how the brain mechanisms that support episodic memory change as people get older. So episodic memory is what we call a type of explicit memory. Something that you're consciously aware of. And it's usually memory for events or episodes that have happened to you. For example, what you had for breakfast or where you parked your car. These memories can often last for relatively short periods of time. So a few minutes or hours or potentially a lifetime. And they can be distinguished from other types of memories, like semantic memories. So for example, that Paris is the capital of France. So at some point that was an episodic memory. Someone told you the first time, but over time it became a semantic memory. So those are the two main forms of long term memory.

Katie - I would assume that episodic memory gets worse with age. Is that just the stereotype or is that true?

Duncan - Indeed it does. But the question is whether the same brain mechanisms support episodic memory, as you get older as they do when you're younger. So in this study, they recruited two groups, a sample of people who on average are around 70 years of age. And another group who were on average around 23 years of age. They were both put inside an fMRI scanner and they were shown particular items like "barn" or "door", which they were instructed to remember. Then they had a period where they had to try and recall them. So they would show a label and they had to bring to mind the item and rate how confidently they were able to retrieve it. And then later on, they actually showed people examples of barns to see if they could pick out the exact one that they'd been shown.

All the time, they're scanning their brains. And in particular, during the retrieval phase, they looked at brain activity within multiple different brain areas. And in particular, they looked at activity within the hippocampus and also the prefrontal cortex, which is right at the front of the brain. And what they wanted to explore is whether or not the prefrontal cortex is kind of kicking in as you get older, to support your failing hippocampus.

Katie-Oh, okay. So some sort of compensation going on?

Duncan - Exactly. So some people have shown already that on these kinds of tasks, older subjects have increased activity in the prefrontal cortex relative to younger subjects. So one argument that's been made is that that's because they're doing it to compensate, but it's quite hard to demonstrate that explicitly. And that's what this study set out to do.

So to start with, they're able to show the older individuals did indeed show stronger prefrontal cortex activation, but they went a step further and they constructed what's called a connectome. So that's where you estimate the extent to which different brain areas are functionally connected with other brain areas. And what they're able to show is during this retrieval phase, the older subjects' prefrontal cortex was more strongly integrated with other nodes within the brain. So it wasn't just more active. It seemed to be more plugged in to activity elsewhere in the brain. And those subjects for whom the prefrontal cortex was most plugged in, were the subjects who did best on the task. So that would be consistent with the idea that it's somehow compensating.

Katie - Does that have any additional relevance? Because I tend to think of memory as something that fails or sort of weakens relatively as we get older. But the brain being more connected with other bits sounds like an advantage.

Duncan - Ah, we'll get onto that in a moment, but yeah, you're right. The final piece of the puzzle is to demonstrate that you get these sorts of systematic increases in prefrontal cortex integration, as you get drops in hippocampal integration. And that was what they finally showed, is that you get this sort of prefrontal cortex stepping in, in the older adults, more so as the hippocampus sort of steps out as it were. Which is sort of the direct evidence of the kind of functional compensation account.

Now you are asking about "surely, that's a great thing?". It's a compensation strategy that probably works to some extent early on, but if you were to start to experience much more progressive hippocampal damage, for example, like say in the case of a dementia, then what you might well find is that as that damage becomes more extensive, the prefrontal cortex would no longer be able to compensate.

Katie - Does this transition on which bits of the brain are being relied upon for this memory, does that have any practical difference to the quality of the memory? Would you be able to tell?

Duncan - I think in this task, probably not, cause it's such a basic task. But one thing you notice with subjects who do have episodic memory problems - so for instance, mild cognitive impairment or something much stronger like Alzheimer's - is in the very early phases, what subjects or what patients are able to do is sort of fill in the gaps. So the prefrontal cortex we know is an area of the brain that's specialised for highly flexible thinking and problem solving. And so what people are possibly doing is rather than sort of remembering the exact episode, they're maybe remembering little bits and then they are sort of filling in the gaps with this sort of more flexible brain system, the prefrontal cortex system. And so I suspect that ultimately it will result in a loss of quality of the memory, or it's indicative that the quality of the memory is lower. And that's why the prefrontal cortex is having to step in.

Three brown kittens sitting side by side, looking to their left

Rough and tumble: animal play
Sarah Heath, Behavioural Referrals Veterinary Practice

Why is play so important for the wellbeing of companion animals, like cats and dogs? Katie Haylor spoke to vet and behavioural medicine expert Sarah Heath...

Sarah - Play is hugely important for development as it is in human children. They’re not only rehearsing actual physical movement and developing physical responses. But also play is important in terms of emotional development. And when you’re playing in this intra-species way that you’re seeing in your kittens when they’re rough and tumbling around, they’re also learning how to read each other’s body language effectively. Social skills, through play. And also cognitive health - they’re learning what responses are successful. So play is crucial in their physical, cognitive and emotional development.

Katie - Watching my kittens grow, it’s evident to me that play can fall into different types, so first up I asked Sarah to break this down.

Sarah - So we talk about locomotory play, object play, social play. We can also define it in terms of its underlying motivation. And when we do that, we tend to use the concepts of Jaak Panksepp the Estonian biologist, who worked on affective systems in human animals and nonhuman animals and identified seven different systems, which are responsible for behavioural output. And of the ones that are important for play context, we've got the desire-seeking system, which is related to object play. And then we have the social play system related to that physical intra-specific form of play, where there's rehearsal of behavioural sequences, particularly threat sequences.

Katie - Ah, so would that be the difference between one of my cats pouncing on a ball and chasing the other one around?

Sarah - Absolutely. Yeah. So when the cat is playing with a toy, and cats are a lovely example of disaster seeking object play - predatory behavior is also desire-seeking motivated - when you see cats playing with objects, they are engaging in the same behavioural sequences as they use in predation. So we love to see that when we're playing with an object with these fishing rod type toys, where you stimulate the same motivation that you have with a bird flying through the air or a mouse running along the ground, pouncing, stalking ambushing. And then we also have the social play between cats. And social play is most effective and successful intra species because of the fact that social play is very contextual. So your kittens are rolling around with each other, or your puppies are engaging in this rough housing type play. And social play is important in rehearsing behaviours that are specific to the species. So it's important in learning how to communicate as a dog or as a cat. But also, very importantly, for rehearsing threat, both in terms of giving threat successfully, but also responding to threat successfully.

Threat is actually quite a difficult behaviour to reverse. As soon as you threaten another, then you have the risk that they're going to be motivated by fear-anxiety, and worried about the interaction. But if you have a safe context in which both individuals know that the threat that's being delivered is not real, then you can rehearse how to respond to it and how to perfect the giving of that threat without there being any detriment.

Katie - Is it possible to map those kinds of play behaviours onto different bits of the brain?

Sarah - This is an area that is massively increasing in the non-human animal world. A couple of areas that are very active at the moment in research are cognitive research in non-human animals. So trying to understand more about cognition. But also emotional research. So the seven systems I was talking about that Panksepp talks about in his research, looking at how we both protect ourselves emotionally, those are the negative emotions in Panksepp's model, fear, anxiety, frustration, panic, grief, pain. And then on the engaging side or positive emotions, we've got the desire seeking emotional system that drives us to gain access to those things we need in order to survive, the lust system, which kind of does what it says on the tin, the social play system with this ability to rehearse species-specific behaviours in a play context. And then you have the care system, which is the emotional drive that leads us to nurture others. And by us, I mean mammals, not just human animals. So those systems are the ones that we're talking about with Panksepp's model. And I think this is an area where we are really only beginning to start to put together the information we have about emotional systems. We know of course, about the role of the amygdala in the production of those responses. But actually getting to the point of dividing that up into specific behavioural responses is an area of quite active research.

Katie - Sarah explained that, unlike dogs and humans, cats are not what’s called socially obligate. So whereas a pooch needs social play throughout life, once cats get socially mature, between 2-3 years old, cats have a lowering of thier motivation for social play once they’re grown up.

And understanding differences like these is important. As Sarah told me the key to playing well with your pet is to respect that, as they are different species to us, they may play differently to us. It may sound obvious, but they're not little people, and playing with a dog like you're a dog, Sarah says, can lead to problems.

Sarah - The way we do that most successfully with non-human companions is using toys. And actually thereby creating object play rather than social play. So social engagement during play is different from social play. You can play socially with an object. So there are forms of team sport, which are object play, but have people involved. So tennis, for example, you're playing with objects, but you're playing with objects together. So there is a social component, but that social component is actually desire-seeking motivated, not social play motivated. You're still getting good social interaction, but there's not the scope for the confusion and the perception of threat from the animal.

Katie - Are there any consequences of lack of play in cats and dogs altogether? Just not enough?

Sarah - Yes, absolutely. And let's stick with your beloved species, the cat for a minute, they give a lovely example of this. They engage in predatory behaviour in order to actually kill, consume, eat, etc. But they also have a requirement for playing in that style. So it's not just predation for the purpose of actually getting to the point of killing and consumption, but also as a way of rehearsing the skills that are necessary for that. So if they don't have enough outlets for predatory style, so desire-seeking, motivated, object play, then we have a couple of possible consequences. One of them is that they lose confidence. And we can use object play in cats to boost self-confidence as well. But remembering that the object play is different again for cats than it is for people in things like duration. So cats engage in object play for very short bursts, often between one and five minutes. People can interpret that as disinterest. Whereas actually, yes, he is interested. That's good quality play for a cat, one to two minutes! And so we find that unfortunately, kittens who are more motivated to engage in the play and sometimes do it for longer, and you tend to go, "Oh, they love playing. So we'll play with them". And when they get to be adults and they're only doing it for one or two minutes, they go, "Oh, she doesn't really like it. I won't bother anymore". And so we see a lot of adult cats who don't actually get sufficient opportunity for play.

Katie - Do you think that play can tell us anything about the emotional relationship, or maybe bond is a better word, between pets and their owners? Because of course it's non, I guess the majority of it, is nonverbal.

Sarah - Yes. The quality of play, definitely, has some link to the caring relationship. The care or nurturing part of the relationship. And that's why, again, play in adult dogs and cats is different. Because the basis of the relationship is also different in an obligate social adult and a non obligate social adult. You're going to have a different basis of the relationship as well. So a human-dog relationship into adulthood has a caring component, a nurturing component to it, which is not as evident. I'm not saying it doesn't happen, but not as evident in the adult-cat relationship. One of the things we do with our domestic cats, particularly here in the UK, is that we pre-puberty newter them. So that influences their emotional development of course, and their transition from kittenhood into adulthood emotionally. So we're going to have some extension of juvenile behavioural responses into adulthood that you wouldn't see if that cat was entire and was not in a domestic environment. So there are some complications, if you like.

So, yes, it is an indicator of the relationship. And it can also be a way of improving the relationship, particularly in dogs. One of the things that's also important to remember is that when play is appropriate, as in it's stimulating, or it's being motivated by the social play system, the desire-seeking system. We also need to remember that we have to consider the aspect of arousal as well, quantity of emotional input, that's happening for that individual during play. And that can be somewhere where we can get problems, because if you have a low level of emotional capacity, and the play you're engaging in is highly arousing, that can actually be emotionally detrimental. Not because it's a protective emotion, it's still an engaging emotion, but because the arousal level is incompatible with the individual's emotional capacity.

Katie - You mean if, say, it's just all a bit too overwhelming for your pet and they, perhaps they freak out and end up scratching you or something like that?

Sarah - Yeah. Two things can happen that make it then become detrimental. Yes, there is a situation where you can be engaging in a play interaction, which is either diesire-seeking or social. And then for some reason, either you get overwhelmed by the level of arousal and you no longer have the capacity to cope with that. And therefore you kick into feeling overwhelmed and therefore protective, and then you become repelling in your behaviour to make that stop. Or, perhaps more commonly in the play context, you actually start to get frustration building up, because the play is not succeeding for some reason, maybe because the communication breaks down between the two individuals and the play is not successful, then frustration would become involved. And frustration is also characterised by confrontational behaviour. So you can get actually confrontational or repelling behavioural responses coming out of what was a play context.

Katie - So my kittens are playing together a lot right now, and this will likely change as they get older. But what about elderly pets. Do they still need play? And how do you cater for this, whilst accommodating age related changes?

Sarah - If we sit with the social obligate dog for a minute, and then we need to think about whether there are physical health issues, which mean that mobility associated with play may be difficult. And therefore we need to make sure they can play without moving about too much or without putting strain on joints, etc. We also might need to think about cognitive decline and therefore that may have influenced their ability to read context. So in dogs particularly, they may find it more difficult to engage in social play as they get older if they have some form of cognitive compromise.

And then in cats, we need to remember that probably what we're looking at for them is more object play than social play. And we need to remember the fact that their bodies can be struggling. Particularly we know osteoarthritis, degenerative joint disease, is incredibly common in cats over 92, 93% of cats over the age of 12 will have some form of osteoarthritis, degenerative joint disease. So therefore are likely to have some element of chronic pain. So yes, the way we play with them is going to be different. Because we want to engage them in object play to have those senses, their sight and their hearing, still engaged, but not wanting to instill high levels of actual physical movement.


Hi there, if you click on the chapter title, so "The neuroscience of mediation", and then scroll down to the bottom of the interview, the references are listed there.

Where can I find the references for the two studies mentioned at the beginning?

Add a comment