Loading
From the impact of giving a child a time out, to the latest research on punishment-free parenting, Amanda Ruggeri investigates the fascinating science of discipline.
Y

Your toddler smacks another child at play group, upset that she won't share her toy. Your pre-schooler needles you at the store, whining that he really wants that sweet. Your 12-year-old refuses to set the table, storming off to her room and slamming the door.

Ask any parent, and they’ll tell you that dealing with situations like these can be among the most challenging – and common – parts of parenting. If the frustration of the moment itself weren’t enough, there’s the difficulty of how to respond. Do you give a time out? Threaten to take away privileges? Negotiate? Or do you follow the approach being promulgated by an increasing number of advocates, which holds that connecting with a child in their moment of distress, not punishing them, leads to calmer, better-behaved, and emotionally more in-tune kids?

"The word 'discipline' has been misconstrued as 'punishment', meaning inflicting pain as a consequence of doing something," says Dan Siegel, clinical professor of psychiatry at the UCLA School of Medicine and the author of six books on parenting, including No-Drama Discipline. He points out that the word "discipline" comes from the Latin word disciplina which means instruction or teaching.

"Parents say, 'Parents should be good disciplinarians.' We go, 'Yes – and a disciplinarian is a teacher.'"

Like any parenting decision – from whether to sleep train to the dilemma of screen time – how we discipline (or "teach") our children is deeply personal. Our beliefs around discipline have been shaped by our culture, the attitudes of those around us, how we were raised, even our current stress levels. Whether we think discipline is needed at all is also situational, depending even on which rules we set: a three-year-old told not to leave the playroom is more likely to "disobey", for example, than a three-year-old who is allowed to come in and out as she pleases.

Even beliefs that are a no-brainer in some societies are unusual in others. "The Anbarra child hears of no rules and receives no punishment," one anthropologist noted of the Aboriginal tribe. Other Aboriginal approaches to discipline include the idea that "the child has the ultimate choice to obey or not and adults are not overly upset if the command is not complied with". Rather than with rewards or punishments, children learn how to behave "through trial and error over a period of years". The Sámi, an indigenous group spread across the Arctic, espouse a similar parenting philosophy, letting children make their own decisions about even when to eat and sleep. Instead of punishments, there are intricate, unspoken rules and communal activities that nudge children towards desired behaviour, such as going hunting or fishing together.

In other societies, parents take a harsher approach. One Unicef report found that more than 90% of children in countries including Ghana and Egypt have experienced either physical or psychological aggression as forms of discipline in the last month.

One thing we know is that every caregiver will, at some stage, find themselves dealing with a child not doing as they're told. Anywhere from 25% to 65% of parents say that their children are at least sometimes noncompliant, with 1% to 9% saying this is a frequent or severe problem. This peaks in toddlerhood, after which children start to use tactics like negotiation to get what they want – underscoring the idea that in younger children, at least, disobedience and pushing limits is a part of normal development.

But what does the science say about how to respond? And is there really a way to guide children towards kind, thoughtful, cooperative behaviour that doesn't involve punishment at all, as some experts argue?

Some forms of punishment have been shown to worsen, not improve, behavioural problems (Credit: Getty Images)

Angry parent, angry child?

What the vast majority of scientists, paediatricians and psychologists now agree on is that harsh parenting tactics and corporal punishment, including spanking or "smacking", are unhelpful – and can cause harm to the child even when parents think the spanking is mild and justified.

The American Academy of Pediatrics changed their policy statement in 2018 to warn parents against spanking, and its use at home is now banned by 63 countries and territories ranging from Wales to Colombia. (Spanking at home remains legal in the US, Canada and Australia while in the US, 19 states also allow its use in schools, something banned in nearly every other nation.)

But despite signs that such bans, along with broader social changes, have helped reduce spanking, many parents do it. In the UK, 42% of parents said they've smacked their child over the course of the previous year. In the US, almost one-third of parents reported spanking their one-year-olds over a given month, while one study in North Carolina found that 70% of the mothers of two-year-olds – and 5% of the mothers of three-month-olds – admitted to doing so in the previous year. One 2013 poll found that 78% of US parents thought spanking could be appropriate – though only 22% of US paediatricians do.

Yet research suggests that physical punishment only makes the problem worse.

"It turns out the kids who are spanked get more aggressive over time. If it was working, that would go down. But it doesn't," says Elizabeth Gershoff, a professor of human development and family sciences at the University of Texas at Austin, who studies the effects of corporal punishment on children. "It actually is doing the opposite of what parents want.

"We've shown that over and over by now."

One robust review of high-quality research co-authored by Gershoff found spanking in childhood was associated with the child being more aggressive, antisocial, and having poorer mental health, with a worse parent-child relationship and a higher likelihood of mental health problems or antisocial behaviour as adults. It also found children were more likely to have lower cognitive abilities and lower self-esteem – and that spanking was not associated with any improvement in child behaviour.

"No study has found physical punishment to have a long-term positive effect, and the vast majority have found negative effects," concluded another review of 25 years of physical punishment research.

It's not just physical punishment that can backfire, but also "psychological aggression" like  shouting and name-calling. Such verbal harshness towards a child at age 13 predicted more child behaviour problems and depressive symptoms over the next year, for example, even if parents were warm the rest of the time. Harsh parenting tactics including spanking, raising one's voice, or getting angry may even change children's brains, activating a pattern of heightened fear and vigilance even when there is no actual threat.

Still, as with spanking, such tactics are common. In the US, one survey found 98% of parents of five-year-olds report having engaged in some form of psychological aggression in the last year. Unicef has found that nearly seven in 10 children aged two to 14 worldwide were "shouted, yelled or screamed at" in the last month. Four in 10 were called "dumb or lazy".

If attacking the child physically or verbally doesn't improve their behaviour and can have such negative consequences, why do parents do it?

Some may be re-enacting the way they themselves were brought up. Others may be reacting out of anger or helplessness, unable to control their own emotions. And some may simply think it's the only or best way to deal with a misbehaving child.

As science has shown these tactics to be less than ideal, other forms of punishment have come to the fore – such as "time out" (withdrawing attention from a child for a certain amount of time) or "taking away a privilege" (such as not letting the child watch their favourite TV programme).

Some parents may react harshly out of anger or helplessness (Credit: Getty Images)

The time out controversy

According to some experts, there are other, more effective ways to encourage good behaviour than any punishment at all (more on that later). But even if parents are told this – and believe it – it doesn't always change their response. Punishment isn't always about changing a child's behaviour, but stems from a parent's desire to express disapproval, satisfy their sense of justice, or even just vent their feelings.

As a result, it's unlikely we'll ever get to the point where most families practise "punishment-free parenting", no matter the potential benefits to the child, says Alan Kazdin, professor of psychology and child psychiatry at Yale University and director of the Yale Parenting Center in New Haven, Connecticut.

"You could build the behaviours you want without punishment. But at the same time, one has to be realistic. Parents are not going to abandon punishment," he says.

"So as professionals we say to ourselves, what, within the research, is the most mild punishment that is as effective, gives parents a tool, and avoids all the negative effects of hitting, shouting, screaming? A brief time out is one. And taking away a privilege is another."

But researchers define time out fairly narrowly, and not in the way that many parents practise it.

For time out, "the child is simply told that their behaviour is inappropriate, and that the consequence is time out", says one review. While most parents think of it as happening on a "naughty step" or in a child's room, researchers define time out not as where it takes place, but as a stint of time where the child is not given attention. Time out doesn't even have to be away from the caregiver, experts say; it can be quiet time in the same room.

Importantly, the parent doesn't add other, more aggressive punishments – so no yelling, calling the child "naughty", or even telling them to use the time to think about what they've done.

"We want time out not to be shaming, but just boring. We want it to be so much more boring than whatever else is going on in the environment," says Corey Lieneman, a postdoctoral fellow in child and adolescent psychology at the University of Nebraska Medical Center and co-author, with West Virginia University child psychology professor Cheryl McNeil, of the forthcoming book Time Out for Child Behavior Management.

"If you add things on top, like yelling, or allowing your child to escape the time out chair repeatedly, then it becomes exciting for the kid… And I think it's a common misconception to think that children have to sit in time out and think about what they've done, or they have to feel badly about what they've done. It's more about just seeing a difference: it's fun when I listen, it's boring when I don't."

In terms of length, a time out should last for one to five minutes – there's no evidence that a longer stint has more of an effect.

Time out also should be used sparingly, researchers say. It should be used for clear disciplinary situations, not ones that are "emotionally upsetting" for the child, or where their attachment system, and need to be close to a caregiver, are activated. And it should be in the context of an otherwise positive parenting relationship.

If parents are trained to use time out in this expert-approved way, does it actually work?

What punishment does, at best, is immediately suppress the behaviour. The trouble is, the research is unequivocal: what happens is the behaviour returns at the same rate – Alan Kazdin

There is relatively little research on time outs and similar punishments, says Gershoff. This makes it a more complex question than, say, spanking. In particular, there are few robust RCTs – randomised controlled trials, the gold standard of scientific research. In the first known review of time out's effectiveness, published in 2020, for example, the authors found just six relevant randomised controlled trials done between 1978 and 2018. All had small sample sizes – a maximum of 43 participants.

From the available data, they concluded that time out improves children's behaviour in the short term. One meta-analysis of programmes that help teach parenting skills, for example, found that if parents participated in a programme that included time out, there was a greater average improvement in children's behaviour than programmes that did not include time out. 

The evidence for any long-term benefit, however, is thinner.

"What punishment does, at best, is immediately suppress the behaviour. It stops it – much of it by a startle reaction, but it stops it," says Kazdin. "The trouble is, the research is unequivocal: what happens is the behaviour returns at the same rate."

One of the few long-term studies done on time out followed children from ages three to 10. Once everything was taken into account – such as whether a parent used other forms of physical punishment – there were no differences in behaviour between the children who received time out and those who did not. The researchers interpreted this to mean there was no evidence that time out harmed children. But it did not seem to help them long-term, either.

Other studies suggest that time outs may help when used as part of broader programmes that support parents and teach them non-violent, gentle ways to encourage good behaviour. But it's not always clear if the time outs themselves were beneficial, or the gentler approach as a whole, which also included techniques like praising the child more, and listening to them.

"It is really hard to isolate any one strategy," says Lieneman. "It'd be like isolating one vitamin: is one vitamin helpful or not helpful? We have a lot of evidence that [time out] is helpful, but it's mixed in with all of the other health behaviours, so all of the other positive relationship aspects."

The child's personality may also play a role. Robert Larzelere, professor of human development and family science at Oklahoma State University and one of the co-authors of the 2020 review of time out's effectiveness, looked at both the short and longer-term effects of various disciplinary responses including time out, privilege removal, and reasoning.

He found that, for the 12% most defiant toddlers he studied, punishments and warnings improved behaviour over a two-month period, but only if mothers used this tactic relatively rarely – in up to 16% of their total disciplinary responses. "The most effective parents prefer to use other tactics, such as mutually acceptable compromises and age-appropriate reasoning, but will back those methods up with a single warning followed by a timeout if toddlers persist in being defiant," he says. "Then these defiant toddlers learn to pay more attention to the reasoning that they usually ignore and to the single warnings."

For less challenging toddlers, who made up 35% of the children he saw, punishments and warnings backfired: they worked immediately, but at the two-month check-in, they had increased the child's behaviour problems. And on average, for the remaining majority of children in the middle – 53% of the total – infrequent timeouts had no effect on later behaviour, positive or negative.

Adding to the complexity, it can, of course, be difficult for a parent to know which of these camps their child falls into.

Some experts don't recommend using time out at all, unless it is as a punishment-free way, such as separating two fighting children so they can take a break, calm down, and then regroup.

One criticism is that in practice, parents don't use time out in the way it is recommended. One study of the parents of 400 US children found that while more than 75% of parents reported using time out, 85% of them were using it in a way that ran counter to the evidence.

"Yes, the research way of using time outs can teach a child skills of self-regulation – but the common use of time out is, in our experience anyway, not done the way the researchers say it should be done. And instead the word is used – 'Take a time out!' – in fury and exasperation," says Siegel. "It's used as a way of punishing a child, meaning to inflict emotional pain on them."

Some also have concerns over the potential risks of using a punishment like time out.

"Although we have, to a large extent, moved on from hitting – thank goodness – we've in many cases replaced it with other things that hurt the child psychologically, affect that parent-child relationship, erode trust and contribute to the child feeling completely powerless and helpless," says Joan Durrant, child-clinical psychologist and professor of community health sciences at Canada's University of Manitoba, as well as the creator of the programme Positive Discipline in Everyday Parenting.

"If I slept in or something, and my husband's reaction was to say, 'You can't drive your car for a week', or 'Go sit in the bathroom for an hour' – it's ridiculous, when you think of applying that to ourselves. But that's what we do to children all the time."

One concern, she says, is the lesson we do teach children by punishing them: that they must do what an authority figure tells them, "or else". "It sets children up to learn to be submissive, which is extremely dangerous. To do what adults tell them to do right now – the first time ('I don't want any back talk!')," she says. "So when my coach, or my priest, or my teacher wants to coerce me into something, it's a whole lot easier. I've learned that I have to submit."

Indeed, a child's overcompliance – if inspired by fear of an authority figure, such as a fear of punishment – is itself considered a mental health problem.

But for non-physical punishments, if given rarely and in the context of otherwise warm, responsive (not authoritarian or overcontrolling) parenting, those risks remain theoretical.

The 2020 study that followed toddlers until they were 10 didn't just find that being given time out made no difference to their behaviour in the long term; on its own, time out also had no impact on their risk of having a poor relationship with their parents, behaving poorly or being anxious or depressed. That did not hold true for physical punishment like spanking, which was associated with increased rule-breaking and aggression as the children got older.

While this study was widely reported to prove that time out doesn't cause any harm, however, it had flaws. In particular, almost half of the families dropped out of the trial over time, which could skew results: if the families whose children really struggled with their behaviour long-term were those who found it hardest to make it to the study's follow-up assessments, for example.

Ultimately, it's not so much that a punishment like time out is risky as that it's a missed opportunity, says Siegel, who – with his co-author Tina Payne Bryson – earned so much notoriety in 2014 for a story on time out, he wrote a follow-up to clarify his position.

"Rather than focusing on a worry about what time outs do, I would say, 'What do I want to teach? What's the benefit?'" says Siegel. "Children learn from what we do. They learn from how we are."

"So the question then is, what do you want your kid to learn? You want your kid to learn that you don't have skills that allow you to stay calm and clear, and instead, that you're exploding out of frustration, and feeling incompetent? Or do you want them to see that you have skills that, whatever happens, allow you to stay present?” says Siegel.

Harsh parenting tactics can cause harm to the child even when parents think they are mild and justified (Credit: Getty Images)

Emotion coaching

Even if we don't want children to obey every single command, and while there's a plethora of research indicating that overcontrolling parenting can be harmful for children, consistent misbehaviour comes with its own long-term risks, at least in older age groups (not, for example, toddlers). One study found that 13-year-old children whose teachers and peers said they misbehaved were more likely to commit criminal offences by age 27, for example.

But there are punishment-free ways to encourage good behaviour, some experts say.

One pillar is to have a little more empathy. Children are told what to do an awful lot – one small study found mothers gave their children an average of 41 instructions per hour (the researchers were looking at children who were referred to a family therapy programme for not complying with their parents' instructions, so this might not represent every family). But they aren't as developed as adults in terms of processing that information (it takes around 5.5 seconds from hearing a parent's instruction to a child's complying). Nor do they always know how to do what they're told – which may be why talking a child through a task leads to their complying more than simply telling them what to do.

Think of how we would feel in some similar situations to those in which we often punish children, suggests Durrant.

"Let's say that I really want to build a cupboard, so I go to somebody who knows how to do this,” says Durrant. “He hands me a power saw and says, 'The first thing you do is saw this wood. Here's the line – cut along that line.' I've never held this thing before. I flick the switch and it goes all over the place. And I damage the wood," she says. "And then they come back and say, 'I told you what to do. How could you create all this damage? Look what you've done. Go to your room for an hour.' I go to my room. I come out. They hand me the saw and say, 'Cut along that line.’ But I've learned nothing. All I've learned is that I should be ashamed. I should have known better. It's all my fault."

One alternative that has arisen is emotion coaching, or helping children understand and express their emotions. This approach hinges on the idea that whining, a tantrum or even hitting isn't merely misbehaviour to be "corrected", but a sign that a child is emotionally dysregulated. Since children (like adults) cannot learn when they're distressed, its advocates say, they need to be brought back to a state of calm before being taught a lesson will have any effect. (Read more from BBC Future about the healing power of emotion coaching.)

Children behave in socially appropriate ways, when they feel connected. When they feel loved, respected, supported - Sophie Havighurst

"Children can't work through solving problems and even hear you giving them guidance about what else might help if they're still really emotional. And yet that's when we often try and give instruction and guidance about behaviour. So our whole approach is that you've got to wait till your child settles and calms, and you're connected," says Sophie Havighurst, professor of child clinical psychology at the University of Melbourne.

"Children behave in desirable ways, or socially appropriate ways, when they feel connected. When they feel loved, respected, supported and validated in their emotional world."

Research also shows that the more emotionally in-tune with their children parents are, and the less disapproving and critical of their child's feelings, the better children are able to regulate their emotions and behaviour. Children who receive "emotion coaching" are less physiologically aroused (an indication that they are better able to regulate their nervous systems), and even are less likely to get sick, than those who do not.

One study of 94 children showed that there may also be a self-reinforcing cycle between the way parents perceive their children, and how they respond to them: if parents said that they were likely to minimise their child's emotional reactions, the children were more likely to exhibit problem behaviour later – and if young children were perceived by their parents to have poorer emotional regulation, they were more likely to be punished by their parents as they got older.

While there have been a number of RCTs on the impact of emotion coaching, or of similar techniques that replace punishments with more of a problem-solving approach, they have been based on relatively few datasets. But the available evidence shows that the "positive parenting" approaches worked better than nothing when used with children who had behavioural problems.

There also were signs that how effective they were depended on the individual child. For children who had depression or anxiety in addition to misbehaviour, for example, one study found that collaborative problem solving was more effective than conventional methods that included time out. Another study, co-authored by Havighurst, found that emotion coaching worked better than behavioural parent training for eight- to nine-year-olds and for children whose parents reported having poorer psychological wellbeing. Meanwhile, behavioural parent training worked better than emotion coaching for younger children and for parents who had better psychological wellbeing.

In general, however, the "alternative" approaches – which eschewed not only spanking but disciplinary actions like time out – were just as effective as those who used time out.

One such example is the emotion coaching programme called Tuning into Kids, designed by Havighurst and her co-author Ann Harley. It encourages parents to respond to the emotion underlying a behaviour – including by connecting with the child, communicating empathy, helping the child understand their emotions, problem solving, and setting limits ("I know you're frustrated, it's really hard when you don't want to put your things away. Let's see if you can stomp your feet really hard to let out that steam." Later, after the child has calmed: "It's not easy when you don't want to stop playing. I know it is hard for you. I wonder what you can do next time instead of hitting Daddy?".)

If the child's anger or misbehaviour escalates, parents are discouraged from using a punishment like time out. Instead, they are told to use a "time in": staying with a child who is angry or distressed and providing a quiet, calm presence with minimal talk, and a rub of the back, if desired.

"We need to teach kids to learn that their emotional needs lead to connection – not disconnection, or punishment," says Havighurst. "One of the things that we support is this idea that when kids are very angry, they often fear rejection. They fear abandonment. And the activation of attachment is that they still need to be attached, even though they're very dysregulated," says Havighurst. (This only works, she adds, if the parent isn't very angry or distressed themselves – if so, they may need to step out of the room to avoid verbally or physically harming their child.)

At a follow-up 10 months later, the parents who learned about emotion coaching were less likely to dismiss their child's emotions and had greater empathy for their children. They also reported fewer "negative expressiveness" (like fighting) in their families. Meanwhile, parents and teachers reported "significant" changes in how the children behaved, as well as in their understanding of emotions. Of the children who had clinical behaviour problems such as hyperactivity or oppositional defiant disorder at the start, 27% no longer had these problems at follow-up with a clinician, compared to 18% of those in the control group.

There are caveats. As is common with studies like this, there was a high drop-out rate – follow-up data wasn't available for 31% of parents, 22% of teachers and 20% of direct child assessments – which could, again, make results seem either better, or worse, than they were. (Havighurst points out that in other studies she's done of Tuning into Kids, which had similar results, the retention rate was much higher.)

And, like behavioural parent training, emotion coaching is a suite of interventions, not simply "time out versus no time out". Any positive results might not have had anything to do with dropping time out. They could have been, for example, from simply teaching the parents about their own emotions, helping them to become more regulated and calmer.

This may be why the research shakes out the way it does, says Larzelere. "Those programmes emphasise different aspects of positive parenting. So the better, more skilled way that parents can use those kind of skills the better for the child," he says. "But the evidence isn't against time out."

Researchers who advocate for punishments like time out as part of a positive parenting relationship also add that it's not an either/or proposition. There isn't any reason that parents can't problem solve, express empathy or talk through emotions with their children as well as issuing a punishment, they say – which some approaches explicitly advocate.

The idea, ultimately, is to provide the parents with a way to respond that is safe for the child. Telling them not to punish the child at all, Kazdin says, risks leaving them at the mercy of their impulses.

"We want parents to have tools in their pockets. Because once they're impulsive, it's a smack. It's a shouting, it's a shake, it's a scream, it's something," says Kazdin. "You don't want that kind of stuff."

Durrant, who believes punishment is never productive, also has concerns about escalation, but from the other side. Take time out. "It often gets very, very ugly, where parents are holding the doorknob, the child's pulling on the other side, there's all kinds of screaming," she says.

"What we're doing is giving parents a recipe for violence. There's going to be yelling, there's going to be hitting, there's going to be grabbing that child and forcing them down on that chair. Why are we given a prescription for coercion, when we know that every human being resists coercion? And then the more they push back, the more it escalates."

Punishment-free parenting advocates also point out that when we talk about a behaviour intervention's "effectiveness", we normally mean, "Did it make children behave better?" That's what research studies usually look at, too. That may make sense – but it isn't the whole story.

"You get rough measures, like, Is the child behaving better?" says Havighurst. "But what happens inside if a child has a parent being much more emotionally responsive to them? What does that build internally in a child? And that's the part I think it's really hard to measure."

Siegel agrees. You can change a child's behaviour by focusing just on the behaviour, he says. But if you look at what's underlying it, helping your child to hone skills of insight and empathy, you can go beyond that.  

"That, yes, allows a kid to guide their behaviour, but it's much, much more than that. It allows them to know themselves and others in a way that allows them to thrive, not just survive by regulating their behaviour," he says.

"I'm interested in something much deeper than the behavioural outcomes. I'm interested in the mind of the child."

Warm, positive parenting can encourage children to express their emotions and regulate them, experts say (Credit: Getty Images)

Family Tree

This article is part of Family Tree, a series of features that explore the issues and opportunities that families face all over the world. You might also be interested in other stories about children's behaviour and development:

You can also climb new branches of the Family Tree on BBC Worklife – and check out this playlist on changing families by BBC Reel

Common ground

They may come at discipline in different ways. But experts on both sides of the equation overlap in many of their findings about what works best for children.

For one, they agree that not only is harsh or authoritarian parenting not optimal – permissive or "indulgent" parenting isn't, either. Permissive parenting, which is normally defined as being supportive but with low levels of control or expectations, tends to be a mixed bag. Studies have found that it's associated with high levels of self-confidence in children, but also with more substance abuse and behaviour problems, for example.

The style of parenting that consistently comes out on top is "authoritative", where parents have high levels of responsiveness, but also high levels of demandingness.

Indeed, despite the stereotypes around "gentle parenting", a punishment-free approach doesn't mean doing away with boundaries, Havighurst says. It should be authoritative, not permissive.

"But there's a way you can do that without using dominance, or withholding or punishing," she says. "It's not about the child doing everything the child's way. It's about coming to some balance. Children learn best when you respond to them with empathy before guiding them in their behaviour."

On the other hand, researchers who are in favour of consequences like time out, like Larzelere, say these consequences need to be used rarely and in the context of otherwise warm, positive parenting.

"It'd be the goal to be as positive as possible," he says. "Parents shouldn't do those things as much as possible and should emphasise the positive relationship with their children." Negative consequences, he says, should be a "backup" to teach a child to cooperate after other attempts have failed, rather than a go-to response to misbehaviour.

Most researchers also agree that not only should punishments not be the crux of any parenting strategy – but they're not the best way to teach a child a lesson, either.

"Spanking and things like are very attractive for many parents because they often get a quick reaction. You feel like you've dealt with it, and then you can move on. Versus taking time to really talk with children and explain things to them," says Gershoff. "Time out and spanking are these kind of quick fixes that don't really deal with the underlying issue, and they don't help children learn."

Reasoning, rather than simply commanding, is something you can do with children from a young age, Gershoff says – it's something they really start to understand when they're two or three.

In fact, Larzelere's comparison of disciplinary tactics found that for the majority of toddlers, the best way to get a toddler to comply in the short term is to offer alternatives – reasoning works best for children who are whining or negotiating.

Again, however, it depends on the child. In the longer term, offering alternatives decreases misbehaviour for the 7% least challenging toddlers. For the 20% most defiant toddlers, offering alternatives too often increases misbehaviour – although frequent reasoning, two months later, had helped them behave better. And for the remaining majority of children, offering alternatives had no clear effect either way on behaviour two months later, on average.

"We use mild punishment, because parents are going to punish. And mild punishment is only better than regular other punishment, because the side effects are so few. But it's still not going to teach the child what to do," says Kazdin.

Among other elements, his approach includes a specific sequence of practise and praise. Say a child is throwing tantrums while hitting you. You should tell your child to "practise" having a tantrum, without the hitting. Then give effusive, specific praise ("That was such a great tantrum! I bet you can't do that again!"), along with some physical affection. Practise over and over.

Another relatively practical strategy is to look at when and how parents issue commands. Just as adults don't like being pulled out of something they're working on, neither do children. When mothers were told to direct their toddlers to play with certain toys, for example, it worked best if the child was already turning towards that toy or if the mother was able to attract them to that toy herself, rather than simply making a command. "A request for action coming out of the blue has little chance of succeeding," the researchers write. If you do have to give an instruction, make it direct and clear.

Both sides also emphasise the importance of emotional regulation – in the parent. And they underscore the importance of something often completely overlooked: how caregivers act themselves.

"Parents leave the most important tool off the table, which is systematic modelling," says Kazdin. That means showing the behaviour that you want to see in your child, and pointing it out in others. "The old thing of my generation was, 'Do what I say, not what I do.'" If that’s a parent’s approach, he says, "The research shows – good luck."

--

Join one million Future fans by liking us on Facebook, or follow us on Twitter or Instagram.

If you liked this story, sign up for the weekly bbc.com features newsletter, called "The Essential List" – a handpicked selection of stories from BBC Future, Culture, Worklife, Traveland Reel delivered to your inbox every Friday.