How do i teach my dog to sit and stay?

A 30-second stay in a position isn’t just a longer 3-second stay. Something different is happening in your dog’s brain and understanding it changes how you’ll train duration.

A quick note on the word “stay”: most professional trainers don’t teach stay as a separate command. If a dog sits, the expectation is that they hold it until you release them. “Stay” is redundant, and teaching it separately may lead to other problems. What people mean when they say “stay training” is what this article is about: duration. How long can your dog hold a position – and why does it fall apart?

The sit that falls apart at the worst moment

You’ve been working on sit. In the kitchen, three seconds – easy. Five seconds – fine. Ten seconds – still good. Good job, you’re patting yourself in the back.

Then one day you push it a bit further. Fifteen seconds. Twenty. And somewhere around the twenty-five second mark, your dog just… gets up. Wanders off. Maybe shakes themselves off like they’re resetting. Or they hold it, but you can see something is happening.

The tempting interpretation is that they got bored, or decided the rules don’t apply today. What you’re watching is a physiological process, one that has nothing to do with obedience and everything to do with how arousal and self-regulation work in the brain.

There’s a sweet spot – and it’s smaller than you think

In the 1950s, psychologist Donald Hebb mapped out something that’s become foundational to how we understand learning and performance. The relationship between arousal and performance isn’t linear – it’s an inverted U. Too little arousal and you get a disengaged, switched-off dog who can’t really be bothered. Too much arousal and you get an impulsive, reactive dog who’s physically incapable of the kind of controlled, careful behaviour you’re asking for.

In between is an optimal zone – a window where the nervous system is alert enough to engage but calm enough to think. Not wound up, not zoned out. Somewhere in the middle. That’s where learning happens. And that window is narrower than most people expect.

Here’s what matters for duration training: holding a stay is not passive. It’s an active cognitive task that requires sustained attention and impulse control. That mental effort has a cost, and it accumulates the longer you ask your dog to hold position. As it builds, it destabilises where your dog sits in that optimal window.

The direction it tips depends on the dog and the situation.

In a stimulating environment like other dogs around, interesting smells, things happening – the cognitive load of holding still can push arousal upward, into overwhelm. The dog gets fidgety, increasingly tense, and eventually breaks dramatically. In a quiet, slightly boring environment, the same sustained effort can tip the other way. Arousal drifts downward, engagement fades, and the dog slowly melts into a kind of glazed disconnection. It is more like gradually stopping participating in it.

At that point, breaking the stay isn’t a choice, but at some point, it just becomes too hard.

And there’s something else going on underneath

Even when arousal is well-managed, duration applies a second kind of pressure. Research on how learned behaviours work has shown something important: older, previously trained behaviours never fully disappear from memory. They get quieter, pushed to the background by whatever has been more recently and consistently reinforced. But they aren’t gone. Under certain conditions, they come back.

Researchers call this spontaneous recovery. What triggers it? Time. Pressure. Both.

A dog who used to pace or whine before they learned to hold a stay still has that pacing and whining stored in memory. Under normal conditions, the stay behaviour wins because it’s been reinforced more recently and more consistently. But as a stay stretches on, as arousal climbs and cognitive load increases, the newer behaviour becomes harder to maintain and older patterns can start to surface.

‘He was doing so well and then it all fell apart’ is often less about regression and more about a system under strain. Perfect time for old habits to resurface.

This is also why stay duration collapses so easily when you stack other demands on top of it. Add a distraction while your dog is already at the edge of their duration capacity and you’ve just loaded two demands onto a system that was already running close to full.

Duration is a fitness variable, not just an obedience one

Think of it like physical fitness. You wouldn’t expect someone who just started running to hold a 10-minute pace indefinitely. You’d build up gradually, expect some bad days, and understand that stress, sleep, and general state all affect performance. Duration in dogs works exactly the same way. It’s something you build incrementally, it varies day to day, and it can temporarily drop when other demands are high. A dog who reliably holds a 20-second stay at the park, after 40 minutes of exercise, with other dogs nearby.

Duration is something you build, not something you unlock. And like fitness, it needs to be maintained and not just achieved once.

Why this is where most training plans quietly fall apart

People often don’t train duration as its own variable. They get a solid 10- second stay at home, take it to the park – new location, higher distraction and ask for 10 seconds again. That’s three variables changing at once: location, distraction, and duration all shifting simultaneously. If it falls apart, you genuinely can’t tell which one broke it.

The same logic applies here as in Article 1, where we looked at how location and distraction each need to be trained deliberately as separate variables. Duration is no different. Stack them before any one is properly solid and training starts to feel broken – inconsistent, random, hard to diagnose. Usually there’s a specific variable that hasn’t been built properly yet. And it’s dragging everything else down with it

Adding any of the other Ds while duration is still shaky is the fastest way to make training feel like it isn’t working – when really, you’ve just skipped a step.

The fix isn’t complicated, but it does require being honest about where your dog actually is, rather than where you’d like them to be. Build duration in calm, familiar conditions first. Get it solid, not ‘usually fine’ solid, but reliable across different days and different moods. Then start moving the other variables, one at a time.

How to build duration

Treat duration like a separate skill with its own training plan and not like an extension of whatever behaviour you’ve already been working on. There’s a less obvious factor worth knowing about here: self-control appears to draw on a shared biological resource that can be depleted across a session – and across a day.

Research with dogs found that animals required to hold a sit-stay before a problem-solving task persisted significantly less on that second task than dogs who hadn’t had to exert self-control first. When given a glucose drink in between, the deficit disappeared. The same pattern had already been documented in humans. The exact mechanism is still debated, and the glucose story has proved more complicated than early research suggested. But the core finding is robust: self-control tasks make subsequent self-control harder. Session order matters. Balance breaks are not optional.

A few practical things that actually help:

• Start shorter than you think you need to. If your dog can hold something for 10 seconds, train at 5.
• Build in small increments – add 2 or 3 seconds at a time, not 10.
• End every session before your dog breaks position, so the last thing their nervous system registers is success, not failure.
• Pay attention to what “almost breaking” looks like for your dog. The goal is to end the behaviour before that point, not after. Finishing on a break is how you accidentally train your dog to break.

And on days when duration is suddenly worse than it was last week, consider what else is going on.

• Did they sleep badly?
• Is it a hot day?
• Have they had a lot of mental stimulation already?

Duration is genuinely sensitive to overall state. A dog who held 30 seconds easily on Tuesday might struggle with 15 on Thursday.

The fix isn’t complicated, but it does require being honest about where your dog actually is, rather than where you’d like them to be.

Next up: Distance

Distraction, Duration – and now the third D: Distance. Of the three, distance is the one that surprises people most, because it turns out to be about much more than how many meters away you’re standing.

When you step away from your dog, the attachment system activates. And what looks like a simple spatial variable is one of the most emotionally loaded asks in training. That’s what I plan for Article 3.

The research behind this article

I don’t expect everyone to want to read the original papers, but I always think it’s worth knowing that what I write about has solid science behind it. Here’s where this one comes from:

HEBB DO. Drives and the C.N.S. (conceptual nervous system). Psychol Rev. 1955 Jul;62(4):243-54. doi: 10.1037/h0041823. PMID: 14395368.

Bouton ME. Context and behavioral processes in extinction. Learn Mem. 2004 Sep-Oct;11(5):485-94. doi: 10.1101/lm.78804. PMID: 15466298.

Delamater AR, Westbrook RF. Psychological and neural mechanisms of experimental extinction: a selective review. Neurobiol Learn Mem. 2014 Feb;108:38-51. doi: 10.1016/j.nlm.2013.09.016. Epub 2013 Oct 6. PMID: 24104049; PMCID: PMC3946186.

Thrailkill EA. Partial reinforcement extinction and omission effects in the elimination and recovery of discriminated operant behavior. J Exp Psychol Anim Learn Cogn. 2023 Jul;49(3):194-207. doi: 10.1037/xan0000354. Epub 2023 Jun 1. PMID: 37261748; PMCID: PMC10524675.

Miller, H. C., Pattison, K. F., DeWall, C. N., Rayburn-Reeves, R., & Zentall, T. R. (2010). Self-Control Without a “Self”?: Common Self-Control Processes in Humans and Dogs: Common Self-Control Processes in Humans and Dogs. Psychological Science, 21(4), 534-538.

Leonardi RJ, Vick SJ, Dufour V. Waiting for more: the performance of domestic dogs (Canis familiaris) on exchange tasks. Anim Cogn. 2012 Jan;15(1):107-20. doi: 10.1007/s10071-011-0437-y. Epub 2011 Jul 16. PMID: 21769501.