flooleyoobley
New member
The Look At That (LAT) training method was popularized mostly in Leslie McDevitt's Controlled Unleashed and is a commonly-used force free technique for reactive dogs. I've long been a proponent of this method, but it's not always easy to implement (as with anything with a reactive dog) and I find it's often very misunderstood.
After struggling with my reactive pooch for the last 7-8 months, we've been having more breakthroughs with this method, and I thought I'd summarize some of the key points to keep in mind when trying to implement LAT. In no particular order:
After struggling with my reactive pooch for the last 7-8 months, we've been having more breakthroughs with this method, and I thought I'd summarize some of the key points to keep in mind when trying to implement LAT. In no particular order:
- Forget about what your dog is doing when starting out. Although in LAT we are teaching an operative behaviour or 'pattern' to help the cope with their environment, i.e. looking at a trigger as a form of targeting, the over-arching goal of LAT is classical counter-conditioning, i.e. changing your dog's emotional response to the trigger. This is the core of the technique but I think it's often mis-judged or not well understood. I've seen people using this technique, as well as other methods, where they focus a lot on getting their dog to heel, sit, look at them, look at that other dog etc., and only provide reinforcement when the dog is not reacting. This is often counter productive to the goal of changing the dog's conditioned emotional response to the situation because they are quickly over-threshold. Instead, forget about your dogs behaviour and focus on providing gushing rewards as soon as your dog clocks their trigger. If I could bring a whole roast chicken with me and throw it in front of my dog the second they see another dog, I would. It doesn't matter what your dog is doing. They may be trying to react at their trigger but you can still provide lots of rewards for them to indulge in -- you can't reinforce an emotion. You want your dog to realise that when they see the trigger, the heavens open with glorious things, and as soon as the trigger disappears, the rewarding stimuli disappear. You could also think of this as a form of non-contingent reinforcement where the learner is reinforced regardless of their behaviour, and via that process, problem behaviours will often be discontinued.
- LAT gives your dog chance to self-soothe. I feel strongly that LAT was much better than teaching my dog to look at me because my dog is so fiercely vigilant. Nothing is going to be more important to my dog than keeping track of the scary things in their environment, so why not let them? We can allow them to look, provide lots and lots of nice things, and build this behaviour into our counter-conditioning.
- Start in a familiar, safe environment. In practice, we know keeping our dogs under-threshold is best for learning, so I'd always start LAT in a familiar environment. If you can start somewhere where they can see the stimuli that they react to but feel safe (e.g. it could be a family member in the back garden or familiar dogs in a safe area) then you can start to teach them that what a triggering stimulus relates to more quickly. I've done this with my dog who is very leash reactive, starting in a training facility with dogs he knows, and just rewarding for any look at the other dogs.
- Vary the rewards, if possible. My dog is very food motivated, but poisoning the reinforcer is very easy to do. We've had a lot of success recently by switching up the training food every couple of days, which has kept my dog's interest much more than using a single very high value food for multiple days.
- Don't skimp on the jackpots. When starting out, feel free to use loads of treats. To really get your dog to realise that the trigger is predictive of nice things, giving one piece of food isn't going to cut it for many dogs. Think handfuls of food as soon as they see the thing they don't like. You might be done in a very short amount of time, but that's fine, because...
- Short sessions go a long way. I've been driving my dog to new spots for 2-5 minute LAT games, e.g. a large intersection where there are dog walkers. We've also been finding large car lots excellent places to watch people and dogs go past and practice. We have the car right there if we need to abort or a have a breather, which is more comforting for him and me. I feel like these short but frequent sessions have been a game-changer.
- Practice LAT on other stimuli. We still practice LAT with people, even though my dog is no longer reactive to them (but still wary). I feel like this helps with LAT on other dogs, because my dog might already have had a few trials of the behaviour before seeing a dog that day.