In my opinion and experience there is no tried and true method to deal with reactivity posative or negative. It all depends on the dog and its motivation for the behavior.
My current little project is a classic case of genetic weak nerve.
She sees another dog while wlking on the leash she focuses on it head up ears forward, tail lashing, the closer the dog get her head drops growls and then lunges. Same reaction every time, if I get her at the right moment a quick correction throws her off and stilts her escelation. If we sit/stand close to the object of her reactivity she relaxes after a minute or so and then wants to play. Every time no matter what type of dog.
Take her to the dog park, throw the E Collar (She is collar conditioned) on her let her loose. She spots a dog hackles lunges, gets stimmed breaks off, even apporaching another dog to intensly leads to a stim, she always breaks off the approach then re approaches and uses her nose properly. Then she wants to play, same pattern as when we are on leash.
Her overwhelming response is exitement mixed with an equel if not greater dose of insecurity. Which leads to the lunging growling etc.
Her excitement level when around other dogs generally precludes the use of treats as her fixation on the other dog or dogs is to high, only a correction can refocus her attention. Then she is open to an alternate message.
People is a bit different. Generally she is much more sharp indoors, so I will use her responses in those settings as she generally ignores people outside.
Any individual apporaching us leads to growls, or if it was allowed an abbreviated lunge. Correction of this reaction yeilded a submissive yet still fear aggressive dog. As in the dog is in a down but still hackling growling. There is no tail waging in this response, instead you see hackles up, ears back, head low and tail tucked.
I diagnose this as fear / insecurity
Once the object of her reaction has stopped moving and is relatively still she is now (with a lot of work is able to relax) as in sniff the person, take treats do some light OB etc.
I mainly treat this reaction exclusively with treats, as in as soon as she relaxes I give her a treat, then the person in question gives her a treat. We are now moving to having the stranger talk loudly, move suddenly, move arms around etc. Any aggressive response recieves a light correction more of a redirection, then once she relaxes immidiately get rewarded by myself and the helper. Also, I will do light OB around the helper then hold the treat right against the helper or at the helpers feet so she make posative associations.
These are just some examples, I have found that a mix of correction and +R timed appropriately is the most effective approach.
I dont know why people insist on applying absolutes to dog training simply to suit their own biases and perception of how the world should be. IMO use what works plain and simple, most world class trainer/handlers use a healthy mix of both approaches. Political correctness has no place in working with dogs as they unlike human beings are immune to BS.