Most Kiwis think New Zealand should have a proper earthquake early warning system – but that doesn't mean we'd all respond to it correctly in the seconds before a big shake.

An insightful new study has suggested that, for all of the clever technology built into these platforms, how we'd react to getting an early heads-up to an incoming quake could come down to our personal experiences with them - especially when these systems were new.

Earthquake early warning (EEW) systems work by detecting and reacting to an earthquake's fastest-travelling P-waves before the slower but more damaging S-waves arrive, giving people moments to prepare.

New Zealand is one of few countries to host a Google-pioneered system that pings messages to peoples' Android phones before the shaking, telling them the estimated strength and how far it was from them.

Kiwi company Jenlogix also operates a network of P-wave-detecting Palert units that's used by several councils, universities, district health boards, ports, and power companies.

The units all streamed data to a central server, but also had triggers than could activate local emergency systems – such as stopping a lift at the next floor and open the doors, or turning off the gas supply.

Potential use of other low-cost sensors, such as Raspberry Shakes, is being investigated by Massey University researchers.

But unlike Japan, Mexico and now parts of the US, the country doesn't have a nationally-funded EEW system – something that could cost $25m a year to run here, and perhaps $50m to prepare.

A 2013 GNS Science report found creating one would require a massive upgrade to our geodetic sensor network, along with a "significant increase" in funding.

All the while, researchers have been interested in just what Kiwis would do if they suddenly got an alert from one.

"We were interested in understanding whether people's prior experience of earthquakes or earthquake drills might affect their intended responses to a warning," said Dr Julia Becker, of Massey University's Joint Centre for Disaster Research.

In a study just published in the journal Frontiers in Communication, she and colleagues surveyed more than 3000 people on what they'd think of an EEW system, and how they might respond.

They also asked whether they'd been in strong quakes before, or had taken part in drills like the nationwide ShakeOut exercise.

The responses were fascinating: doing exercises or drills didn't seem to prompt the right actions for earthquake warnings, such as drop, cover, and hold.

There were several possible reasons for that: one being a lack of education in linking warnings with drills like ShakeOut, which has been shown to boost people's use of protective actions in actual quakes.

Rather, the researchers found the more personally relevant a person's experiences were – like having a family member previously injured from a quake – the more likely they were to intend to take a useful action when they got an alert.

Those whose loved ones had been hurt were more likely to move somewhere safe, mentally prepare, tell others, or pull over and stop if they were driving.

Researchers found the more personally relevant a person's experiences were, the more likely they were to intend to take a useful action when they got a quake alert.  Photo / Mark Mitchell

Regional experiences with bigger quakes also seemed to play a part: people in Canterbury and Wellington reported a greater likelihood to mentally prepare for a quake if they got an early warning.

While having an EEW system was seen as more useful among people whose family or friends had been affected, this effect was statistically small – and there were no observed differences for other types of experience, including among those who'd been hurt themselves.

"This is interesting, as we'd probably expect people with more experience to see the system as more useful," said study co-author Dr Lauren Vinnell, also of the Joint Centre for Disaster Research.

"It could be that people aren't sure about exactly how the system would work and what benefit they could get from it; in our study earthquake early warning was hypothetical.

"It could also be that people were showing either fatalism or normalisation bias – if people had really negative experiences, they might think a warning couldn't possibly help, or if they've had earthquake experiences that weren't too bad, they might not see the point of a warning."

Ultimately, the study highlighted how public education would be crucial if New Zealand set up a state system – just as it was already with Google-run alerts now accessible for about half the country.

"For a system to help people, they should know how it works and how it might go wrong as well as what they should be doing when they get a warning," Vinnell said.

"Secondly, we should also as much as possible bear in mind the experiences people have had with earthquakes. The more specific we get with this the better."

"It's important to consider how different people might perceive and use early warnings differently, so that we can make sure they aren't doing more harm than good."