For 38 excruciating minutes, Hawaii feared the worst. In January 2018, phones all over the Aloha State were buzzing with an emergency alert of unprecedented, catastrophic scale.
"BALLISTIC MISSILE THREAT INBOUND TO HAWAII," the text message said. "SEEK IMMEDIATE SHELTER. THIS IS NOT A DRILL."
Fortunately for the more than 1 million Hawaiians lying in the path of this grave missile threat, there was no danger after all. It was all a big mistake, a false alarm triggered by accident during an impromptu drill.
Not that anybody knew that then. At least, not for 38 long minutes, at which point the false alarm was corrected with a follow-up alert clarifying there was "no missile threat or danger".
But just because no explosion ever came doesn't mean the false alarm was harmless. A new analysis of Twitter activity before and after the event indicate the 'crisis' had an unexpected effect on anxiety levels.
After all, what does such an epic shock do to a whole population of people on a psychological level? Hundreds of thousands of residents and holiday-makers who had been told by the government, in no uncertain terms, that a weapon of mass destruction had been fired directly at them?
To find out the psychological impact of the false missile alert, researchers combed through more than 1 million tweets posted throughout the faux emergency, to gauge a proxy of real-time Hawaiian anxiety during this unforgettable ordeal.
"Can a false alarm of an impending disaster itself be a form of trauma?" says psychologist Nickolas Jones from the University of California, Irvine (UCI).
"Our results suggest that the experience may have a lingering impact on some individuals well after the threat is dispelled."
Jones and fellow UCI psychologist analysed 1.2 million tweets from almost 15,000 Twitter users who were identified as probable Hawaii residents based on their account activity.
With these users in focus, the researchers analysed the language in their tweets from six weeks before the false missile alert up until 18 days after the event, looking for terms that are associated with anxiety expression, such as afraid, scared, worried (from a list of 114 anxiety words).
To be clear, tweets aren't the best measure for people's overall mental health, as they're simply something people have chosen to outwardly share. What's going on in people's heads might be worse, better, or totally unrelated. It's impossible to tell from a tweet, and so these results need to be taken with a grain of salt.
However, they're a good starting point for investigating the impact of the incident.
And as expected, the results showed an increase in anxiety during the false missile alert – with anxiety terms rising by 3.4 percent every 15 minutes until the all-clear message was transmitted, and spiking by 4.6 percent overall on the day of the incident.
Once the all-clear was broadcast, though, it's clear the event – and the stress it caused for people in Hawaii – wasn't simply over and done with.
"What surprised us was that the anxiety persisted even after the state's emergency management agency and a local congressional representative issued corrective tweets – which were retweeted by 35,000 users – that the initial alert was a false alarm," says one of the researchers, Roxane Cohen Silver, a social and health psychology scientist.
"This suggests that cancellation of a threat doesn't immediately calm reactions to the situation. Amazingly, some people did not know whether the corrective tweets were believable."
Even more intriguingly, the data suggests that the psychological impact of the event on people's anxiety levels differed greatly depending on how anxious they were before it happened – but not in the way you might expect.
Based on the Twitter users' relative usage of the anxiety words prior to the false alert, each user identified in the study was categorised as either low, medium, or high anxiety.
Strangely, it was the least anxious group who seemed most negatively affected by the fake missile crisis, not the users who were already identified as being highly anxious (to the extent this kind of language analysis technique can be said to indicate actual anxiety).
"For the group of users in this sample who did not express any anxiety before the event … anxiety increased the most and lingered the longest, relative to other groups, before stabilising to a new baseline level 2.5 percent higher than what it was before the missile alert," the authors write in their paper.
"Insofar as anxiety expression on Twitter can be assumed to be reflective of a user's life experience… this pattern is consistent with evidence demonstrating that people who are likely to have had lives devoid of psychologically impactful negative experiences are at increased risk of negative psychological outcomes following a traumatic event."
By contrast, people who appeared to be the most anxious before the false alert seemed to actually become measurably less anxious after the temporarily traumatic episode, with a post-alert baseline that was 10.5 percent lower than their pre-alert level.
"The literature suggests that people who experience negative psychological states, like anxiety, before a large-scale trauma, are at an increased risk for negative psychological consequences afterwards," Silver says.
"However, those individuals who before the alert generally expressed much more anxiety on a daily basis than anyone else in the sample seem to have benefited from the false missile alert instead."
There's a lot more research that needs to be done to explain that result, but the researchers suggest it could reflect a capacity of those people to recognise how much worse things could have been if the missile event hadn't been a false alarm.
Alternatively, it could be a form of relief experienced after a near-miss, the team says, or a kind of balancing of perspective in people predisposed to anxiety.
In any case, there's a lot we still don't know about how people emotionally process catastrophic crises like this (even ones that turn out to be non-existent), but thanks to research like this, we're getting closer to understanding how people think about the unthinkable.
The findings are reported in American Psychologist.