Texting has become a part of our everyday life.
It’s how we communicate with our family, friends and colleagues.
But the words and emojis we use can indicate a great deal about the state of our mental health.
Our texts can show if we’re under stress, suffering from depression, or even having suicidal thoughts.
The words and emojis we send through text messages can indicate if we’re feeling depressed, stressed or anxious. Texting the crying face emoji could be a better indicator of someone being in distress than texting the word ‘suicide’
The Crisis Text Line (CTL), a text messaging–based crisis counseling hotline, has based its entire business on analyzing communication.
Data scientists at CTL are using machine learning, a type of artificial intelligence, to pull out the words and emojis that signal a person at higher risk of suicide or self-harm.
The computer also tells them who on hold needs to jump to the front of the line to be rescued.
According to Bob Filbin, the chief data scientist at CTL, the service has exchanged 33 million messages exchanged with texters in crisis.
And depending on how we start a conversation, the words they use, or even the emoji faces can say a lot about what problem we may be facing.
When we use words like ‘nervous’, ‘sometimes’, ‘hard’, ‘feeling’, they are usually all indicators of anxiety.
Stress messages often include words like ‘bad’, ‘guess’, or ‘anymore’.
Both the words ‘mom’ and ‘parents’ are most likely to be sent via text when the texter is feeling stressed
Filbin says that when someone texts the crying face emoji, it’s often a better indicator of someone being in distress than texting the word ‘suicide’.
When we use words like ‘nervous’, ‘sometimes’, ‘hard’, ‘feeling’, they are usually all indicators of anxiety. Stress messages often include words like ‘bad’, ‘guess’, or ‘anymore’
CTL’s data has turned up all kinds of interesting insight.
For instance, Wednesday is the most anxiety-provoking day of the week, and crises involving self-harm often happen in the darkest hours of the night
‘Before we used the computer, we had a list of 50 words that [we thought] were probably indicative of high risk. Words like ‘die,’ ‘cut,’ ‘suicide,’ ‘kill,’ etc,’ Filbin told Vox.
‘When a data scientist ran the analysis, he found thousands of words and phrases indicative of an active rescue that are actually more predictive.
‘That’s the whole idea and the power, really, of AI – it gets smarter over time.’
And many scientists believe indicators can be found in a number of social media posts as well.
A study published this month by Qntfy, a startup mental health analytics company, used Twitter.
Researchers estimated the emotional content of tweets from hundreds of users who had talked openly about a suicide attempt, and tweets from a control group that did not display suicidal thoughts or feelings.
While nearly everyone in their sample included emoji in their tweets, the researchers found that some in the group that talked about attempting suicide employed a narrow range of emoji representing sadness more frequently, such as blue or broken hearts.