Suicide
Social Media Can Identify Those Who Are at Risk of Suicide
How to save a life: Suicide, social media, and technology.
Posted October 9, 2022 Reviewed by Jessica Schrader
Key points
- Patterns in text messages can identify future suicide attempts.
- A pattern of past behavior, increased anger, and decreased positive emotion are among the most alarming potential risks.
- Two-thirds of people who died by suicide shared their plans on social media but denied suicidal thoughts to their therapist.
Most of them are tweens. Or teens. Or 20-somethings.
Once upon a time, they were someone's children.
Their parents held their breaths as they took their first steps.
With clumsy fingers, the children learned to tie their shoes.
Their parents reminded them to say please and thank you, to be nice to the other children, and to make friends with people who were nice to them in return.
But as children grow older, some become crueler, and social settings become more unsettling than the adolescent brain is prepared for.
And sadly, according to the National Alliance on Mental Illness, the feeling of fitting in and of having friends who support and understand you is one of the top factors in whether someone decides if they want to live or die.
Suicide is the second leading cause of death among people aged 15-24 in the United States. And the first leading cause of death of those aged 15-24? "Unintentional injury."
Not "accidental injury," but "unintentional injury."
The wording is curious, and perhaps given with grace, a possible gift for those who cannot stomach the thought of losing a loved one by their own hand. And yet, we so desperately need to talk about suicide and suicidal thoughts in order to decrease the stigma and leave an open road for people to travel to find the mental health support they need.
Approximately 50% of All Lifetime Mental Illnesses Begin By Age 14
And while 20% of high school students report thinking about taking their own lives, at last count 9% report having already made an attempt to take their own life.
This is the problem we are facing. This is the reality of what we are dealing with.
This is why we are beyond sugarcoating death with vague obituaries and obscure, unanswered explanations.
We Don't Talk Anymore, But We Do Text
Although most of us text with a frequency we never could have imagined when we held blackberries in our hands so many years ago, teenagers text to such a great extent that some have posited that they use their phones to create their environment. Co-construction Theory (Subrahmanyam et al, 2006) believes that this world that adolescents create is as real, if not more real, than the world offline. More importantly, the most important developmental challenges they face, such as sexuality, identity, substance abuse, etc. are navigated online—with words their parents aren't privy to.
Which explains the shock and disbelief that make the grief all the worse for the loved ones of an adolescent who has taken their own life.
In an article published in 2022 in Clinical Psychology Science1 Glenn, Nobles, Barnes, and Teachman, language analysis software was used to examine 200,000 text messages to identify texts from people who later attempted suicide. The software was programmed to flag similar texts with the hope that this would help identify other people who might be at risk of suicide.
All 33 participants in the study admitted having made at least one previous suicide attempt. All were over the age of 18 and were recruited from the University of Virginia Psychology Department and the Charlottesville community at large.
The results of the study were quite clear:
- You cannot determine by the words someone types whether they are planning to attempt suicide. Slang expressions such as "I just want to die!" or "Kill me now, I'm so embarrassed!" are used regularly by our youth.
- A pattern of increased anger and decreased positive emotion may be a potential indication that an individual is moving towards a greater likelihood of taking their own life.
- If you use language analysis software equipment to track texting behavior and emotions over time, you can identify patterns that increase the likelihood of a future suicide attempt, particularly in individuals who have attempted or exhibited strong suicidal ideation in the past.
![Yan Kruvov/Pexels Yan Kruvov/Pexels](https://cdn.psychologytoday.com/sites/default/files/styles/article-inline-half-caption/public/field_blog_entry_images/2022-10/pexels-yan-krukov-7640773.jpg?itok=koa9DrdY)
Text messaging someone you only know online potentially increases the likelihood of danger, since you cannot rely on your other senses such as body language, pheromones, eye contact, and vocal intonation—all of which convey a sense of whether someone can be trusted.
But, there is another side of the same coin: because adolescents and 20-somethings have grown up with online communication being the norm, some research indicates many people in this generation actually feel more comfortable revealing their hidden secrets or fears in this more invisible form of communication.
This makes those online communications, and the ability to track changes in online communications over time significantly more important as a potential assessment of someone's mental health. Unfortunately, at this point in time, the computer program is much more successful at this task than the human eye alone.
Tweets Aren't Always Sweet
Statistically speaking, the connection between social media and suicide is somewhat appalling. Several studies have found that 8 out of 10 people with suicidal thoughts will disclose their plans online. 3, 4, 5
Unfortunately, the world of online socialization tends to fall into one of two extremes: an extremely close-knit loyal-to-a-fault clique, or competitive friendships that quickly devolve into vicious, painful bullying.
A 2021 paper on computer linguistics and suicidality, "Suicide Ideation Detection via Social and Temporal User Representations using Hyperbolic Learning," created a computer program that sought to find a connection between the last thing someone tweeted and their potential suicide risk.
The computer program, or algorithm, utilized the work of earlier models to create a sort of super-system that could do three things at once:
- Analyze the most recent tweet of an individual who had expressed suicidal thoughts or ideas.
- View this most recent tweet in both a social context (how this tweet compared to this person's online peers) and view this tweet in an emotional context (how this tweet compared to the individual's previous tweets in terms of being an emotional extreme).
- Factored in the mental health status and suicidal ideation of the Twitter user's online peer group as a whole.
Interestingly, it is the mental health status and suicidal ideation of the peer group that seems to best indicate an individual's likelihood of taking their own life. In fact, one study found that 60% of people who died by suicide had denied having suicidal thoughts to mental health practitioners, but had shared their plan with their peer group.
How Do We Help?
The good news is that research examining technology and suicide proves that there are breadcrumbs we can follow—once we know what to look for. And hopefully, these breadcrumbs can prevent the death of some of our children.
But there is more we can do.
- Encourage your children to be kind—especially to the people who don't seem to fit in, or lack a friend group, or are often the subject of emotional or physical bullying.
- Ask the tough questions: Have you ever hurt yourself? Have you ever thought about hurting yourself? Have you ever thought about suicide? There is no empirical evidence that asking someone if they are depressed or suicidal will cause them to be depressed or suicidal.
- If you are concerned that your child is depressed or anxious or struggling socially and they aren't comfortable talking to you, ask them if they would like someone to talk to. It's really simple: "Do you think it would help for you to have someone to talk to, like a therapist?"
- Encourage your child to be an advocate for others who cannot advocate for themselves. Encourage your child to recognize others who might be suffering, and teach them the importance of empathy.
What if we could save someone's life, and what if we could teach our children to do the same?
If you or someone you love is contemplating suicide, seek help immediately. For help 24/7 contact the National Suicide Prevention Lifeline, 1-800-273-TALK, or the Crisis Text Line by texting TALK to 741741. To find a therapist near you, see the Psychology Today Therapy Directory.
References
1 Glenn JJ, Nobles AL, Barnes LE, Teachman BA. Can Text Messages Identify Suicide Risk in Real Time? A Within-Subjects Pilot Examination of Temporally Sensitive Markers of Suicide Risk. Clin Psychol Sci. 2020 Jul;8(4):704-722.
Burgoon, J. K., and Le Poire, B. A. (1999). Nonverbal cues and interpersonal judgments: participant and observer perceptions of intimacy, dominance, composure, and formality. Commun. Monogr. 66, 105–124.
Flek, L., Harshit, J., Ratn Shah, R., Sawhney, R., (2021) Suicide Ideation Detection via Social and Temporal User Representations using Hyperbolic Learning. 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2176–2190 June 6–11, 2021. ©2021 Association for Computational Linguistics
3 David D Luxton, Jennifer D June, and Jonathan M Fairall. 2012. Social media and suicide: a public health perspective. American journal of public health, 102(S2):S195–S200.
4 Glen Coppersmith, Mark Dredze, and Craig Harman. 2014. Quantifying mental health signals in Twitter. In Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality, pages 51–60, Baltimore, Maryland, USA. Association for Computational Linguistics.
5 Robert N Golden, Carla Weiland, and Fred Peterson. 2009. The truth about illness and disease. Infobase Publishing.
Subrahmanyam K, Smahel D, & Greenfield P. (2006). Connecting developmental constructions to the Internet: Identity presentation and sexual exploration in online teen chat rooms. Developmental Psychology, 42, 395–406.