Skip to Content
Uncategorized

Spotting Suicidal Tendencies on Social Networks

Suicide rates in Japan are among the highest in the world. Could the analysis of behaviour on social networks help?

Japan has one of the highest suicide rates in the world. Indeed, for men aged 20-44 and women between 15 and 34, it is the leading cause of death. The rate equates to about 26 deaths per 100,00 people. By contrast, the rate in the US is just 11 per 100,000. 

Which is why the Japanese government has invested heavily in programs to understand the causes of suicide and reduce the number of resulting deaths. Its plan is to cut the rate by 20 per cent by 2017. 

Psychologists have studied suicide for many years. One focus of research is identifying and studying people who have regular thoughts about suicide, so-called suicide ideation. The evidence gathered to date suggests that people with suicidal thoughts tend to be socially isolated, meaning they have not just fewer friends but are also less likely to be members of  friendship triangles in which three people are mutual friends.  

However, these types of studies have been difficult to do accurately. For young people, the data comes largely from questionnaires filled out by students at a particular school or university. The problem here is that when students have friends outside this environment, the outsiders’ role in the social network cannot be properly accounted for. 

This doesn’t influence the data for the total number of friends for each person but it may well influence the calculation of the number friendship triangles.

Today, Naoki Masuda at the University of Tokyo in Japan and a couple of pals address this problem. Instead of studying suicide ideation at a school or university, these guys  looked at in an online social network called Mixi, a major Japanese network with over 25 million members.

Mixi allows users to become members of online communities on various user-defined topics. There are some 5 million of these topics, of which several are about suicide. Members of these groups might reasonably be thought to be prone to suicide ideation. So an interesting question is how these people differ from others and whether this information can be used to target them for help.

Masuda and co they simply compared the members of these groups, some 10,000 of them, with a control group of over 200,000 who are not members of these groups.  

The results are in some ways surprising. It turns out the people prone to suicide ideation have about the same number of  friends as the control group. This alone does not seem to be a defining characteristic in the online world, where ‘friends’ are easy to come by.  Neither does age or gender seem to be an identifying chaacteristic, which flies in the face of previous research.

The warning signals are more subtle, say Masuda and co. For example, people prone to suicide ideation are likely to be members of more community groups than the control group. That may be the result of spending longer online and of a desire to want to interact. 

But a key indicator seems to be that these people are much less likely to be members of friendship triangles. In other words, they have fewer friends who also friends of each other.  This low density of friendship triangles appears to be a crucial.

That’s an interesting step in the study of suicide ideation online, which is relatively unexplored. The approach  has considerable future potential. Online social network data offers the potential to study users over time and how their thoughts and behaviours change. That could conceivably reveal much more detail about the factors that influence suicide ideation and change behaviour.

However, there are still huge gaps in our understanding  of the link between online and offline behaviour. This needs to be filled, and not just in the study of suicide ideation. The obvious Japanese interest in this area and the government’s desire to intervene may provide the impetus for  change.

Ref:  arxiv.org/abs/1207.0561: Suicide Ideation Of Individuals In Online Social Networks

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.