actorartathleteauthorbizcrimecrosspostcustomerservicedirectoredufoodgaminghealthjournalistmedicalmilmodpostmunimusicnewsworthynonprofitotherphilpolretailscispecialisedspecializedtechtourismtravelunique

I’m Dr. Henry Lieberman, principal research scientist at the MIT Media Lab, investigating cyberbullying. Ask me anything.

Mar 11th 2013 by HLieberman • 11 Questions • 621 Points

Henry Lieberman, here, starting the AMA at 12:30pm EDT on Monday 11 March until 2:30pm. I had already answered a few questions, below.

I am studying online bullying behavior, along with my students Karthik Dinakar and Birago Jones. We are working on software that can detect bullying language and behavior. Here's a very short video I did recently with NowThisNews about our work: http://nowth.is/Y9GUH1. See also this article in The Atlantic, http://www.theatlantic.com/magazine/archive/2013/03/how-to-stop-bullies/309217/

We hope that the outcome of our work will be to help educate kids about bullying, and to get everybody involved (bullies, victims, and bystanders) to reflect on their behavior. We hope this will help make every social networks a more welcoming and safe environment to interact.

Q:

Where I teach many of the older teachers say "Kids always do a better job at policing themselves then we can ever do."

Can you address this statement?

A:

Yes, that's true. What's heartening to me is the courage of many kids who have started anti-bullying groups in their schools, gay-straight alliances, and other positive social movements. I don't consider that "policing", but just a way for kids to help set the direction of their own social groups. The message is stronger and more effective when it comes from kids themselves than from adults.


Q:

I know you are investigating cyber bullying, but what about if it is happening on and off the web? My nephew is going through this right now, and the rules on how to handle it are so delicate. It's hard to get some justice.

A:

I'm sorry to hear about your nephew. What we can do online is to provide education that explains to kids why it occurs, and to give them resources, help, and emotional support to cope with it when it does. Hopefully, this will change attitudes so that, ultimately, there will be less face-to-face bullying as well.


Q:

What was the worst case of cyberbullying you had? How did it result?

A:

I don't know if I can choose "worst cases". But by definition, the worst ones are the ones that result in suicide, and there are plenty of those reported by the media. I believe that most of these are preventable.


Q:

Has any part of your bullying research ever made you laugh?

A:

Yes! Sometimes humor is a great way to deflect bullying, and we've seen kids be incredibly clever in their responses. Attempts at humor by bullies, in contrast, are mostly crude and unfunny.


Q:

Thoughts on video games and violence?

A:

I think violent video games do have a negative effect, but I do not support censorship. I think the key is to promote positive alternatives. I actually think the worst things on TV are not the violent cop shows (which many people can dissociate from reality), but the situation comedies and "reality" shows, where bullying and humor at the expense of others are shown as normal or positive.


Q:

What approach are you taking to investigate cyber bullying? do you think that the bully is all at fault(saying that the bully is abused at home or something where bullying is an escape to some or joke to others)? do you think social media is doing enough to help stop cyber bullying?

A:

For a short introduction, see the Atlantic article in my introduction. For technical readers, a journal article is here: http://www.media.mit.edu/~lieber/Publications/Bullying-TiiS.pdf

We're not interested in "blame", be it the perpetrator or society as a whole. We're interested in what we can do to either prevent the problem or mitigate its consequences.

I don't think social networks are doing enough right now. As our work shows, the software of the social network itself can help change people's attitudes and either help create a welcoming environment, or not.


Q:

How is Reddit in terms of cyber bullying?

A:

It's like any other group. Mostly, I've gotten thoughtful and respectful questions and discussion. I've gotten a few hostile people, as you can see here, but I'll just ignore them.


Q:

How does the program react to someone saying something mean about themselves?

A:

The goal of our program is to detect possible cases of cyberbullying, we can't make a definitive judgment about whether something is cyberbullying or not. There are many ways to "fool" the program. We just want to reach the level of confidence where it makes sense to offer the person some kind of help. If our judgment isn't correct, and the person doesn't really need the help, there's not much harm done.


Q:

Are you simply using a keyword search, or is your program going to be able to identify tone and context of a message?

A:

No, it's not just a keyword search. We're using some sophisticated machine learning, trying to identify the overall themes of the post, and a knowledge base specific to bullying. The full technical details are in:

http://www.media.mit.edu/~lieber/Publications/Bullying-TiiS.pdf


Q:

Have you collaborated with helping professionals such as clinical social workers/psychologists? If so, how? If not, is this something you would like to explore? -Helping Professional in Training

A:

We'd like to more, if we find the appropriate people. Sometimes I find books written by psychologists or educators on the topic to be somewhat vague or unhelpful. They give generic advice like, "tell an adult", which sometimes doesn't feel effective to kids actually in the situation. What's necessary is to be able to customize the advice to each kid's situation, and listen and respond to the kid's actual concerns. We're looking at ways to deliver personalized education appropriate to an individual situation.


Q:

kids have bullied others since civilization started; do you really feel this warrants research? i mean if the kids don't want to be bullied more, just turn your damned computer off or go to another website. people aren't nice to you on Facebook; block them. It's not a hard concept.

Seriously, I grew up in a time where if you got into a school fight, you usually ended up as friends after. I don't see the big reason for whining about it now.

A:

Yes, I do feel this warrants research. The negative social consequences of bullying have now, as a result of research, been conclusively identified. Kids who are bullied (or who bully) do worse in school, have poorer health outcomes, and many other real consequences for our society. It needs to be addressed. Bullying is not "inevitable", or "normal" and anti-bullying efforts are not "whining".

I'm an old guy too, and I remember when attitudes were as you say. Attitudes are now changing, thankfully. If you survived bullying, congratulations. Many others weren't so lucky. Don't wish your misfortunes on others.