The PSYCHOLOGY of grooming: an interview with ian elliott 


Dr. Elliott is a Forensic Psychologist, an expert on sexual predators, and an expert on the underlying psychology of grooming.

In this interview he discusses sexual grooming and how it can relate to violent extremist radicalization and online safety. 

Dr. Elliott presenting his Universal Model of Grooming

Dr. Elliott presenting his Universal Model of Grooming

Does a groomer always reach out to a child, or does the reverse ever happen with a child reaching out to a groomer?

It’s difficult to say, going back to my background in sexual exploitation and how that relates to extremism, a lot of grooming behavior and sexual exploitation tends to occur within households. In the vast majority of sex abuse cases the abuser is known to the child. It’s quite rare that child abuse is perpetrated by complete strangers. But obviously the perpetrator being a stranger is more likely in online grooming.

In terms of children reaching out, in my experience it doesn’t often happen. You tend to find in the majority of cases that it is adults going to spaces on the internet where children are. So in that way it’s similar to offline situations. It tends to be that adults go to those places and attempt to identify and single-out individuals that they think might be vulnerable to grooming. So it tends to be more that adults would go to places on the internet where they could find children, rather than children going to places where they are more likely to be groomed.

It’s also worth noting that when you are talking about online recruitment of children for any reason, you are really talking about adolescents. We tend to see with the sex offenders that they are contacting children over 11 or 12 years old. Not only is it less likely that children younger than 11 are going to have access to a lot of the websites and technology, but also one element of what you are trying to do involves persuasion and developing relationships, it is more difficult to do that with much younger children.

“It tends to be more that adults would go to places on the internet where they could find children, rather than children going to places where they are more likely to be groomed.”

So one of the issues we’ve had to face within the sex offender field is around understanding and recognizing that adolescents have agency on the internet, that they are able to take risks on the internet and do take risks on the internet. Certainly in terms of adolescent pathology they are sometimes slightly more predisposed to taking risks than the same people would be later on in their lives.

So the issue is that we need to recognize that children have that agency but at the same time recognize that the recruitment of children for malevolent purposes online is exploitative and it’s still not their fault. So the challenge is around framing that correctly. In violent extremism that would mean recognizing that although there is violent material out there, and that some children and adolescents might be naturally more tempted towards riskier online behavior than adults might be, that the responsibility for any exploitation always lies with the exploiter.

So where do you think children are most vulnerable to being groomed online?

In the sex offender field, we’ve got 2 decades worth of research on online grooming in chat rooms. A lot of the work I’ve done has been based on transcripts that we’ve taken from adults talking to children in chat rooms. But the difficulty now is that an awful lot of that might be quite outdated, given the way that technology has evolved, given the way that social technology has moved on. Chat rooms are probably an alien concept to most children, they probably don’t even know what they are these days.

So I think that there is an issue now around the newer technology. The danger is now shifting away from some of the work that we’ve done on chat rooms and we need to work out now whether that’s applicable to where children are now which tends to be within comment sections on YouTube within facebook and snapchat. And apps like snapchat and yik yak also have the element around the lack of permanency of what you put on it. So we don’t currently know how well the concepts apply to those, to be honest with you. It’s a bit of a minefield for us as well.

And I presume that’s because technology moves so quickly? It evolves so quickly and children adapt with it.

Oh absolutely, it tends to be that the technology moves on and academia is probably about 10 years behind that and then the law is probably 20 years behind academia. So it’s always kind of an arms race. But we’ve got a good idea of the kind of psychological processes that go on, it’s just whether or not we can apply them or whether the same things do apply when you look at this new technology.

So you said academia is 10 years behind and the law is probably 20 years behind that, where do you think the schools are? Do you think that they are in tune with this? Do you think that they are equipped for this?

I think they’ll naturally be behind the technology because that always drives everything. But I actually think the schools do pretty well with this kind of stuff because it comes down to accountability. Obviously they are fully accountable for the children within their schools and school districts. I actually think they are reasonably good at it.  And I think that probably comes down to the fact that they are closer to the ground, and just see a lot of this stuff much quicker. They’ll see their pupils interacting with the technology. And also this idea that they are probably directly accountable for the safety of the children. So there’s a little more pressure for them to actually do something.

From your research do you think that children often opt out of the radicalization (or grooming) process? Do they ever recognize the position they are in, that they may be getting manipulated, or does it always seem to carry through to the point where the adult sees what’s happening and has to get involved?

Again my background is working with sex offenders, other than a brief review of literature that was a sweeping exercise, I don’t know a huge amount about recruiting children for violent extremism, apart from trying to apply some of the ideas across.

It’s a difficult question to answer because you are trying to count unknowns. I would imagine that a pretty sizable majority of children would be able to recognize that they were being exploited or would be able to recognize the intentions of the person who’d approach them and would be able to recuse themselves from the situation.

One of things that I will point out is that in the literature and in public opinions of people who would engage in this kind of behavior –  (I imagine that it would be the same for extremism) – there is a hindsight bias that believes that these individuals are somehow particularly good at finding vulnerable children. And I think that the only time we tend to see it is when something bad has happened. So it tends to look like targeting is something they’re very good at. That groomers are very good at identifying vulnerabilities and being able to exploit those vulnerabilities.

I think that’s a hindsight bias – looking at cases where things have gotten to that stage and therefore it looks like they are very good. And naturally I think the vast majority are pretty poor at it, I think most children probably have the intelligence and the agency to get themselves out of the situation. Obviously we have to then care about the ones that can’t do that. But my guess would be that the vast majority of children would be able to identify that.

So you seem to think that the recruiters that are successful just happen to stumble upon one vulnerable child out of many?

They have systems in place. We’ve noticed that with – and this is going back to online transcripts around the 2006-2010 time period – what they would do is go on, start conversations with as many children as possible, and then have some early questions around the supervision of that child, the mental welfare of that child, their integration at school, things like that. Things that they can use to then identify where the most likely success they’ll find is, in terms of how much resource to put into each child that they are talking to. That allows them to target children that they believe are vulnerable.

 It’s an incredibly unsophisticated kind of scatter gun approach, where they’ll just approach as many children as they possibly can and then pick two or three of those children to put their conversational resources into. I think the vast majority of the times they are probably wrong in the children they pick. But again we only tend to see the ones that end up in exploitative or abusive positions. So I think they have ways to do it but they are quite unsophisticated ways of doing it and I think that they fail more often than they would succeed.

 And of course we think they are highly successful because we see that in the media.

Exactly. It’s the cases that you see are the successes, for a lack of a better word, but for the groomer they are the successes.

Some work that I have been doing recently has been looking at transcripts so I don’t know how applicable they’d be to the new technologies.

I don’t know if you are aware of the TV show “To Catch a Predator”? The organization that acted as decoys in those cases actually published the transcripts online. So we went through and had a look at those transcripts to work out the proportions of certain words we considered to be important and at what time period during those time periods they occur.

And one of the things that we found is that in the first 20% of conversations, there’s very little talk - again these are sex offenders – about sex or meeting up. But there is an awful lot of talk related to flattery and friendship, and an awful lot of talk about security. By security I mean there’s a lot of talk about where are your parents, where is your computer, is anyone watching you? And at the same time, are you a cop? Is this a sting? Are you real? All that kind of stuff.

So what we think is that there is a period of probably the first quarter or fifth of the conversation where they have what we call it a triage system. That’s a medical term, where when you go into a hospital they take all your information down and prioritize the patients in terms of how at risk they are. So it’s a really simple process that they start conversations with as many children as possible and then whittle those children down to one or two.

People often say “you are one click away from stumbling on inappropriate (or harmful) material.” How true do you think this is? Are kids really only one click away?

No, I think a one click away makes for a nice ‘sound bite,’ but it’s not particularly true. I think people already have to be partially down the ‘rabbit hole’ to some extent.

That rabbit hole is probably much shallower with extremist material. I think so much pressure has been put on organizations – take Facebook for example – that you would never find illicit sexual images of children on their site, and if you did it would be immediately removed. It would be incredibly difficult to promote sexual material like that.

But with extremist material, it’s still available on those sorts of platforms. I think you already have to be in an interesting place but currently you could be a magnitude of clicks closer to extremist material than you would be for other kinds of illegal material on the internet.

I think that is because we have had almost 30 years to respond to sexual material but the violent extremist material is much more recent and we haven’t yet fully adapted.

So to clarify it is unlikely that any random child (one who isn’t necessarily directly vulnerable) would come into contact with a groomer? A child would likely have to already be partially down the rabbit hole to be exposed?

Yeah, I think it would just be a case of having to think of it the other way around.

In that scenario the logical conclusion would have to be that a child accidentally ‘stumbled down the rabbit hole,’ and I don’t really buy into the idea that you can accidentally get into some of these situations. Now, that is not to say it’s someone’s fault for getting into that situation, they may have not necessarily incriminated themselves by taking certain steps that have led them to that point. But, the idea that someone can ‘accidentally’ find this stuff is a little unbelievable to me.

I mean I have had sexual offenders I’ve worked with who say they first saw this type of material in a pop up, I mean its 2017 are there still even pop ups on chrome? Are they even technologically possible anymore?

I guess stumbling upon the material is the best way to describe it, but you have to have already been moving in a direction that has led you to a certain place. It might not be that the path you were on was ‘extremist,’ but you can find your way onto that path through other avenues I think.

I am sure there are cases where people have been reached out to out of the blue, but I think in the majority of cases people were down the rabbit hole to some extent.

What you think? I think you all know more about the extremist material issue than I do!

I think I agree with your sentiment. I believe most children have started their own journey down that hole, and a recruiter found them along the way and pulled them all the way down.

I also think that not all of these kids (on an extremist path) were heading down an extremist path, they were just out there looking for something, but they didn’t know what they were looking for (possibly identity, community). A recruiter finds them and gives them something, and it may not have necessarily been what they were looking for originally, but it may address their needs:

I find it quite interesting that you mentioned the aspect of children looking for an identity. It’s an interesting concept. I am going to default to sex offenders because that is what I know – there was a huge project funded by the EU that looked into not only perpetrator behavior but also victim behavior, to see whether they could pick out any vulnerabilities that might be shared across that group.

Their goal was to see if they could try and find prevention opportunities. One of the things that came out of that – and again it was controversial because some accused them of victim blaming (which is ridiculous) – but it was the idea that adolescents have a burgeoning sexuality and that they then go and explore on the internet. That then can lead them into places where they are more likely to have conversations with adults. Or they are going into chat rooms and having conversations with their peers, but adults find them and exploit it.

So yeah it could be the same thing, you have a group of individuals with a burgeoning interest. Say for example you have someone who has a burgeoning interest in a certain religion or a certain social movement, this could lead them to places that could make them slightly more vulnerable in the sense that their curiosity in a topic could be exploited. Again the goal of that individual is perfectly legitimate but it could lead them to a crossroads with people who would be interested in exploiting them.

In the conversations that you have examined with sex offenders online, does there seem to be a check list of things that groomers will typically talk about in conversations? Or is there not one? Is every single case seem to be different?

I was working on a project a couple of years ago, I was working to distill the models of grooming in the sex offender literature. A lot of the previous work that was done was qualitative, my concern became (or my realization was) that this sort of behavior is completely goal-driven – the individual has a goal they are trying to achieve. The goal in many instances was to meet to engage in sexual behavior. With a violent extremist recruiter you are trying to recruit targets to a violent extremist ideology and towards violence for their group. You have goals that you are attempting to achieve.

One of the criticisms I had with the previous literature (surrounding online grooming) was that they were overly interested in what the themes were of the conversations. I suggested those themes could be completely different depending on the person, it is about the goals. If you have something you want, you will happily say whatever it is you need to to achieve that goal. And so, trying to find out what the common topics of conversation are might just send you in a misleading direction. If you come out and say ‘well they’re going to come out and say x, y, z’ it isn’t going to apply to all cases.

Recently there has been a push more towards processes (advocated largely by John Horgan). The idea is to understand the underlying psychological processes going on (on the part of the offender) as opposed to what are the surface level conversations. I was looking for the common processes that you see across all types of grooming, and if there was any way we can apply these to more general psychological concepts as opposed to just forensic psychology. And I wanted to see if we could take the sex offender out of it and find more universal grooming tactics that we would be able to apply to recruitment for violent extremism, gangs, etc. The processes are generally the same, it’s the surface conversations that are going to be different.

“Themes can be completely different depending on the person, it is about the goals. If they have something you want, you will happily say whatever it is you need to to achieve that goal.”

There almost always seems to be an effort (regardless of the timing in the process) to build rapport with the child. There’s this almost constant relationship-building.

There was also a lot of incentivization financially whether it be with money or gifts. It could also manifest as coercion with bribery or blackmail. The groomer might also try to convince them that sexual contact is good for them, that they would somehow benefit from sexual contact, and convince them that it could be somehow educational or instructional.

There may also be an attempt to alter the child’s ability to make effective decisions and to their ability to identify risky situations. So in offline grooming we’ve seen the use of alcohol or drugs to disinhibit the victim. The internet itself might have the same effect. It’s possible that groomers choose the internet to exploit the fact that people (including children) act differently on the internet if they feel anonymous.

The last element that we identified (as a universal aspect of grooming) was that there was some sort of security management. Ensuring the target being spoken to is vulnerable and that the likelihood for the groomer of being caught is limited.

All four of the aspects I just talked about before are part of this idea that these initial goals can serve as a conduit to accomplish other higher-level goals. The higher-level goal for sex offenders is to desensitize the target to sex. With violent extremism it would be desensitization to violence and desensitization to extremist ideology.

In a typical desensitization process, what you do is you slowly start to introduce goal-relevant information in larger quantities over time. It will eventually desensitize the target to that information. Say you wanted to desensitize someone to violent extremist material you would slowly drip-feed ideas of violence and violent imagery over a period of time. Or you could introduce it very quickly in high quantities and maintain those high levels over a longer period of time. Ultimately, it’s about exposing people, over time, to relevant information related to the goal that you are trying to achieve. 

You said researchers have been having issues with viewing discussions/chats with more recent groomers and their victims. What do you think the reasoning is behind that? Why is it more difficult now, has it always been that way?

I think it is a combination of the fact that technology is just naturally changing and you don’t have chat rooms as they used to be 10 years ago. With the rise of social media, a lot more of that conversation occurs in private, you just don’t get access to it. Now we are seeing a lot of bad stuff happens not only in private but on networks/ apps that automatically delete messages (snapchat).

It’s going to be a lot more difficult to do research on this, in the sense of you basically have no data to go on. People rarely use chat rooms anymore, where the information is right there in front of you. It’ll be more difficult to get access to that kind of material. As a researcher it has become difficult, but also for law enforcement.

People’s concerns about online safety is making everyone more private online. That is making it a lot more difficult for researchers and law enforcement to get access to information about people’s behavior on the internet…. as technology becomes more encrypted and allows for more private use, its extremely difficult to do any sort of tracking. It’s going to become more difficult to document what’s happening, and therefore more difficult to design preventative measures.

So it is already severely affecting research. Current grooming models seem very accurate in how they address the psychological evolution of the child, and the groomer likewise. Do you think the models will evolve (with technology) or is it likely that the model will adapt to the platform (sites, apps, etc)?

The idea of the model was to draw the specific crime out of it and look at what the underlying general psychological processes were.

We did this because of the following criticisms we had of current sex offender literature: 1 so much of it overlapped with general ideas of seduction and adult relationships, and 2 there were instances of just general everyday behavior of people who engage in this kind of stuff.

If, for you to achieve a goal, you need to recruit someone else to join you because either the goal involves them or because you need their assistance, you’re going to have to provide them with some incentive to do it, going to have to be nice to them, etc. These are processes we use in everyday behaviors, but processes that can also be used to achieve illegal or exploitative goals.

So the idea behind the model is trying to explain those processes in a way that would be generally applicable to more of these exploitative behaviors – not just online or offline, and not just for sex offenders. So, I would hope that it would allow us to apply it to the new technologies

What scares me more on a practical level is the inability of law enforcement and individuals who are interested in preventing these type of crimes, to access this data. Police do have tools to at least monitor the sharing of material online, but there are obvious balances to make with personal freedoms and civil liberties. There is a concern of chasing people away [from websites/material they are allowed to have access to] and trying to balance good law enforcement and trying to prevent people from viewing horrible but legal material.

Operation250 would like to thank Dr. Elliott for providing us with a great overview of online sexual grooming and its effects on online safety behavior. For more on Dr. Elliott's work he can be found at: