 I'll just like to start by thanking my colleagues that are listed here, Zoe, Elsa, Elsa's in the audience actually, and a few others. The work I'm going to talk about today is in collaboration with my fantastic colleagues. There's no doubt that academic research has helped to improve policing over the last 30 or 40 years. Susan and Jean have both described the new collaborative, innovative research centres that we work with. We've worked really closely with practitioners, but in the past there has been a gap between the way that academics work and the way that practitioners work. That gap is starting to close now, thankfully, and there's a huge amount that we can learn from each other. So, as academics, what can we bring to this relationship? There's been a lot of talk lately about evidence-based policing. What does that actually mean? The police deal with evidence every day. Surely, they're already evidence-based practitioners. When we talk about being evidence-based, what we mean is the questions that we ask of evidence. So, the way you evaluate evidence, the fact that you never stop asking questions of the evidence and you never take anything at face value. So, it's what we spend our time as academics, as researchers, as teachers drumming into our students. We don't necessarily ask where is the evidence. You can find evidence everywhere. We're surrounded by evidence. You can usually find evidence to fit any story you like. We ask, what is the evidence? It usually comes from multiple sources. What is it based on? How good is that evidence? For us, there is no proof. There is only evidence. If we're lucky, we can get a large body of evidence which might suggest that one thing is slightly more accurate than another. So, these questions are so important because we're surrounded by evidence, as I said earlier, and a lot of it is absolute nonsense. Here we go. So, I found this. Ice cream consumption leads to murder. So, I've solved crime on the third slide. All we need to do is get rid of, stop people from eating ice cream and then one reduced the murder rate. So, not only have I solved crime, I thought I'll go the whole hog and I'd solve global warming at the same time. So, a pirate shortage caused global warming. You can never have too many pirates, that's what I say. So, we can laugh at this and think it's ridiculous, but what we mustn't do is evidence-based researchers and practitioners is dismiss it out of hand completely. I know that it's unlikely that the information in these graphs is inaccurate because I can understand statistics, I understand research methods, and I know that just because there's a rise in one thing and it doesn't cause a rise in another thing, things can just be related without one causing a change in another. So, issues like this and this way of thinking is one of the big things that we can bring to our relationship with police and practitioners. It's not about teaching the police how to deal with forensic evidence at the scene of a crime. They know how to do that, they're trained in that, but it's about becoming evidence-based, open, critical thinkers. Okay, so as well as that, we can also provide evidence about issues that are important to police. Okay, quite a lot of the time we research things that interest us, but they're not necessarily important for police and practitioners. The rest of my talk is going to be focused on some of this research that we've been doing as part of our policing collaboration that Jean mentioned earlier. The police approached us and said they were worried about issues concerning social media and could we investigate it, so we've started to do that now. We're looking at a number of different things, the role of the citizen, the way that the police engage with social media, and we're using a broad range of evidence. We're doing surveys with public and police, we're looking at in-depth analysis and we're using experiments. Unfortunately, I can't talk about the results today because we don't quite have them yet, but I'm going to talk about the issues and what we're doing. Okay, so social media has had a huge impact on policing. They now have access to huge amounts of information, amounts of intelligence, they can monitor posts, they can analyse language on Twitter. You can look at connections and influences, look at friends, photographs, video. It's quite often put on Facebook with a time tag, a place tag, other people are tagged in it, and it can give insights into behaviour and crime. So it can help to solve crime. Also, as Karen was talking earlier about prevention, this type of information may help to predict and prevent crime. Not in a huge way, I'm not saying it's the answer, but it may. So this type of information is available to everyone. It's available to the police, but it's also available to us. So what are we doing with this information? So, a question for you. If you were at a party and you got home and realised that someone has stolen your phone, what would you do? Okay, hopefully you'll find police. But would you do anything else? This is becoming increasingly common. People are starting to conduct their own investigation on social media. So they go home, search for the person, perhaps search for a party or an event, search for photos of the party, look for friends of friends, look at profile photos, and they sit and try and piece together their memory of what happened where they were at a particular time when they thought their phone was stolen when you saw certain people. What's happening is that people are doing this. They're turning up at the police station with information, with Facebook information, with photographs, and they're saying to the officer on the desk, I have my bag still on this last week. This is the person that stole it. So the issue for the police when they spoke to us is what do we do with that information? Is it reliable? Will it impact on formal investigation? So, in response to these issues, and to this increase in citizen investigation, the Association of Chief Police Officers published guidelines in July of 2014. And this is for Facebook IDs. Now, when I looked at them, I thought they're really detailed on procedures in terms of how to document the procedure, how to document the way the evidence was obtained. So I've got questions here like what was shown on the screen which allowed the ID to be made. Was there any prompting? Was there anyone else there? How long was that screen viewed before the ID was made? This is just a few of them. There's a big, huge long list. Now, these are quite good, but there's a but, and it's quite a big but. There are issues. Now, the obvious one here is if you look at some of these questions, they rely on memory, don't they? If I asked you an hour ago, you were looking at your phone, and how long did you look at a particular screen for? How many screens did you look at? It's relying on memory, and we know that's fallible. There's also an assumption here, in this part of the guideline, that a social media ID is no different to any other type of informal identification. So it says it's comparing it here with an accidental sighting, a street ID, seeing a picture on crime watch, or by any other method. Now, the issue for me as a psychologist with expertise in face recognition, I look at this and think, oh, okay. Actually, in some circumstances, it might be. It might be the same, it might be similar. In some circumstances, you may know the suspect. You may go home, you may look at Facebook, grab a few photos of the event, show that that person was there, grab some information, take it to the police. That's fine. If you know the person, that makes a robust, reliable identification. The issue comes if you don't know the person, and you only met them that night at that event. Okay, we know that we're really, really bad at recognising people that we've only seen once before. So out of all the issues that I've listed at the bottom, familiarity is the biggest one. If people are searching through dozens and dozens, sometimes photographs, it's also going to contaminate memory. So what do I mean by familiarity? I'm just going to give a little example here. Are these the same or different people? Different, it's easy, isn't it? You don't know them personally, but they're famous. It's an easy task. What about this one? Are these the same or different people? Same but different age. That's really interesting. Okay, they are the same person. I thought this was a tricky one. You're too good. But the photographs were taken at the same time. Okay, they're not different ages. So quite often what we find is two photographs of the same person can look more different than two photographs of different people. And we're really, really poor. There's a whole wealth of evidence to show that if we haven't seen the person, we're very poor at matching. Images like this is about 30% error rate. And if you think about that in a criminal investigation, 30% of the time you're going to get the wrong person. So it's just not really good enough, is it? How about verifying identity from different photos? So you might come across photos of an event that you were at, and you're trying to compare those photos with somebody's Facebook profile. These images, for people that don't know the person, which is actually three images of me, believe it or not, I thought I'd be brave. My passport, my terrible driving licence photo, and a Facebook photo. Okay, it would be really difficult to verify that this is the same person in these three photographs. And that's exactly what people will be doing on Facebook. So there are huge issues around this. So at best, what we think at the moment is that a Facebook identification could render a formal identification redundant. So if that person is then asked to go and do a video line-up or any form of formal ID, they'll more likely be accurate and pick the wrong person. Okay, at worst, this kind of ID could result in unsafe evidence. So the guidelines are as good as they can be at the moment, and they're quite detailed in some aspects, but there is a huge lack of research on identification on social media. So this is where we come in, this is what we're looking at now. So we're currently investigating different forms of identification on social media by using mock Facebook procedures. We can't use Facebook unfortunately, but we can set up experiments that look very similar with different profile photos and lots of different event pages. So that's one of the things that we're currently working on. Okay, so citizens are engaging with social media. They're trying to solve crime, they're trying to be helpful. Karen's talked about solving crime isn't just about the police, it's what we can do as citizens. People are trying to help you, but there are issues. We don't really understand the full impact of what they're doing yet on formal procedure. So citizens are engaging with Facebook, but the police are also engaging with Facebook. Ever since 2008, there's been a real interest in engaging with the community. Back in 2010, the National Police Improvement Agency published a document called Engage, and it was about procedures to do with using social media, Facebook, Twitter, etc. And it contains some nice information. There's some nice case studies in there. But it tends to be focused primarily on given information. So given safety information, given community information, given crime updates, rather than engagement as we would view engagement. If you look at the research as well, that kind of idea of given information is borne out and a lot of the research around that time as well. So I've got a nice little example here. Greater Manchester Police in 2010 decided to tweet every incident over a 24-hour period. So over 3,000 tweets. Their Twitter followers increased from 3,000 to 17,000. And the chief constable said, well, our aim of engaging the public and informing the public has been achieved. Has it? Have they really engaged with the public? So engagement was seen very much as given information. And this type of engagement is seen all around the world. Just about every police force around the world has an online social media presence now. And it seems to work very much as a way of delivering messages. But it can be used for proper engagement. There are avenues to build relationships with the community, to communicate directly with the public. And there's some nice research coming from Canada where O'Connor looked at not just the amount of tweets and how people were interacting with tweets but what was in them. And he came up with two themes, the informational information, giving out the safety messages, giving out the crime updates. But there are also interactive ones where there was a real effort to try and engage with the public. So the police were asking for responses, asking people to attend events. And they not only did that, but they were trying to engage in a meaningful conversation. So they were responding to messages that other people have posted. And when they did that, they often mentioned others. So there is research showing that it can be used in a really meaningful way. But social media opens up the possibility for engagement, but also public scrutiny. The fact that the police have a Twitter presence, have a Facebook presence, shows that they've really opened though to this public scrutiny in a way that they haven't been in the past, which is really good. But there are risks. There's negative publicity, there's criticism, sometimes criticism comes from anonymous sources as well on Twitter. But there are also benefits. And we need to think about whether there's risks outweith the benefits. So it's a couple of nice examples, which I think of as benefits, is that social media can put a human face on the police. It can put a human face on a huge organisation that's often seen as faceless. So the first little image here I've got is back from 2012, and this is when the New York officer bought some new boots for a homeless man he encountered in Times Square. This was tweeted and retweeted hundreds of times, and it endeared him to the whole population. This recent one, a very recent one, which I've absolutely loved, partly because I've never seen Scottish Police describe this fun-loving before. And that was by the Daily Record. It's the Running Man challenge. If you haven't seen it, go home and have a look on YouTube. It's hilarious. But our Scottish police force took on the challenge with full force, if you like. But these are nice public interest stories. It shows the police are human, and they are taking their message directly to the public, bypassing the media, and bypassing all the interest that the media have. OK, so I'm 1-0, but I'm on my last slide. OK, so I'm just going to finish it again about what we're doing now and how we're engaging with the police. I've talked about some of the issues, and these are really under-researched issues. One of the things we're doing is we're doing surveys, looking at all aspects of social media, and we're surveying police forces in our consortium and also the public. We're also doing in-depth analysis of Facebook posts of some of our consortium members, some of our police forces. What we want to do is really go beyond pre-research and consider how information and how identity is constructed on Facebook. So as a memory researcher, I know that every time I tell a story, it changes. And my memory of what happened to me originally is different because I've told the story. And the way I've told the story and who I've told it to will impact on how that changes. So that same change occurs when you talk online, when you write stuff. Your memory changes. So, we want to look at this in terms of how the police construct their identity, how they construct their story, and how that's interacted with the people that are responding. So, big questions here to ask. What does social media do? Does it help to construct identity? Does it build multiple identities? And now I think this is an ideal time for police Scotland to think about issues like this. All the forces have come together, they've formed into a single force. So, does that mean it's going to be a single identity? A collection of identities? And what should the narrative be for police Scotland? How will it be constructed? And what role does social media play? There we go. Thank you very much.