Interview
James Bonta
Ph.D. (Clinical Psychology), Co-author of “The Psychology of Criminal Conduct” (RNR model), Canada
Widely recognised for being part of the transformation of how we understand and deliver effective rehabilitation, Dr. James Bonta has spent decades researching how to reduce reoffending through evidence-based practice. In this interview, he reflects on the origins and evolution of the Risk-Need-Responsivity (RNR) model, common misunderstandings about what works, and where research and practice must focus next to deliver lasting change.
Based on current state-of-the-art, what do we reliably know
works in offender rehabilitation?
JB: I’m sure it will come as no surprise that I’m going to talk about the Risk-Need-Responsivity model (RNR). That model arose out of our concern in the 1980s about the growing “get tough,” punishment-oriented movement. People were almost accepting the viewpoint that nothing works. We wanted to make a statement in response to the “nothing works” movement, which had already been gaining momentum. So, in 1990, we published a review paper in Criminal Justice and Behaviour, where we introduced the RNR model.
In that 1990 paper, we introduced the risk principle, which states that more intensive services should be provided to individuals at higher risk of reoffending. We also emphasised the importance of targeting criminogenic needs rather than non-criminogenic needs.
We then described the responsivity principle, which has two parts. The general responsivity principle – which suggests that cognitive behavioural interventions are generally more effective – and the specific responsivity principle – which considers individual motivations and personal characteristics, meaning that the approach may need to be adjusted based on those characteristics.
Shortly after, we published a meta-analysis which provided empirical support for the model. That second 1990 paper showed that the more closely a programme follows the three RNR principles, the better the outcomes in terms of reducing recidivism.
Since then, a large body of research has been published. There are now meta-analyses supporting each of the principles: the risk, need, and responsivity principles. The evidence spans across many different groups: women, youth, violent offenders, and even sexual offenders. So, what reliably works? Treatment that follows the RNR principles.
In your view, what are the most common misconceptions the public or policymakers have about what works in offender rehabilitation?
JB: I think the biggest misconception is this idea that a reduction in recidivism of 10 or 15%, which is what we typically see in real-world studies, “doesn’t sound like much.”
My response to that is to compare it with what we see in medical sciences. Take the famous aspirin study, for example, which looked at giving people a daily dose of aspirin to prevent heart attacks. The reduction in deaths was 3%. In the context of cancer treatment like chemotherapy or radiation, the improvement in survival rates is often around 10%. Most people will not pass on treatment because it only helps 10% of the time.
The changes we are seeing are important. The big problem is that policymakers don’t seize on that progress, and they don’t advertise that information.
The other big misconception coming from policymakers and politicians is that they misread public support for rehabilitation, assuming that people only want punishment and imprisonment. But when you look at surveys, the public, in general, supports rehabilitation.
I think that part of the answer comes down to the media. Crime has become entertainment. Researchers even have a term for it: “crime theatre.” You turn on the television and it’s wall-to-wall detective shows, forensic dramas, mafia movies, you name it. That creates a distorted image of who justice-involved individuals are. That is not the typical person you encounter in prison or probation departments. The vast majority of people involved in the justice system have complicated life stories that have brought them into conflict with the law. They’ve made bad choices at different points in their lives, but they’re not Hannibal Lecters.
“Evidence-based practice” is widely used in the field today. To what extent do you believe this is reflected in genuine adherence to empirical research and rigor in practice?
JB: Before I retired, I was Director of Corrections Research at Public Safety Canada. Wherever I went, I would talk about the RNR model, and I’d explain that it’s the model that shows the most evidence for effectiveness. In most of the places I visited, whether it was probation systems or prison systems, the heads of those agencies would often come up to me and say “We already do that.“
One year, I pitched a study at the Heads of Community Corrections meeting that asked a simple question: “What do probation officers actually do when they sit down with their clients?”.
One province agreed to participate, so we started looking for volunteers. We asked probation officers to audio-record their supervision sessions with clients, and to give us access to their case notes and files. We ended up with about 72 officers participating and published a paper on the study that some people call the infamous “black box study”, because we looked into the black box of community supervision.
We found that low-risk clients were being seen just as frequently as high-risk clients, which means the risk principle wasn’t being followed. When we listened to the audio recordings, did they address criminogenic needs such as substance misuse, or pro-criminal attitudes? Well, there was some moderate adherence to the need principle. They would talk about employment and substance use, but not about criminal attitudes or pro-criminal associates, so the need principle wasn’t followed very well. And what about cognitive-behavioural techniques? That was virtually absent. Across all the audio recordings, we found no evidence of cognitive-behavioural interventions.
So, is evidence-based practice actually being practiced? At least back in 2008 in Canada, the answer is that it certainly wasn’t. And it’s not just Canada. Other studies have used tools like the Correctional Program Assessment Inventory and applied them in different settings. What they find is that evidence-based practice is not being used, whether it’s in Canada, the United States, or England. So yes, there’s a lot of room for growth.
It took a lot of bravery to participate in that earlier study I mentioned. It required courage for that head of probation to say, “Okay, come into our province and do your study. Whatever we find out, we find out.” Because they might discover that they’re not as effective as they thought.
Fear is the biggest barrier to implementing evidence-based practice. There’s a general nervousness about exposing yourself and admitting that perhaps you’re not doing evidence-based practice.
How important are the skills and approaches used by correctional
staff in day-to-day interactions? How can these skills be integrated
into current training?
JB: I think it’s highly important to reflect on what staff actually do with their clients. Take probation, for example. There are thousands of probation officers around the world. Many of them meet with their clients regularly. If a client needs treatment and is referred to a psychologist or psychiatrist, they may only see that professional once a month – if they’re lucky to access that service at all. So, wouldn’t it be better to train probation officers to be more therapeutic, rather than simply policing their clients?
In the “black box” study I conducted with provincial community corrections in Canada, when we presented our findings, they realised that maybe their staff were not following RNR, and began looking into how to train probation officers to apply the RNR principles.
We developed a training model called STICS – Strategic Training Initiative in Community Supervision. It was a 3–4-day programme that included monthly meetings and refresher sessions. Three provinces came forward and volunteered. Probation officers agreed to audio-record their sessions, attend the training, and participate in monthly meetings.
We found out that officers who received STICS training significantly increased their focus on criminogenic needs. Most importantly, these officers demonstrated cognitive-behavioural intervention techniques: pro social modelling, homework assignments, role-play, cognitive restructuring techniques. In a two-year follow-up, we found reductions in recidivism.
Ultimately, we implemented STICS across the province of British Columbia and, in our evaluation, we found reductions in both general and violent recidivism, again, in the 14–15% range, which is consistent with real-world treatment outcomes.
And now we’re looking forward to the next step. British Columbia is in discussions with Statistics Canada to possibly conduct a cost-benefit study. They continue to apply the STICS model. There have been adaptations specific to their context, but they’ve retained the monthly meetings, expert feedback, and ongoing professional development.
What areas of research or innovation do you believe hold the greatest promise for further enhancing rehabilitation outcomes?
JB: There are several important developments happening in relation to the RNR model. There’s a growing body of research aiming to better specify these principles, particularly to examine how well they generalise to different groups of offenders and across various situations. For example, when it comes to the risk principle, Kimberly Sperber and her colleagues are doing some interesting work on the question of dosage. Specifically, how much treatment is required for a medium-risk client compared to a high-risk client?
In the early stages of this work, dosage was typically quantified in terms of number of hours. But now, researchers are going beyond just counting the number of treatment hours. They’re asking about what exactly is needed more, is it role-playing, homework assignments at the end of a session, so on. This area is developing quite quickly.
With respect to the need principle, there’s active discussion around the Central Eight criminogenic needs and how generalisable they are across different populations. So far, the Central Eight have worked quite well across different offender groups and situations.
Cognitive-behavioural approaches have strong empirical support. Not only is it effective with criminal justice populations, but it’s also widely validated in the psychotherapy literature for treating issues such as depression and anxiety.
However, the real gap in the literature lies in specific responsivity. We need closer investigation into what specific adaptations are needed for subgroups, for instance, how it applies to trauma informed care. What exactly needs to be done differently with someone who has experienced a trauma than with someone who has not? Similarly, you can look at women or marginalised groups and ask the same question.
What we’re seeing now is the RNR model being applied to different offender groups. Researchers are currently testing it in cases of intimate partner violence to see whether it works in that context. Personally, I’d really like to see more work in sex offender treatment.
On the other hand, we’ve just published a new study applying the RNR model to individuals with serious mental illness. That’s a growing area. Ten or fifteen years ago, there were no studies looking at RNR with this population. The focus was mostly on symptom alleviation, not on changing criminal behaviour, so to speak. But what we found in our study was that applying RNR principles can make a difference also with this group. So there’s still more to learn, and that’s where I see things moving in the years ahead.
Dr. James Bonta
Ph.D. (Clinical Psychology), Co-author of “The Psychology of Criminal Conduct” (RNR model), Canada
Dr. James Bonta is a clinical psychologist and one of the leading figures in evidence-based offender rehabilitation. He earned his Ph.D. from the University of Ottawa in 1979 and began his career at the Ottawa-Carleton Detention Centre, where he later became chief psychologist. From 1990 to 2015, he served as Director of Corrections Research at Public Safety Canada. Dr. Bonta has held various academic and professional positions throughout his career and was a member of the Editorial Boards for the Canadian Journal of Criminology and Criminal Justice and Behaviour.
He is best known as co-developer of the Risk-Need-Responsivity (RNR) model and co-author of The Psychology of Criminal Conduct. His contributions have earned him numerous awards, including the Canadian Psychological Association Criminal Justice Section Career Contribution Award (2009), the Queen Elizabeth II Diamond Jubilee Medal (2012), the Maud Booth Correctional Services Award (2015) and Community Corrections Award from the ICPA (2015).
Advertisement
