Can we rethink nicotine addiction? Is the word "addiction" helping—or hurting—the conversation? In this episode of THR Unfiltered, recorded during the Global Forum on Nicotine 2026 #GFN26 we speak with behavioral scientist Dr. Arielle Selya about the science behind nicotine dependence, addiction, and tobacco harm reduction. The discussion explores why researchers often prefer the term "dependence" over "addiction," how these concepts differ in scientific and clinical contexts, and why language matters when discussing people who use nicotine.
Transcription:
00:03 - 01:30
[SPEAKER_00]
Good morning from Warsaw. And hello, whatever time of day it happens to be, wherever you happen to be watching this. This is the second edition of Tobacco Harm Reduction Unfiltered. And I'm here with Arielle Selya, who's a behavioural scientist. So I better watch how I behave myself. And we're talking about how we should think about nicotine addiction. And we'll get on to maybe more controversially whether we should be talking about addiction at all, even if it's the right words to even be using. Good morning. Good morning. Nice to see you here. Likewise. And I got all this from ChatGPT, so it must be right, you see. We're having this chat now, and I had to look at ChatGBT to sort of suss out what you'd been doing, having a look at your CV. And it seems like all your early work was focused on nicotine dependence in adolescent smokers. That was how you started. But I noticed that you, at least in the headlines and the summaries, you didn't use the word addiction. You used the word dependence. So can you explain that choice?
01:30 - 04:16
[SPEAKER_01]
Yeah, so a couple of reasons for that. So when I... The work you're referring to is when I was first getting started in the field of tobacco behavioral research. And this was during my postdoc back in 2010 and 2011. And cigarettes were still the predominant product there. And that's how I kind of got my research skills, got familiarized with the area. And the project at the time under my postdoc advisor was basically trying to understand why some kids who experiment with smoking stop after the transient experimentation and then they just leave it versus why they keep continuing and eventually become smokers. So there was a large project, a longitudinal survey that we worked with a lot trying to get adolescents at that early stage of experimentation to try to understand why some of them continue on and others don't. And nicotine dependence was one of the strongest predictors. And there's different ways of measuring this. I think initially the reason why I use the word dependence was because it's precise scientifically, because there's different validated scales that you can use to measure dependence. And it wasn't addiction, like there's no real formal term for addiction that has specific criteria and the way that dependence is. It's a nicotine dependence scale. So that was the original rationale for it. But of course, lately, things have shifted a lot. E-cigarettes have become the dominant product in a lot of countries, including the U.S. And the word addiction doesn't care. It's not as relevant when talking about e-cigarettes. because first of all, the dependence scores are lower. If you're an exclusive e-cigarette user, you're likely to have lower dependence symptoms than cigarettes. People that switch from cigarettes to e-cigarettes can lower their dependence levels, but it's tricky to measure this sometimes too, because not all measures are comparable across products. One of the best examples or one of the best ways to measure dependence for cigarette smoking is how soon do you smoke after you wake up in the morning? And within five minutes is the most highly dependent category. But that item doesn't really translate well. If you look at the science on how people respond to that and how well that captures the construct of dependence, it doesn't really work the same for e-cigarettes. But you have all these studies out there that use that as a measure, assuming that it transfers to e-cigarettes as well. So it's tricky to measure, but that's... kind of all goes into why I've always used the word dependence. And now today it's more because I'm more cognizant of like the stigma of calling somebody an addict.
04:16 - 04:59
[SPEAKER_00]
Yeah, we're definitely gonna come back to that in a minute. So it tends to suggest to me that there is no actual clinical definition difference between addiction and dependence. Because I have read somewhere that dependence is more of a physical thing, addiction is somehow regarded as more of a behavioural thing, compulsive use and all the rest of it. But from what you're saying, there is no, you know, established clinical difference between the two, and people seem to use them interchangeably.
04:59 - 06:24
[SPEAKER_01]
Yeah, so I think definitions have evolved as well. So if you look at the Diagnostic and Statistical Manual of Mental Disorders, that you can go through a mistake that like junior psychology students make, for example, is like once they read the DSM, they're like, oh, that applies to me. I must have OCD. I must have depression. I must have bipolar because everybody has the occasional symptom of some of these things. But the question in the mental health literature is how do we distinguish somebody that has a clinically meaningful disability barrier of these symptoms versus somebody that has the occasional symptom, but it doesn't really impact their functioning of life. And across many of these diagnoses, the consistent criteria for determining whether it's clinically relevant is does it impact your daily functioning? Does it impact your quality of life? So I think that there have been attempts to fit tobacco dependence into that mold. And I think the analogy there or the parallel in tobacco dependence is, are you using it despite the harm to yourself? And cigarettes obviously cause a lot of harm, so there is a case to be made that maybe it qualifies as an addiction or a compulsion if you're using it despite experiencing negative health effects. But if you take most of the negative health effects away, what's left?
06:25 - 07:02
[SPEAKER_00]
Well, quite. Yeah. Yeah, absolutely. You've obviously moved into the tobacco harm reduction space from your work. So how has your work developed? Have you carried on? transferring what you've been doing with cigarettes into the world of e-cigarettes or easy tobacco. Has it changed? How has it developed in the context of dependence or addiction? Or have you been doing completely different things from a behavioural point of view?
07:02 - 08:13
[SPEAKER_01]
It's mostly different things. The way that I found myself in tobacco harm reduction is by looking at the gateway hypothesis, which I have a session on later today. And I initially, being surrounded by sort of tobacco control academics to begin with during my postdoc, and reading the literature and seeing the news stories, I had never heard anything positive about e-cigarettes in 2014 when they started getting attention of the research community. And I completely bought the gateway hypothesis. So I thought along the way, you know, a junior faculty member is under a lot of pressure to publish. So I thought of a paper that I could use to prove the gateway hypothesis. And I thought, oh, great, I'm going to have a high impact paper. I'm going to, you know, look good on my next annual review by publishing in a good journal. And when I ran the analysis, I didn't find, the results were the exact opposite of what I had expected to find. So I started looking more at that. And at this point, I kind of left the dependent story behind because I was more interested in behavior and population level trends and what's happening with this gateway. Right.
08:13 - 08:22
[SPEAKER_00]
Okay, so how did you suddenly have a sort of epiphany, like, oh my God, I've got this wrong?
08:23 - 09:46
[SPEAKER_01]
Yeah, the first crack was that I had this longitudinal sample that I was working with to sample, to try to study that experimental phase of smoking and understand why some youth continue and some youth stop. It was a great longitudinal study, multi-year time points, and what I... My goal for this paper that ended up being like the first crack in my worldview was that I heard about the gateway, I believed in the gateway, but I was seeing assumptions being made about it that didn't have specific studies backing them up. So for example, the idea that the gateway happens through nicotine dependence. So if a kid starts using e-cigarettes, develops dependence, then that motivates them to start smoking. So it was about the mechanism through nicotine dependence that I wanted to prove. And what I found in this analysis was that prior nicotine dependence and prior smoking affected e-cigarette use, but not vice versa. Like e-cigarette use was not associated with later smoking. and it was not associated with later cigarette smoking. In other words, I couldn't prove causality, but what my results were showing was that e-cigarette was more likely to be an effect of prior smoking and the nicotine dependence from prior smoking than a cause of smoking in the future.
09:46 - 10:12
[SPEAKER_00]
right okay now i came across something when i was looking i had no idea what it is but according to my mates at chat gbt this was one of your signal contributions to tobacco harm reduction and i can't even remember exactly where something like a simulation model something or other yes i can't even remember what the phrase is now um but it seemed quite significant
10:13 - 11:57
[SPEAKER_01]
Yeah, so that was the next stage. The study I just described was kind of the first crack in my worldview. And for any given study, start second guessing like well maybe this isn't a real test of gateway because of this so i started thinking of how else can i test this gateway versus common liability idea and i thought of population level trends so if you think about it what the gateway would predict logically is that as e-cigarette use starts increasing in the youth population it should cause more smoking than there would would have been otherwise so but the comparison to the alternate world without e-cigarettes, you can't directly observe that. So you have to have some kind of some way of modeling what that hypothetical world would have looked like. So I used a simulation model where I trained it on, like I was able to replicate youth smoking trends from 2000 to 2010 before e-cigarettes became popular. And I projected that forward under three assumptions. One, no change. So just kind of carrying forward the prior decline in smoking. Two, the gateway. So for every e-cigarette user, there's a certain number of a certain number of them also become cigarette smokers, and then a diversion scenario. And the difference is remarkable. Like, the Gateway vastly overpredicts the number of actual youth smokers that we see in the population. Even the... kind of no effect, the no difference, vastly overpredicted. So it takes a diversion and a pretty strong diversion to be consistent with the real data.
11:58 - 12:00
[SPEAKER_00]
Right. When you say a diversion, what?
12:01 - 12:07
[SPEAKER_01]
Yeah, sorry. So what I mean by that is e-cigarette use might be preventing them from ever starting cigarette smoking.
12:08 - 13:18
[SPEAKER_00]
Oh, okay. All right, okay. Gateway I find fascinating in this context because the original gateway theory happened in the 1950s around drugs. And the idea perpetuated by the, an early version of the American DEA was that, I mean, the original idea was cannabis drives you mad, right? Reefer madness and all of that. And then they realised it didn't. So they had to come up with something else to keep the congressional funding going. So they said, okay, all right, well, it doesn't drive you mad, but once you start smoking on cannabis, you need something stronger. You need something more powerful. And so therefore you'll gravitate onto heroin in that case. in relationship to to what we're talking about it almost suggests that vaping by definition is safer because if you start with vaping and then you move on to smoking on the gateway theory it's an escalation so vaping is by definition because like you're right the gateway
13:18 - 13:30
[SPEAKER_01]
when people express their gateway concerns, like if their main example of what's bad about vaping is that it leads to smoking, that implicitly recognises that smoking is a lot more harmful.
13:31 - 16:45
[SPEAKER_00]
Indeed, indeed. Just getting back to this idea of addiction, we had a conversation yesterday about the the Latin root of the word. And the Latin root of addiction, I'm going to pronounce this right, is adictere. I got it wrong, didn't I? I wrote it down. Adicere. Adicere in Latin, which means enslaved. And the original idea was that if you owed somebody a debt, money, for example, in Roman society you'd have to go and work for them as a servant or a slave in order to work off the debt and over time that became the idea that you know if you were addicted to something you become enslaved to whatever and by the time you get to the 19th century then you're talking about opium morphine and all the rest of it And in the sense of tobacco harm reduction, as far as tobacco control is concerned, you're a failed quitter. It's what you basically are. But I suppose, and we talked about the stigmatising effect of a word like addiction. The trouble is that in the public narrative, Talk about addiction. People talk about, think about, immediately think about the worst consequences of the most serious drug use that they see. They've watched the French Connection or movies or Narcos or whatever it is, you know, on Netflix. And that's the impression you get of addiction. And then you get to that, to addict, which means, you know, you're a bad, not only a bad drug, but a bad person. But I think in relationship to nicotine is interesting because, you know, it was long established that, you know, cigarettes were bad news. Smoking was not a good thing. And I think, I don't know, get a sense that the narrative was more to do with what smoking causes cancer. really kind of get their head around the idea, well, it's like the nicotine that's making you want to smoke to get to the point where eventually in 30 years time you get cancer. But once you take the nicotine out of the cigarette, which is obviously what's happened, nicotine of itself becomes the new demon drug. And that's what people are talking about these days. I mean, is that the sort of thing... from a behavioural point of view, does that impact on the work that you do or the people that you talk to just in terms of, well, you know, all the focus on nicotine causes teenage brain damage, you know, whereas we've had cigarette, you know, you've been doing and other people have been doing analysis of teen smoking for decades and nobody's come up with the idea that nicotine causes brain damage. All of a sudden, nicotine has been kind of isolated. It's like, ah.
16:46 - 16:46
[SPEAKER_01]
Right.
16:47 - 16:47
[SPEAKER_00]
Yes.
16:47 - 18:08
[SPEAKER_01]
Yeah. Yeah. So the, uh, there's a lot to unpack there, but essentially there's, yeah, yeah, no, you're right. Like if you take away, if you ask skeptics about e-cigarettes, like what, what are you actually, what do you think the harms are? Um, they'll, they've sometimes they'll say long, you know, unknown long-term health effects, but like, as the evidence emerges, e-cigarettes have been out for almost 20 years now. Um, And there's very good safety profiles. They might be riskier than non-use, but they're vastly safer than conventional cigarettes. So if you take away the specific health harms, a lot of the objections now are backing up to addiction itself. But that's kind of, it's like tautological or maybe an empty definition of addiction, because if you take away the harm, what's left is just physiological dependence, but that applies to caffeine. alcohol too I think is maybe more problematic than nicotine and even at tobacco control conferences they often serve alcohol like free open bar you know at the end of some of the events and of course coffee is flowing freely throughout the day I think it looks over time to me as if the concept of addiction is as much a social and political construct
18:09 - 18:55
[SPEAKER_00]
as it is, because we haven't been, like you say, there's no proper medical definition of addiction. And I have noticed in some of the work that I've done that WHO in the 1950s divided drugs into habit forming and addicting, and they put tobacco under habit forming. And then successive US Surgeon General reports have kind of ramped things up. So I think the 79 report talked about nicotine as being dependence producing. And by 1988, all of a sudden the whole report was around nicotine addiction.
18:55 - 20:02
[SPEAKER_01]
Addiction, right, yeah. And I think this ties into misperceptions too, because I have a recent paper out with my Penny colleagues showing that 94% of adults who exclusively smoke and don't plan to quit falsely think that e-cigarettes are just as bad or more harmful than smoking. And there's some other work showing that if you dive down and you say you ask people, OK, why do you think they're worse? A lot of them cite addiction. A lot of people think that switching to e-cigarettes is just trading one addiction for another. And they don't like the enslavement aspect that you mentioned. And that's, you know, if somebody doesn't want to be addicted to a substance, that's That's fine for them, but other people don't have a problem, you know, being dependent on a cup of coffee in the morning. But there's also like different dependence potential. And there are a number of studies coming out now that if you switch from cigarettes to e-cigarettes, you can reduce your overall level of dependence when you look at the right measures that are comparable across products.
20:03 - 20:28
[SPEAKER_00]
So from a sort of behavioural point of view, somebody is a smoker, but they don't want to switch to e-cigarettes because of the addiction issue. I mean, they're not saying that I'm not addicted to smoking, but I might become addicted to e-cigarettes. They're not saying that, are they? Or are they?
20:28 - 20:44
[SPEAKER_01]
No, no. I mean, I think probably there's a good chance those people think that e-cigarettes are just as bad as well. So the health issues tie into that too. But the addiction itself bothers some people.
20:44 - 21:02
[SPEAKER_00]
And maybe that's because of the... They haven't bought into the idea that... the whole point about an e-cigarette is that you're not setting fire to anything. There's no combustion there. And so they have all the smoke nasties. You're not going to be in hate. That hasn't registered.
21:02 - 21:14
[SPEAKER_01]
I don't know if they understand that because you still see academic papers and news articles referring to e-cigarette smoking. That kind of language conflates and perpetuates the misperceptions.
21:15 - 21:33
[SPEAKER_00]
I mean, it always strikes me that the addiction argument is kind of what I call the get out of jail free card for people who oppose tobacco harm reduction. I mean, originally it was going to be, well, you know, kids will start vaping and then that will lead them to smoking. No.
21:33 - 21:34
[SPEAKER_01]
Right.
21:34 - 21:36
[SPEAKER_00]
No. It'll normalize smoking.
21:36 - 21:39
[SPEAKER_01]
That's increasingly not back to data.
21:40 - 23:02
[SPEAKER_00]
No. They're more dangerous than, at least as, if not more so than dangerous than cigarettes. No. So every box you tick, you can dismiss. And then they'll turn around and go, ah, ah, but nicotine is addictive, isn't it? And then you get into that whole argument. So it does resonate quite strongly, even though it's essentially meaningless, isn't it? you know, there's a public perception of what addiction means. People understand it in certain ways and they apply it in all sorts of different ways. They go, well, you know, I'm addicted to shopping or I'm addicted to looking. Actually, it's quite interesting now because you've now got a whole load of, I mean, big tobacco used to get it in the neck for the whole addiction thing and harms. And now it's big tech. Big tech are getting it in the neck about young people and addiction and how they're spending all their time looking at TikTok and Snapchat and all the rest of it. So addiction has kind of taken on a new dimension, really, in the public narrative.
23:02 - 23:55
[SPEAKER_01]
Yeah, I think the definitions have really evolved. I remember when I took my first psychology course and we discussed the difference between addiction and physiological dependence. And the reason, I was trying to understand in this class why caffeine wasn't considered an addiction. And the answer was that at the time that it's only a physiological dependence. It doesn't have the psychological component of addiction. And so if you apply that, I don't know if that's still the thinking in the field, but I think that applies to a non-combustible product use as well because it's, There might be physiological dependence symptoms, but you're not spending all your time pursuing this in the same way that an opioid-dependent person would.
23:55 - 24:30
[SPEAKER_00]
Which is interesting because the whole problem with opioid painkillers in the States, you know, Oxycontin and all of that, a lot of that was derived from the fact that somebody wrote a letter to the New English Journal of Medicine saying, well, you know, I've got cancer patients in hospital and we give them morphine for pain. And they go home and that's it. They don't feel too good for a while. They have some withdrawal symptoms, but they don't become addicts.
24:30 - 24:31
[SPEAKER_01]
Oh, right.
24:31 - 25:23
[SPEAKER_00]
They don't rush out on the streets looking for more. So they were dependent on morphine because they were given it for pain, but they didn't actually become addicted. So it sounds like in some disciplines... there is a differentiation that's made between the two. But in the world we operate in, as we say, if you've got a habit and you've got harm, you've got a dependency, shall we say, or some people would say an addiction. But if you take the harm out of it, and nicotine is, most people at this conference would agree that nicotine is relatively harmless, what's your problem?
25:23 - 25:24
[SPEAKER_01]
Mm-hmm. Right.
25:25 - 25:39
[SPEAKER_00]
I'm talking to the person in my ear. So I think I probably need to wrap this up now. It's been lovely chatting to you. Yeah, likewise. Thank you so much. Thank you for having me on. You're going to be busy for the rest.
25:39 - 25:42
[SPEAKER_01]
Yes, I've got two more. This is the first of three back-to-back sessions.
25:42 - 26:35
[SPEAKER_00]
Yeah, so this is your... Okay, this is your... What do we call it? Rehearsal. This is a rehearsal for what you've got left to do. That's great. Okay, everybody, thank you very much for tuning in to this edition. Tomorrow morning at the same time, I shall be talking to David Swenor. And we're going to be talking about prohibition, which is a theme of this conference, and more specifically, kind of what lies beneath. Is the title of tobacco harm reduction unfiltered tomorrow morning? Is there a moral underpinning? Well, there is a moral underpinning to prohibition, but we'll have a good discussion about that tomorrow morning. And in the meantime... Have a good day or evening, depending where you are. Thank you very much.