1. Introduction

A widely held assumption is that young people engage in smoking and other risk behaviors (e.g., alcohol or cannabis use) because their peers pressure them to do so. This assumption taps into one of the frequently applied theoretical models of peer influence, implying an active, explicit form of peer influence. As a result, most mass-media campaigns and school smoking-prevention programs focus on countering peer pressure by teaching young people refusal and resistance skills. Nevertheless, susceptibility to peer pressure in young people is not limited to adolescents but also includes young adults (see also review of Borsari and Carey, 2001). So far, the findings of survey studies, focusing on this active peer influence, show inconsistent findings (Perrine and Aloise-Young, 2004, Slater, 2003, Urberg et al., 1990) and experimental studies are lacking. Moreover, scholars question whether the outcomes of survey studies are valid and reliable (Arnett, 2007, Michell and West, 1996). Thus, we still know little about the effects of peer pressure on adolescent and young adult smoking. An important question that needs to be addressed is whether this assumption and theory of active peer influence is valid.

An alternative explanation for the influence of peers is found in the imitation hypothesis which taps into a different theoretical model of peer influence, implying a more passive, implicit form of peer influence. Adolescents and young adults observe and imitate the smoking of others, without being urged to do so. There are two explanations of imitation that have found support in the literature. One of the explanations is provided by the social cognitive/learning theory of Bandura, 1977, Bandura, 1986, which suggests that individuals observe and imitate (also called ‘modeling’) other's behavior that may intentionally lead to (immediate) positive rewards such as belonging to the group or being liked. Another explanation is provided by the perception-behavior link paradigm (Chartrand and Bargh, 1999); stressing the fact that individuals often imitate (also called ‘mimicry’) the behavior of others spontaneously and unintentionally. Moreover, empirical evidence has consistently shown that during interaction with another person, individuals unintentionally mimic his/her postures, mannerisms, facial expressions, eating behavior, and other behaviors (Chartrand and Bargh, 1999, Tanner et al., 2008). A small number of experimental studies, focusing on passive peer influence, have shown consistently that students are more likely to smoke in the company of a heavy-smoking than a non-smoking peer (Antonuccio and Lichtenstein, 1980, Harakeh et al., 2007, Kniskern et al., 1983, Miller et al., 1979). In the alcohol literature, experimental studies showed similar findings. Students modify their drinking rate in the direction of the drinking rate of the model (e.g., Collins and Marlatt, 1981; see also review of Quigley and Collins, 1999, Rosenbluth et al., 1978). The hypothesis of passive and active peer influence has not yet been put to the test in an experimental design, however. In this paper we report on an experimental study in which we focused on both passive (imitation) and active (pressure) peer influence to assess their relative impact on student smoking. Our hypothesis is that passive peer influence has a much stronger impact than active peer influence.

2. Method

2.1. Study design

The aim of this experiment is to examine whether passive (imitation) and/or active (pressure) peer influence affects young adults’ smoking. An experimental, observational study with a 2 (smoking condition) by 2 (peer pressure condition) factorial design was used. The smoking condition consisted of a confederate smoking zero cigarettes (non-smoking condition) versus three cigarettes (heavy smoking condition). The peer pressure condition consisted of a confederate not offering the participant cigarettes (no peer pressure condition) versus offering the participant verbally and non-verbally a cigarette three times by asking if s/he would like to smoke, along with opening the pack in front of him/her (peer pressure condition).

2.2. Procedure

The Ethics Committee of the Faculty of Social Sciences at Utrecht University gave their approval for this experiment. The principals of seven Dutch schools for intermediate technical and vocational training (in Nijmegen, Arnhem, Utrecht, Den Bosch) were informed about the actual aim of the experiment whereas this aim was masked for the students at these schools. The students were approached in the school to participate in a study on music taste and preference. We asked students to complete an initial screening questionnaire (Harakeh et al., 2010). Only daily smokers aged 16–25 years were invited to participate. Next, students participated once in a 60-min session during schooldays from 10:00 am to 17:30 pm in the period from May 2009 to January 2010. The participants were randomly assigned to the four conditions. Random assignment was blocked by gender and time of day in order to equally distribute males and females to each condition, and to equally distribute the time of the day when the participant participated over each condition. One-way ANOVA showed that there were no significant differences between the four conditions with regard to participants’ characteristics (age, gender, number of cigarettes smoked daily, and carbon monoxide level in their breath).

We created a mobile lab in a camper vehicle which we parked near the schools. One of the rooms was equipped as a relaxing room with a comfortable couch and a table, and the other room functioned as the observation room. In each session, a confederate and a participant participated in same-sex dyads sitting opposite each other. Participants were asked to blow into a device (Smokerlyzer) to measure the CO (carbon monoxide) level in their breath. To disguise the real aim of the device, students were told that the device enables us to assess alcohol consumption. Further, they were told they could eat food and take drinks that were made available, and that they were allowed to smoke in both rooms. Cigarettes were freely available in order to make the condition where the confederate offered cigarettes but smoked zero cigarettes credible. Confederates sat at a fixed place in the camper and, in each condition, the confederate noticed a pack of cigarettes next to him/her on the couch. The experimenter than asked them if they smoked (the confederate always answers positively) and explained that these cigarettes must have been forgotten by a previous participant and that they are allowed to use them. If the participant was in the smoking and/or pressure condition, the confederate directly smoked a cigarette from the pack, offered a cigarette, or both. The 30-min music task consisted of six music clips of pop songs. After each song, they filled in three questions individually in the questionnaire (grading the song) and discussed ten questions. The confederates were trained and instructed beforehand to always have a similar opinion on the songs as the participant, to act in a warm and friendly manner and to smoke cigarettes at a prearranged rate during the music task of 30 min. The confederates again smoked, offered a cigarette or both during the third and fifth song. At the end of the session, both filled in a brief questionnaire taking approximately 15 min. Each participant received eight Euros for their participation. After completion of this experiment, all participants were debriefed.

2.3. Participants

Of the 71 participants in the study sample, three participants were excluded: they were no longer daily smokers when they were participating in the session. Nine participants were only twice exposed to peer pressure, as they lit and smoked a cigarette just when the confederate was supposed to offer a cigarette (five participants in the pressure, no smoking condition; four participants in the pressure, smoking condition). Of these, five participants did not get offered a cigarette the first time, three participants the second time, and one participant the third time. These nine participants were included in the analyses on the basis of intention-to-treat. The 68 participants were assigned to one of the four conditions: (1) no pressure, no smoking condition (N = 15), (2) smoking, no pressure condition (N = 16), (3) pressure, no smoking condition (N = 20), and (4) pressure, smoking condition (N = 17). Participants were 16–24 years-old (mean age = 18.21, SD = 1.71), 38.2% were male. At the end of the session all participants answered the question in the questionnaire on what they thought the study was about. The responses showed that none of these participants suspected the actual aim of the experiment.

2.4. Measures

Participant's smoking behavior during the session. The experimenter coded the number of cigarettes smoked. We examined as primary outcome the total number of cigarettes.

CO level. The Micro+ Smokerlyzer is a breath monitor which assesses the CO (Harakeh et al., 2010; www.bedfontusa.com ). The participants were asked to blow into the monitor after holding their breath, and a digital readout of CO ppm (one part CO in one million parts of breath) is displayed on the monitor.

2.5. Statistical analyses

All analyses were conducted with Stata. We used Poisson loglinear analyses to investigate the main effects of the pressure and smoking condition on the total number of cigarettes smoked during the session, controlling for covariates (participant's CO level and gender). Subsequently, we tested the interaction effect of peer pressure × peer smoking.

3. Results

The majority (77.9%) of the participants lived at home. All participants were daily smokers: 22.4% smoked 1–5 cigarettes/day, 28.4% 6–10 cigarettes/day, 47.8% 11–20 cigarettes/day, and 1.5% 21–30 cigarettes/day. The participant's smoked at various locations: school (98.5%), at parties/pleasantly engaging evenings (98.5%), on the street (89.7%), at the homes of their friends (88.2%), at bars/discotheques (80.9%), at home – kitchen/living room (45.6%), at home – in their bedroom (36.8%), and in the sports canteen (13.2%). The participants all smoked during the music task: 22.1% smoked one cigarette, 36.8% smoked two cigarettes, and 41.2% smoked three cigarettes. The participants’ CO level ranged from 0 to 34 ppm (M = 9.14, SD = 5.65).

The findings depicted in Table 1 show that peer smoking affected significantly the total number of cigarettes smoked by the student. Students confronted with a smoking peer had a higher likelihood to smoke more cigarettes (p = 0.003). However, peer pressure did not significantly predict the total number of cigarettes smoked by the student (p = 0.309). The covariates (i.e., gender and CO-level) did not predict significantly the total number of cigarettes smoked by the participant. Furthermore, we tested in a next step the interaction effect of peer pressure × peer smoking. This interaction effect showed to be not significant (IRR = 0.70, 95% CI = 0.35–1.38, p = 0.301).Table 1. The effects of active (peer pressure) and passive (peer smoking) peer influence on participants’ total number of cigarettes.

Note: Poisson loglinear analyses, 95% CI = 95% confidence intervals.**p < 0.01.

4. Discussion

This study is the first to show the importance of passive (imitation) peer influence over and above the impact of active (pressure) peer influence on young adult smoking in an experimental design. In our study, peer smoking increased significantly young adults’ likelihood to smoke more cigarettes while peer pressure did not. In the literature, peer smoking is suggested to tap into the passive peer influence, and the underlying mechanism in experimental studies and survey studies on smoking is often contributed to imitation. Students confronted with smoking peers are more likely to smoke regardless of being offered a cigarette or not: seeing is doing. Several theoretical models may explain the underlying mechanisms leading to imitation of behavior of others. One of these theories that have frequently been examined in previous studies is social conformity (see also a meta-analysis of Bond and Smith, 1996). Solomon Asch's work showed that in a group setting participants conform to the norm of the group, i.e., they tended to conform to the behavior of the other group members (Asch, 1951). Thus, social conformity may explain our findings and imply that young adults imitating peer smoking have been intentional. However, in our study we tested peer dyads and not peer groups. There is evidence that conformity of people is more likely to occur in groups than in dyads, and thus this explanation may have played a minor role in our present study. Another possible explanation is that imitating the other in human interaction may reflect a basic instinct in human beings that might even be biological in origin, as has been shown by studies on the importance of imitation for social interaction and social development of animals (Hurley and Chater, 2005).

An alternative theory to explain our findings is the cue-reactivity paradigm. According to this paradigm, smokers react to smoking-related cues/stimuli (e.g., handling a lit cigarette, ashtrays, lighters, or smelling another person's cigarette) in their environment by an increase in craving to smoke (see also meta-analyses of Carter and Tiffany, 1999, Conklin et al., 2008). The smoking-related cues of ashtrays, lighters and package of cigarettes were present in all four conditions, although handling a lit cigarette and smelling another person's cigarette were only present in the condition were the confederate smoked. Thus, these latter two smoking-related cues may have elicited craving in the daily smoking young adults and triggered them to smoke. However, in our previous experimental study (Harakeh and Vollebergh, in press) we excluded in our research design the alternative hypothesis concerning smelling another person's cigarette smoke. These findings showed that when the participant interacted with a smoking peer through the internet and webcam (i.e., confederate and participant sat in two different rooms and participant could not smell the cigarette of the peer) they were also more likely to smoke more cigarettes. Thus, finding support for the imitation hypothesis.

Our findings seem to suggest that young adults behave in a particular way because their social environment passively evokes certain behaviors and less because they are actively or explicitly encouraged to behave in a specific way. Thus, our results may imply that passive peer influence may be of more importance to understand young adult smoking than active peer influence. Our findings must be carefully interpreted but seem to suggest that smoking cessation programs and policy should probably target and put more emphasis on passive peer influence (rather than active peer influence) in order to decrease smoking among daily smoking young adults. There may be three possible ways they could address this. First of all, most of the smoking cessation campaigns portray smoking models in their ads which in themselves may induce people to smoke and may therefore be counterproductive. Therefore, smoking models should perhaps no longer be depicted in these campaigns. Second, interaction with smoking models should be prevented. Government policy has been contributing to this goal by restricting smoking in public settings (e.g., trains, airplanes, bars, restaurants). However, smoking is, surprisingly, not yet officially banned in schoolyards worldwide; one of these countries that does not have such legislation in place is The Netherlands. We would recommend stricter school policies in this respect for these countries (Griesbach et al., 2002, Schnohr et al., 2008, Wold et al., 2004). Third, awareness should be increased of the urge to imitate others. Especially young adults trying to quit or reduce smoking need to be alerted to the effects of smoking by others in their presence, and to successfully quit or reduce smoking they should learn to avoid these situations. Smoking cessation campaigns could emphasize and support this message. Nevertheless, future studies are needed to replicate our study to find support for our findings and to gain more knowledge on these two kinds of peer influences.

4.1. Future research

There are several aspects that need to be taken into account in future research. First, we operationalized peer pressure as the verbal and nonverbal encouragement to take and smoke a cigarette but we did not take into account the possibility that in real life situations, this could be accompanied by teasing, taunting and rejection when the offered cigarette is declined. Although there is less evidence for the occurrence of this so-called coercive pressure (Arnett, 2007), future studies nevertheless need to examine its impact on student smoking. Second, more insights are needed on who are more likely to being imitated (e.g., popular peers), who are more likely to imitate (e.g., young adults with little self-esteem), and whether young adults imitate peer smoking intentional and/or unintentional. Thus, future studies also need to examine the characteristics of the confederate and the participant and test participant's awareness of imitation.

4.2. Limitations

The strengths of these two studies are: (1) the experimental design and (2) testing peer imitation and pressure in one design. There are also some shortcomings which should be taken into consideration. First, in our study unfamiliar peers were the confederates, but peer relations usually centre on familiar companions of a similar age, including (best) friends, siblings, etc. It would be interesting to test whether smoking by familiar peers (e.g., best friend, sibling) affects student smoking differently compared to smoking by strangers. This is difficult to examine in experimental studies; observational studies would be more appropriate. Second, our sample is restricted to smoking continuation among daily smokers. Thus, our findings may be helpful for smoking cessation programs but we need to replicate in future studies whether this also applies to preventing and discouraging smoking initiation and experimentation. Third, this experimental study is conducted in a camper van focusing on peer dyads. However, the impact of active and passive peer influence may vary in different environment and setting (e.g., work setting, school setting, or other public places) and may depend on the number of peers and smoking norms in that specific setting. Fourth, in this study design cigarettes were freely available in order to make the condition where the confederate offered cigarettes but smoked zero cigarettes credible. However, this may not have biased our findings because the cigarettes were freely available in all conditions but may explain why in this study all participants smoked at least one cigarette. Finally, we did not measure smoking topography in detail, but only looked at cigarette frequency. Previous studies showed that imitation did not affect puff frequency per cigarette, percentage of tobacco burned, puff duration, and average inter-puff interval, but only influenced the macro-measures of cigarette frequency and inter-cigarette interval (Antonuccio and Lichtenstein, 1980, Miller et al., 1979). We did not include the latter smoking outcome in this present study because the number of participants would decrease in this analysis, and therefore also the power to detect significant findings.

5. Conclusion

Young adults seem to continue to smoke due to passive peer influence rather than active peer influence. Young adults strongly imitate smoking in mere interaction with complete strangers regardless of being offered a cigarette or not. Anti-smoking policy could probably target this passive peer influence by removing smoking models from smoking cessation campaigns, by banning smoking in schoolyards, and by increasing awareness of imitating the smoking of others.

Link to Article

1. Introduction

The influence of peers on adolescent and young adult smoking behaviors is a complex issue that has garnered significant attention. A prevalent assumption is that young individuals engage in smoking and other risk behaviors due to explicit peer pressure, leading to a focus on resistance skills in prevention programs. However, empirical support for this active peer influence model remains inconsistent, with survey studies yielding mixed results and a lack of experimental investigations.

An alternative theoretical perspective posits that peer influence operates more passively through imitation. This perspective suggests that adolescents and young adults adopt smoking behaviors by observing and mimicking their peers, even without direct encouragement. This imitation hypothesis finds support in social cognitive/learning theory and the perception-behavior link paradigm, both of which emphasize the role of observation and modeling in behavior acquisition. Empirical evidence consistently demonstrates that individuals unintentionally mirror the behaviors of others, including smoking, in social settings.

Despite the existing evidence supporting the imitation hypothesis, a direct experimental comparison of passive and active peer influence on smoking behavior is lacking. This paper addresses this gap by reporting on an experimental study designed to assess the relative impact of imitation (passive influence) and peer pressure (active influence) on student smoking. Our hypothesis posits that passive peer influence exerts a more significant effect on smoking behavior compared to active peer influence.

2. Method

2.1. Study Design

This experiment employed a 2 (smoking condition) by 2 (peer pressure condition) factorial design to examine the effects of passive and active peer influence on young adult smoking. The smoking condition involved a confederate either refraining from smoking (non-smoking condition) or smoking three cigarettes (heavy smoking condition). The peer pressure condition involved a confederate either not offering cigarettes to the participant (no peer pressure condition) or actively offering cigarettes three times verbally and nonverbally (peer pressure condition).

2.2. Procedure

Following approval from the Ethics Committee of the Faculty of Social Sciences at Utrecht University, seven Dutch schools for intermediate technical and vocational training participated in the study. Students aged 16-25 who identified as daily smokers were recruited through a screening questionnaire under the guise of a music taste and preference study. Participants (N=68) were randomly assigned to one of the four conditions, ensuring an equal distribution of gender and time of day across groups.

The study was conducted in a mobile lab equipped with a relaxing room and an observation room. Participants engaged in a 30-minute music task in same-sex dyads with a trained confederate. Participants' carbon monoxide (CO) levels were measured before the task using a Smokerlyzer, presented as a device for assessing alcohol consumption. Cigarettes were freely available in both rooms. Confederates in the smoking and/or pressure conditions were instructed to smoke at a predetermined rate and offer cigarettes during specific intervals. Following the music task, participants completed a questionnaire.

2.3. Participants

Seventy-one students initially participated; however, three were excluded due to no longer being daily smokers, and nine were included on an intention-to-treat basis as they smoked a cigarette precisely when the confederate intended to offer one. The final sample consisted of 68 participants (age range: 16-24 years, mean age = 18.21, SD = 1.71; 38.2% male). Post-experiment debriefing confirmed that none of the participants suspected the true aim of the study.

2.4. Measures

• Participant's smoking behavior during the session: The primary outcome measure was the total number of cigarettes smoked by the participant during the 30-minute session, coded by the experimenter.

• CO level: Participants' CO levels, measured in parts per million (ppm), were assessed before the session using the Micro+ Smokerlyzer.

2.5. Statistical Analyses

Data were analyzed using Stata. Poisson loglinear analyses were employed to examine the main effects of the pressure and smoking conditions on the total number of cigarettes smoked, controlling for participant gender and CO level. An interaction effect of peer pressure × peer smoking was also tested.

3. Results

The majority of participants (77.9%) resided at home, and all were daily smokers with varying consumption levels. Participants reported smoking in various locations, with 98.5% indicating smoking at school and during social events. All participants smoked during the music task, with the number of cigarettes smoked ranging from one to three. CO levels ranged from 0 to 34 ppm (M = 9.14, SD = 5.65).

As shown in Table 1, peer smoking significantly influenced the total number of cigarettes smoked (p = 0.003). Students exposed to a smoking confederate were more likely to smoke more cigarettes themselves. Conversely, peer pressure did not significantly predict the total number of cigarettes smoked (p = 0.309). Neither gender nor CO level significantly predicted smoking behavior. The interaction effect of peer pressure × peer smoking was also non-significant (IRR = 0.70, 95% CI = 0.35–1.38, p = 0.301).

Table 1. The Effects of Active (Peer Pressure) and Passive (Peer Smoking) Peer Influence on Participants’ Total Number of Cigarettes

Note. Poisson loglinear analyses, 95% CI = 95% confidence intervals. ***p < 0.01.

4. Discussion

This study provides novel experimental evidence highlighting the stronger influence of passive peer influence (imitation) compared to active peer influence (pressure) on young adult smoking behavior. Our findings demonstrate that observing a peer smoking significantly increases the likelihood of smoking among young adults, irrespective of whether direct pressure to smoke is present.

Several theoretical frameworks can be considered to explain the observed influence of imitation. Social conformity, as demonstrated in Asch's conformity experiments, suggests that individuals align their behavior with group norms. However, our study's focus on dyadic interactions rather than group settings may limit the applicability of this explanation. Alternatively, imitation in human interaction could be considered a fundamental, potentially innate, aspect of social behavior, as evidenced by its role in animal social development and interaction.

The cue-reactivity paradigm offers another potential explanation, proposing that exposure to smoking-related cues triggers cravings in smokers. While our study design controlled for some cues (e.g., ashtrays, lighters), the presence of a smoking confederate introduced additional cues (e.g., handling a lit cigarette, smell of smoke) that might have elicited craving and subsequent smoking. However, findings from a previous study by our group (Harakeh & Vollebergh, in press) using an online interaction paradigm, where such cues were absent, also supported the imitation hypothesis, suggesting that additional factors beyond cue reactivity are at play.

Our findings suggest that young adults' smoking behaviors are more strongly influenced by passive observation of their social environment than by explicit social pressure. This highlights the potential importance of targeting passive peer influence in smoking cessation programs and policies. Specifically, future interventions could consider: (1) avoiding the use of smoking models in anti-smoking campaigns, as this may inadvertently promote smoking behavior; (2) implementing stricter policies to prohibit smoking in schoolyards and other environments frequented by young adults; and (3) raising awareness among young adults attempting to quit smoking about the influential role of observing others smoke and encouraging them to avoid such situations.

4.1. Future Research

Several avenues for future research emerge from this study. First, exploring the impact of different forms of peer pressure, including coercive pressure involving teasing or rejection, on smoking behavior is crucial. Second, investigating individual characteristics of both the imitator (e.g., self-esteem) and the model (e.g., popularity) could provide insights into the factors that moderate the imitation of smoking behavior. Additionally, examining young adults' awareness of their imitative behaviors and the intentionality behind these actions would further enhance our understanding of this phenomenon.

4.2. Limitations

This study has several limitations that warrant consideration. First, the use of unfamiliar confederates as peers might not accurately reflect the dynamics of real-life peer relationships, which often involve closer bonds with friends and family members. Future studies could explore the influence of smoking behavior among familiar peers. Second, our focus on smoking continuation among daily smokers limits the generalizability of findings to smoking initiation and experimentation among non-smokers. Third, the specific experimental setting (camper van) and dyadic interaction might not be representative of other social contexts where peer influence occurs. Fourth, the unrestricted access to cigarettes, while necessary for methodological reasons, may have influenced smoking behavior in all conditions. Finally, our study did not assess detailed smoking topography, focusing solely on cigarette frequency. Future research could incorporate measures of puff frequency, duration, and other topographical aspects to provide a more comprehensive understanding of the imitation of smoking behavior.

5. Conclusion

Our findings underscore the significant role of passive peer influence, through imitation, in maintaining smoking behavior among young adults. Young adults demonstrate a strong tendency to imitate smoking behaviors observed in their social environment, even in the absence of direct pressure. These findings emphasize the need for smoking cessation programs and policies to address the potent influence of passive peer influence by reducing exposure to smoking cues, promoting smoke-free environments, and raising awareness about the automatic nature of imitation.

Link to Article

1. Introduction

Many people think young people smoke and engage in other risky behaviors (like drinking alcohol or using cannabis) because of peer pressure. This idea of direct, obvious pressure from peers is a common theory. It's why many media campaigns and school programs focus on teaching young people how to say no to cigarettes. While we often associate peer pressure with teenagers, it also impacts young adults. However, research on this "active" peer influence shows mixed results, and we lack experimental studies. Some researchers even doubt the accuracy of survey studies on this topic. Therefore, we still don't fully understand how peer pressure affects smoking in adolescents and young adults, and we need to examine if this theory of active peer influence holds true.

Another explanation is the "imitation hypothesis." This theory suggests a more passive, subtle form of peer influence where young people copy the smoking behavior of others without any direct pressure. Two explanations for imitation have support. Firstly, Bandura's social cognitive/learning theory (1977, 1986) suggests people copy behaviors they see (called "modeling") if they think it will lead to rewards, such as fitting in or being liked. Secondly, the perception-behavior link paradigm emphasizes that people often unconsciously and automatically imitate others' behavior ("mimicry"). Research confirms that people unintentionally mirror others' postures, gestures, facial expressions, eating habits, and other behaviors during interactions. A few experimental studies on this passive influence have consistently shown that students are more likely to smoke when with a heavy smoker compared to a non-smoker. Similar findings exist in alcohol research, where students adjust their drinking to match the person they're with. However, no experimental study has directly compared passive and active peer influence on smoking. Our study investigates both types of influence to see which has a greater impact on student smoking, hypothesizing that passive influence is more powerful.

2. Method

2.1. Study design

This experiment investigates whether passive (imitation) or active (pressure) peer influence affects young adult smoking. We used a 2 (smoking condition) by 2 (peer pressure condition) factorial design. The smoking condition involved a confederate either smoking zero cigarettes (non-smoking) or three cigarettes (heavy smoking). The peer pressure condition involved the confederate either not offering cigarettes to the participant (no pressure) or offering a cigarette three times, both verbally and by gesturing with the pack (pressure).

2.2. Procedure

This experiment was approved by the Ethics Committee of the Faculty of Social Sciences at Utrecht University. Seven Dutch vocational schools were informed about the study's true purpose, but students were told it was about music taste. Daily smokers aged 16–25 years completed a screening questionnaire and were invited to a 60-minute session. Participants were randomly assigned to one of the four conditions, ensuring equal distribution of gender and time of day. There were no significant differences in age, gender, cigarettes smoked daily, or carbon monoxide levels between the four groups.

A mobile lab in a camper van was parked near the schools. One room was set up as a comfortable lounge, while the other was used for observation. Each session involved same-sex pairs (one participant and one confederate). Participants provided a breath sample using a Smokerlyzer to measure carbon monoxide levels, but were told it measured alcohol consumption. They were offered food, drinks, and were allowed to smoke in both rooms. Cigarettes were readily available to ensure the "offering but not smoking" condition was believable. Confederates sat in a designated spot and "found" a pack of cigarettes next to them. The experimenter asked if they smoked (they always said yes) and told them they could smoke those cigarettes. Depending on the condition, the confederate either immediately lit a cigarette, offered one to the participant, or both. During the 30-minute music task, participants listened to six pop songs, rated each song individually, and then discussed ten questions about the songs with the confederate. Confederates were trained to have similar music opinions as the participants, act friendly, and smoke at a pre-determined rate. They offered cigarettes again during the third and fifth songs. Finally, everyone completed a short questionnaire (around 15 minutes) and received eight Euros. All participants were debriefed about the true nature of the study.

2.3. Participants

Out of 71 participants, three were excluded because they no longer smoked daily. Nine participants were only exposed to peer pressure twice because they lit a cigarette right when the confederate was supposed to offer them one. These nine participants were included in the analysis based on their initial assigned group. The final sample consisted of 68 participants: 15 in the no pressure/no smoking group, 16 in the smoking/no pressure group, 20 in the pressure/no smoking group, and 17 in the pressure/smoking group. Participants were aged 16–24 years old (average age 18.21 years), and 38.2% were male. None of the participants guessed the study's true objective based on post-experiment questioning.

2.4. Measures

Participant's smoking during the session. The experimenter recorded the number of cigarettes smoked, with the total number as the primary outcome measure.

CO level. The Micro+ Smokerlyzer is a breath monitor that measures carbon monoxide levels (Harakeh et al., 2010). Participants blew into the device after holding their breath, providing a reading in parts per million (ppm).

2.5. Statistical analyses

Stata was used for all analyses. Poisson loglinear analyses were conducted to see if the pressure and smoking conditions predicted the total number of cigarettes smoked, while controlling for participant gender and CO level. We also tested the interaction between peer pressure and peer smoking.

3. Results

Most participants (77.9%) lived at home. Smoking habits varied: 22.4% smoked 1–5 cigarettes daily, 28.4% smoked 6–10, 47.8% smoked 11–20, and 1.5% smoked 21–30. Participants reported smoking in various locations: school (98.5%), social events (98.5%), outdoors (89.7%), at friends' homes (88.2%), bars/clubs (80.9%), at home in common areas (45.6%), their bedrooms (36.8%), and sports facilities (13.2%). All participants smoked during the music task: 22.1% smoked one cigarette, 36.8% two cigarettes, and 41.2% three cigarettes. CO levels ranged from 0 to 34 ppm (average = 9.14 ppm).

Table 1 shows that the confederate's smoking behavior significantly impacted how many cigarettes the participant smoked. Participants with a smoking partner were more likely to smoke more cigarettes themselves (p = 0.003). However, peer pressure alone did not significantly predict smoking (p = 0.309). Neither gender nor CO level significantly predicted smoking. Additionally, the interaction between peer pressure and peer smoking was not significant (IRR = 0.70, 95% CI = 0.35–1.38, p = 0.301).

Table 1. The effects of active (peer pressure) and passive (peer smoking) peer influence on participants’ total number of cigarettes.

Note: Poisson loglinear analyses. 95% CI = 95% confidence intervals. **p < 0.01.

4. Discussion

This is the first study to experimentally demonstrate the stronger impact of passive (imitation) peer influence on young adult smoking compared to active (pressure) peer influence. While peer pressure did not affect smoking, simply being around a smoking peer did. This supports the idea that "seeing is doing." Several theories might explain the mechanisms behind this imitation. Social conformity, as demonstrated in Asch's (1951) work, suggests that people conform to group norms, even if it means going against their own judgment. While this could explain our findings, it's important to note that our study used pairs, not groups, and conformity is more likely in larger groups. Therefore, this explanation might be less relevant in our context. Another possibility is that imitating others is an innate human behavior, potentially with biological roots. Studies on animals have shown the importance of imitation for social interaction and development, suggesting this could be a fundamental aspect of social behavior.

The cue-reactivity paradigm offers another explanation. This theory suggests that smokers react to smoking-related cues in their environment (e.g., the sight or smell of cigarettes, lighters, ashtrays) by experiencing increased cravings. While some smoking cues were present in all conditions, the sight and smell of a lit cigarette were only present when the confederate smoked. These specific cues might have triggered cravings in the participants, leading them to smoke more. However, a previous study by our team found similar results even when participants interacted with a smoking peer through a webcam, eliminating the smell factor. This supports the imitation hypothesis, as participants still smoked more even without being exposed to the smell of smoke.

Our findings suggest that young adults' smoking behavior might be more influenced by the passive observation of others than by direct pressure. This has important implications for smoking cessation programs and policies, which might be more effective if they focus on passive influence. There are three potential approaches: Firstly, anti-smoking campaigns often use smoking models, which might be counterproductive as they could trigger cravings in viewers. Removing these models from campaigns might be beneficial. Secondly, limiting exposure to smoking models is crucial. While policies restricting smoking in public places are helpful, some countries, including the Netherlands, still allow smoking on school grounds. Stricter policies in such settings could be beneficial. Thirdly, raising awareness about the power of imitation is key. Individuals trying to quit or reduce smoking should be made aware of how observing others' smoking affects them, and they should be encouraged to avoid such situations. Smoking cessation campaigns can play a vital role in highlighting this aspect. However, further research is needed to confirm our findings and deepen our understanding of these two types of peer influence.

4.1. Future research

Future research should consider several factors. Firstly, our study defined peer pressure as verbal and nonverbal encouragement to smoke but did not include other forms of pressure, such as teasing, mocking, or exclusion for refusing a cigarette. While this more aggressive pressure, known as coercive pressure, might be less prevalent, its impact on student smoking requires further investigation. Secondly, we need to understand who is more susceptible to being imitated (e.g., popular peers) and who is more prone to imitating (e.g., individuals with low self-esteem), and whether imitation of smoking is conscious or unconscious. Future studies should explore the characteristics of both the influencer and the influenced and investigate if participants are aware of their imitative behavior.

4.2. Limitations

Our study benefits from its experimental design and the simultaneous examination of peer imitation and pressure. However, some limitations should be acknowledged. Firstly, our confederates were unfamiliar to the participants. Real-life peer interactions usually involve familiar individuals like close friends or siblings. Future studies could investigate if smoking by familiar peers has a different impact than smoking by strangers, although this would be better suited for observational studies. Secondly, our study only focused on smoking continuation in existing smokers. While our findings are relevant for cessation programs, future research should investigate if these findings also apply to preventing smoking initiation and experimentation. Thirdly, our experiment was conducted in a camper van with peer dyads. The impact of active and passive peer influence might vary in different environments (e.g., school, work, public places) and could be influenced by the number of peers present and the existing smoking norms in those settings. Fourthly, to maintain credibility, cigarettes were freely available in all conditions, which may have contributed to all participants smoking at least one cigarette. Lastly, we focused on cigarette frequency and did not analyze detailed smoking topography. While past studies showed that imitation mainly impacted cigarette frequency and not puffing behavior, future research could explore these finer details of smoking behavior.

5. Conclusion

This study suggests that passive peer influence, rather than active peer pressure, might be a more significant factor in young adults' continued smoking. Young adults tend to mimic smoking behavior simply by observing it, even when there's no direct pressure to smoke. Anti-smoking initiatives could address this passive influence by: avoiding smoking models in their campaigns; advocating for smoke-free school environments; and raising awareness about the powerful, and often unconscious, influence of observing others smoke.

Link to Article

1. Introduction

You've probably heard people say that teens start smoking because their friends pressure them into it. This idea is part of a common theory about peer influence, which says that friends directly and obviously try to make each other do things. Because of this, lots of ads and school programs try to teach young people how to say no to cigarettes. But adults can be influenced by their friends too! And the research on how much pressure actually makes people smoke is pretty mixed up. Some experts even wonder if surveys can give us the real story. So, we don't really know for sure how much peer pressure makes teens and young adults light up.

Another explanation for why people smoke is that they're just copying what they see, a more subtle type of peer influence. This is called the imitation hypothesis. Teens and young adults see others smoking, and even without being pressured, they do the same. There are two main ideas about why imitation works. One is social cognitive/learning theory. This theory says we learn by watching others and we're more likely to copy behaviors that seem to have good results, like being accepted or liked. Another idea is the perception-behavior link paradigm, which says we unconsciously imitate others (like their movements, expressions, or even eating habits) all the time without realizing it. A few studies have actually shown that students smoke more around heavy smokers, even without being told to. Similar things happen with alcohol. But no one's tested both direct pressure and imitation in one experiment. That's what this study is for! We think seeing someone smoke will have a bigger effect than being pressured to.

2. Method

2.1. Study Design

We wanted to see if young adults smoke more because of imitation, pressure, or both. We created an experiment with four groups:

• Group 1: A researcher acted like a regular participant ("confederate") and didn't smoke, and they didn't offer the real participant a cigarette.

• Group 2: The confederate smoked three cigarettes, but didn't offer any to the participant.

• Group 3: The confederate didn't smoke, but offered the participant a cigarette three times (verbally and by showing the pack).

• Group 4: The confederate smoked three cigarettes and offered the participant a cigarette three times.

2.2. Procedure

The study was approved by the Ethics Committee at Utrecht University. Seven schools in the Netherlands agreed to help, but the students didn't know what the study was really about. We told them it was about music taste. Only daily smokers aged 16-25 were included.

The study took place in a camper van set up like a lounge. A participant and a confederate (the same gender) sat across from each other. They were told to breathe into a device that measures carbon monoxide (CO) levels (we said it was for alcohol). There were snacks, drinks, and cigarettes available. The confederates acted friendly and pretended to like the same music as the participant.

They listened to six pop songs and answered questions about them. In the smoking groups, the confederate lit up during specific songs. In the pressure groups, they offered a cigarette during those same songs. At the end, everyone took a short survey and got paid eight Euros. Afterward, we told them what the study was really about.

2.3. Participants

Out of 71 people, 3 weren't daily smokers anymore and were excluded. Nine people lit up right before the confederate was supposed to offer them a cigarette, so they only got pressured twice. But we kept their data anyway. That left 68 people: 15 in group 1, 16 in group 2, 20 in group 3, and 17 in group 4. They were 16-24 years old (average age 18), and 38% were male. No one guessed what the study was actually about.

2.4. Measures

• Smoking: We counted how many cigarettes the participant smoked.

• CO level: We measured the CO level in their breath at the beginning.

2.5 Statistical Analyses

We used special stats (Poisson loglinear analyses) to see if being in the pressure group or the smoking group made people smoke more, even after accounting for their starting CO level and gender. We also looked for an interaction effect, meaning we checked if being in BOTH the pressure AND smoking group changed things even more.

3. Results

Most people (77.9%) lived at home and smoked in a lot of places: school (98.5%), parties (98.5%), outside (89.7%), friends' houses (88.2%), bars/clubs (80.9%), and even at home (kitchen/living room: 45.6%, bedroom: 36.8%). Everyone smoked during the music part: 22.1% had one, 36.8% had two, and 41.2% had three. Their CO levels ranged from 0 to 34 ppm (average = 9.14).

Here's what we found (see Table 1):

• Peer smoking mattered: People smoked more cigarettes when the confederate smoked too (p = 0.003).

• Pressure didn't matter: Being offered a cigarette didn't make people smoke more (p = 0.309).

• No interaction: Being in both the pressure AND smoking group wasn't any different than just being in the smoking group (p = 0.301).

Note: This table summarizes the main findings.

4. Discussion

This is the first study to clearly show that young adults are more influenced by seeing smoking than by being told to smoke. Seeing is believing, right?

So, how does simply watching someone smoke make other people light up?

• Social Conformity: We know people act like those around them, especially in groups. But our study used pairs, not groups, so this might not be the main reason.

• Instinct: Copying others might be something we do automatically, even without thinking. Animals do it too, which makes scientists think it might be hardwired (Hurley & Chater, 2005).

• Cues: Seeing smoking-related things (like cigarettes, lighters, or even smelling smoke) can make smokers crave a cigarette. In our study, everyone saw cigarettes, but only some smelled the smoke. However, another study found the same results even when people couldn't smell the smoke. So, it seems like imitation is the real reason.

Our study suggests that to stop smoking, we need to focus on the subtle ways it spreads, not just direct pressure. Here's what we can do:

• No More Smoking in Ads: Seeing smoking in ads might actually make people want to smoke!

• Ban Smoking at School: Surprisingly, some schools still allow smoking. We need to make schools totally smoke-free.

• Be Aware of Imitation: If someone is trying to quit, they should be careful about who they hang out with. Seeing others smoke might make it harder.

4.1 Future Research

There's still a lot to learn about imitation. For example:

• Stronger Pressure: We only looked at offering cigarettes. What if people made fun of you for not smoking? That might be more effective, but it's also not very nice.

• Who Copies Whom?: We need to figure out why some people copy more than others and if certain people are more likely to be copied.

4.2. Limitations

Our study was good because it was an experiment, but it wasn't perfect:

• Strangers: We used researchers as the smokers, but usually, friends influence each other more than strangers do.

• Only Daily Smokers: We don't know if the same is true for people who are just starting to smoke.

• The Setting: We only looked at what happened in a camper van. Things might be different in other places.

• Free Cigarettes: We gave everyone free cigarettes, which isn't realistic.

5. Conclusion

This study shows that just seeing someone smoke can make other people smoke more, even without any pressure. To fight smoking, we need to pay attention to how we portray smoking in the media, make schools smoke-free, and teach people that it's easy to pick up bad habits without even realizing it.

Link to Article

Do Friends Make Teens Smoke? Peer Pressure or Imitation?

Lots of people think kids start smoking because their friends tell them to. They think friends pressure each other to try it. Because of this, grownups try to teach kids how to say "no" to smoking. But what if just seeing someone smoke makes a person want to smoke too?

Some scientists believe that instead of being pressured, we copy what we see. This is called imitation. We see our friends smoking, and we copy them without even thinking about it! We already know that people copy each other's movements and actions without realizing it.

A group of scientists wanted to see which was more important - peer pressure or imitation. They asked teenagers who already smoked to hang out with an actor. Sometimes the actor smoked, and sometimes they didn't. They also sometimes asked the teenager to smoke, and sometimes they didn't. The scientists were surprised by what they found!

The Experiment

How They Did It

The scientists used a camper van as their lab. They had teenagers come inside and sit with the actor. The teenagers thought they were doing a study about music. The scientists secretly watched how many cigarettes the teenagers smoked when the actor was smoking and when they weren't. They also watched to see if the teenagers smoked more when the actor offered them a cigarette.

Who Was in the Experiment

The scientists asked 68 teenagers to be in the experiment. The teenagers were between 16 and 24 years old, and they already smoked every day.

What They Measured

The scientists wanted to know how many cigarettes the teenager smoked during the experiment. They also measured the amount of smoke in the teenager's breath.

The Results

The scientists found something very interesting. The teenagers smoked more cigarettes when the actor was smoking, even if the actor didn't ask them to. But asking the teenagers to smoke didn't make them smoke more. This means that imitation might be more important than pressure.

What Does It All Mean?

This study shows us that just seeing someone else smoke can make us want to smoke too.

What Can We Do?

• Don't show smoking in ads: When we see people smoking in ads, it might make us want to smoke too.

• Make schools smoke-free: If we don't see people smoking at school, we might be less likely to start.

• Talk about imitation: We need to help each other understand that we copy what we see. If we know that seeing someone smoke might make us want to smoke, we can be more careful.

Link to Article