Does the introduction of financial incentives for a behavior have the power to form, change, reinforce, or undermine social norms?

The model shows how strategic communication (framing) of social norms and incentives can shape the way people respond to incentives, minimizing unwanted outcomes and even enhancing the effectiveness of behavioral incentive payments.

If payment programs are introduced into a community where there is an existing norm for cooperation (but perhaps not at the level needed), the payment program must be sustained in order for the cooperative behavior to continue.

The findings for H2 show that the relationship between the initial cooperation level of participants and contribution behavior is moderated by the presence or absence of a financial incentive such that the presence of the financial incentive reduced the effects of the initial group cooperation level on donation behavior.

An investment of one unit in the private account yields a return of one unit, and an investment in the public account yields a return of 1.6 units that is divided among the four players such that the individual investor in the public account gains a return of 0.4 from his own investment and also a return of 0.4 from other individuals' investments in the public account. In other words, contributing a token to the public account yields a private marginal return of 0.4 and a social marginal return of 1.6. This means that any given player has an incentive to free ride by investing only in the private account, but the best outcome for the group as a whole is for everyone to invest in the public account

In the two sorted treatments, participants were separated into groups based on their behavior in an initial “sorting” round

All treatments involved three phases, each consisting of six rounds of a standard PG game (the standard PG game is explained below). In the first phase, participants played a standard PG game without BPPs. In the second phase, the groups in the two incentive treatments received a BPP. In the final phase of all treatments, participants returned to playing the standard PG game without incentives.

H1: The relationship between perceived descriptive norms and contribution behavior will be moderated by presence or absence of a financial incentive.H2: The relationship between participants' initial cooperation level and contribution behavior will be moderated by the presence or absence of a financial incentive.H3: Receiving a financial incentive in the past ameliorates the positive effects of descriptive norms on contribution behavior when the incentive is removed.

Group identity refers to feelings of affinity with one's social group and the desire to be connected to that group

These perceptions are formed through observation of, and interpersonal communication about, behaviors.

Perceived norms have the potential to influence behaviors

Does the introduction of financial incentives for a behavior have the power to form, change, reinforce, or undermine social norms?

Others require the entire group to act or else the behaviors of the individual will not result in benefits.

and pay farmers to conserve or protect forests or water resources

In particular, we undertake a public goods (PG) experiment in which subjects receive a temporary incentive to cooperate.

This article addresses these issues; in particular, it undertakes a preliminary investigation of the effects of short-term monetary incentives on normative change and on peoples' behavior.

Descriptive and injunctive social norms communicated among groups are known to influence behavior

Regardless of this, the sub-sample of validatedcarbon monoxide outputs provided at 3-month follow-up correlated very strongly with theself-report measures, supporting previous evidence that self-report measures are highlyaccurate for smokers who are not adolescents, high risk smokers, or medical patients

First, as thesmokers included in the present study were recruited from community-based stop smokingprograms and were prepared to quit smoking within the next 30-days, it is likely that thesesmokers were more motivated to quit smoking than smokers in the general population.

This ispotentially important because 68% of adults in England engage in two or more unhealthybehaviors, of which hazardous alcohol use and smoking are the most commonly clusteredbehaviors [49]

Given that self-incentivizing implementation intentions could be deployed tochange other behaviors, it would be valuable in further research to identify the optimalschedules of self-incentivizing among other populations and behaviors

Contrary to predictions, there was no difference in smoking abstinence between self-incentivizing on a weekly versus a monthly basis, implying that self-incentivizing on amonthly basis is sufficient to change people’s behavior

Consistent with the analysis reported above, self-incentivizing worked equally well towardssmoking abstinence, regardless of whether participants incentivized themselves on a weeklyor monthly basis,

Biochemical verification (i.e., exhaledcarbon monoxide) was provided by 98.74% (157/159) of the sample at baseline and thiscorrelated perfectly with self-reported smoking abstinence,

Additionally, using objective measures such as exhaled carbonmonoxide levels would only provide an accurate representation of smoking status over thelast 8 hours [35

Exhaled carbon monoxide outputs were available from the stop smokingpractitioners to verify the self-reported measures at baseline and 3-month follow-up.

Recruitment and Participants

Therefore, self-incentivizing on a weekly basis (also coinciding with attendance at stop smoking sessions),for example, should therefore be more effective than self-incentivizing on a monthly basis.

“If I reach the end of the week and have not smoked at all, then I will reward myself by...”)

In fact, financial incentives are considered less acceptable to both UK and US populationsthan any other equally effective alternative

the rate ofquitting both in the UK and worldwide is decelerating

On average, 51% of smokers whoset a quit date during their first stop smoking session, committing to this quit date fromsession two successfully achieve and remain abstinent for at least 28-days

self-reported

if they had not smoked at all by the end of: (a) theweek (N = 44), or (b) the month (N = 50). T

Only a few patients described being distraught about the end of the intervention, indicating that they counted on it financially, or suggesting that they were only adherent because of the gift cards.

Some staff and investigators reported that they, or the providers they worked with, felt obligated to schedule visits or lab work around the requirements for gift card eligibility rather than clinic availability, space, or patient needs, and some resented this.

helped them meet their fiscal needs, and they described using the funds for necessities such as groceries, school supplies, bills, transportation, co-pays, or household and personal essentials

Some staff, however, noted a negative emotional consequence in that they felt the financial incentives created a sense of entitlement in a minority of patients who became aggressive or demanding about receiving their incentive. Most of the staff who described these negative experiences still seemed to have overall positive opinions about the financial incentive intervention. However, for a few staff, these experiences seemed to be central and negatively affected acceptability for them.

Some investigators indicated that they appreciated these positive benefits so much that they would gladly implement a financial incentive intervention again in the future, whether or not financial incentives could improve viral suppression.

Three investigators noted that their attitudes regarding the financial incentive intervention had changed from skepticism in the beginning to very supportive over the course of the study.

Nearly all patients, regardless of the number of gift cards they had received or their site, expressed an overall positive attitude regarding the financial incentive intervention.

The majority of patients (76%) reported an annual income of less than $20,000

All but one investigator interviewed were physicians.

Qualitative substudy

focus group discussions were conducted with site staff

Financial incentives have been used, with varying degrees of success, to encourage uptake of a range of health-related behaviors [12], but remain an emerging area of research within the realm of HIV

Patients, investigators, and staff found the intervention highly acceptable, primarily due to the emotional benefits gained through giving or receiving the incentive. Feeling rewarded or cared for was a main value perceived by patients; this was closely tied to the financial benefit for some.

quarterly $70 gift card financial incentives to HIV-infected patients on antiretroviral therapy (ART) to encourage ART adherence and viral suppression

However, a study of patient opinions regarding financial incentives demonstrated less favorable support for penalty-based incentives

allowing individuals to post more overt and explicit comments

Nevertheless, those negative views should be interpreted in the context in which the data were collected—in particular, the analysis of comments spontaneously written in response to newspaper reporting.

Third, this study demonstrates a novel approach to understanding a range of public sentiments through the use of online postings

I’ve read a number of the comments posted here, and I think everyone seems to missing the point of the study and incentives. . . . I know firsthand how big of a problem non-adherence can be. The cost on the healthcare system is astounding (anywhere from $100-250 billion a year). Patients who do not take their medications, for whatever the reason may be, are costing the U.S. billions of dollars in associated costs because they usually become more sick or have emergent problems which means hospitalization.

Even more comments (12%) reflected distrust of the health profession and/or the services it provides:

If this government doesn’t stop coming up with stupid ideas for spending money, I’ll scream! . . . it is NOT their job to dictate individual lives or take care of stupid citizens that cannot take their medicine. . . . Get real! This country is about individual responsibility.

This is the problem when you have free medical care for all

So theoretically this could mean that my premium will, in some form, be used to pay someone to take medications because they don’t have the willpower to do what they have to do on their own?

What if patients were FINED each month for not taking their medication? Certainly the incentive would still be in place.

Many readers recommended alternative incentives to promote healthy behaviors

Because the display name and location fields are open text boxes, readers who comment have varying degrees of anonymity.

Nevertheless, computer-mediated communication has been shown to provide emotional content similar to face-to-face communication along with enhanced sincerity and truthfulness of sentiments conveyed in online comments

Despite the growing interest in the use of financial incentives to improve health behaviors, research in this area is limited

That article reported on a study in which patients participated in a daily lottery allowing them to win a monetary prize—received only if they took their warfarin the day before.

In fact, paying people to quit smoking has an effectiveness that compares favorably to a variety of pharmacologic approaches to nicotine addiction

394 comments from their online audience. The authors systematically analyzed those online responses

However, paying people to improve their health touches on strongly held views about personal responsibility

proximity to the technology tends to diminish the credibility of experts with high-risk opinions

When experts frame an issue in a way that challenges dominant media frames, as is the case with our wind turbines experiment, the public's perception of source credibility may be low and statistically indistinguishable, irrespective of the risk frame and cultural bias introduced by their worldview.

We also hypothesized that hierarchs trust experts with scientific credentials, independently of their association with high- or low-risk frames.

The conditioning role of prior attitudes on framing is particularly apparent on the issue of wind turbines

“negativity bias” in which negative information tends to be more trusted relative to positive messages

Control for cell phone use relies on a question asking whether subjects are frequent (scored as 1) or infrequent (scored as 0) users of cellular telephones.

The scientific knowledge of the subject might also influence the perception of expert credibility.

Other controls include ideology, proximity to wind farms, cell phone use, levels of scientific knowledge, and perceptions of aggregate scientific opinion.

Nonexperimental Survey Measures

Do egalitarians trust such an expert (that is, an expert associated with a plausible high-risk frame on wind turbines), as much as they trust experts invoking high-risk frames on the other issues?

We further produced a treatment for wind energy that is largely inconsistent with the dominant discourse on renewable energy in Quebec

A Survey Experiment to Test the Power of Framing and Worldviews

Indeed, the IPCC is an example of a highly stratified information bureaucracy—it is ordered, regulated by rules, and based on meritocratic membership—which fits neatly into the hierarch's view that society works best when authority is institutionalized and when decisions are made by authoritative experts with the best knowledge of a given problem.

To

United Nations' Intergovernmental Panel on Climate Change (IPCC) argues that the Earth's temperature is increasing, and that such developments are driven predominantly by human activity (Intergovernmental Panel on Climatic Change

Information sources that frame risks in a manner that threatens a given orientation will be perceived as less credible among individuals who are predisposed to that particular worldview

Rather, culture is conceptualized as an attribute of individuals, who share orientations in common with others, and who may also contest the cultural preferences of fellow members in their collectivity

Under these conditions, individuals form their opinion on policy issues from the simple heuristics that are available to them. Expert framing of policy issues supplies such heuristics

These framing effects are, however, moderated by their congruity with one's underlying political worldviews and the external media environment in which they are interpreted.

In this article, we examine the roles of issue framing and worldviews1 in predicting perceptions of expert credibility

but also perceptions of expert credibility as wel

When faced with new information from policy experts, individuals must first decide whether or not they believe the source to be credible.

rational citizens have little incentive to invest their limited time to learn about the entire range of complex issues they face.

administered to 1,507 adult residents of Quebec,

Feeling rewarded or cared for was a main value perceived by patients; this was closely tied to the financial benefit for some.

One key reason for null results may be that because incentives were only offered over two weeks, this may have been too short of a time frame to truly observe changes in, for example, controlled motivation.

Those in the FI conditions were less likely than those in the non-FI conditions to do some form of (at least) moderate physical activity while not wearing the pedometer.

MCCs + FIs led to more full £15 deposits than the MCC-only condition

Those in the MCC+FI condition achieved more steps than both control groups, and there was some evidence of superiority over MCC-only.

All interventions were well accepted by participants; retention was high; measures well adhered to

Given those in the MCC+FI versus MCC-only condition were significantly more likely to deposit £15 versus £0

Assessing data collection procedures

Participants were mostly female (90%)

Measures

Control without goal (CWOG)

Control with goal (CG)

Monetary contingency and financial incentive (MCC + FI).

Monetary contingency contract (MCC).

Financial incentive (FI).

shopping vouchers

lassified as having low/moderate levels of physical activity

This was important because widespread acceptability of an (effective) intervention may maximise intervention effects

quantitative and qualitative

Controlled motivation, by increasing the importance of the behaviour, may also lead to self-monitoring of behaviour which is highly associated with achieving behavioural goals

Burns and Rothman (2018)

MCCs introduce an aversive stimulus (the threat of losing money) which is removed upon the performance of the desired behaviour

Larger rewards and rewards (i) received only upon achieving a goal and (ii) in non-lottery-based structures, appear more effective

Although public attitudes towards the use of FIs to change health behaviours are generally negative (especially compared to alternative interventions) (Promberger, Brown, Ashcroft, & Marteau, 2011), they are judged more acceptable when incentives are perceived as effective (Promberger, Dolan, & Marteau, 2012)

First, individuals offered FIs are likely those inherently low in intrinsic motivation (because they do not exercise regularly) and second, intrinsic motivation may not map strongly onto physical activity because, for example, individuals could exercise for less intrinsically motivating reasons (e.g., to improve appearance).

Perhaps the most accessible way of buffering against inactivity related conditions is to increase walking; a 20% increase in steps-per-day may produce tangible health benefits

It is possible that although financial incentivesare considered a potentially worthwhile health promo-tion tool, among mid-older adults with private health in-surance in our study, improved health had a more intra-personal motivating value

Ex-ploring the acceptability and perceived usefulness ofdifferent types of financial incentives as part of formativeresearch is one way of gaining insight to which particu-lar (if any) financial incentives may be worth includingin a lifestyle intervention with this population.

The appeal of peer supportseemed to stem from the potential of drawing motiv-ation from others who had shared a common experi-ence.

Specifically, being accountable to the pro-gram team and having someone monitoring their pro-gress would reinforce their commitment to make andsustain behaviour changes

Overall, participants seemed to have a sense of pridein, and ownership of, their weight loss and the behav-ioural changes they had made during the 18-week pro-gram

health as an incentive, individual responsibility for healthbehaviour, program support, and peer support

However, if financial incentiveswere to be used, participants favoured combiningthem with other means of ongoing support ratherthan in isolation.

Participants believed that youngerpeople may be more likely to be motivated by moneythan mid-older people.

they felt thatthey already “paid enough” in insurance premiums

improved health and wellbeing was regarded as the ultim-ate incentive for both initial weight loss and maintenancefollowing program completion.

Discussion about financial incentives conveyed a generalfeeling of distrust and indignation at the prospect of be-ing offered a financial incentive.

Yeah, something to maintain what you’ve achievedthrough the program ... I do think you plateau. Youhave a really good experience and then it flattens offfor a while, and sometimes you probably need thatmotivation, I don’t know what it would be, to just getoff the plateau and go down to the next level, youknow. (Female, 69 years, female-only group).Conversely, a few participants expressed no desire forongoing support as they felt they had made sustainableand lifelong changes.I think if you prove that you’ve maintained yourseventy kilo weight and it hasn’t changed, you’re self-sustainable. I don’t need the group. I’ve changed mybehaviour so much that you know what, this is rocksolid. (Male, 65 years, male-only group).

Maintenance program impressions

one third (33.8%, n = 25) answered that $50 would beenough; and 25.7% (n = 19) that $200 or more would beneeded.

The majority of these respondents (85.2%, n = 63) per-ceived that non-cash incentives would motivate weightloss maintenance.

Con-sistent with recruitment, the majority of Study Bparticipants were from the most advantaged socio-demographic areas, whereas Study A participants weremore evenly distributed across the different socio-demographic areas

Participants

Study a: Quantitative phase

who had completed the HWFL programin the year prior to November 2016

The current study comprised participants who com-pleted the Healthy Weight for Life (HWFL) program

We used a partially mixed methods sequential designwhere qualitative data was dominant over quantitativedata

There is a paucity of literature pertaining to the use offinancial incentives in Australian lifestyle interventionsinvolving mid-older adults

With increasing lifeexpectancy [20] and prevalence of non-communicabledisease among older Australian adults increasing

Australian

Whilst we were able to identify some aspects of HPFI that influence acceptability, the exact nature of an ‘acceptable’ HPFI is not clear and appears to be context-specific.

No attempt was made in any included papers to explore whether and how acceptability varied according to socio-demographic characteristics of members of the public.

Whilst quality appraisal was undertaken for the empirical papers, this was not possible for the scholarly writing as no relevant tool could be located.

A lack of relevant data meant we were not able to come to any clear conclusions on whether acceptability varied according to the nature of the behaviour incentivised or the population rating acceptability.

The potential for a perpetuating cycle of personal failure was also discussed. Lack of success in behaviour change, emphasised by failure to gain a reward, may lead to demotivation and even greater difficulty in future attempts at behaviour change.

If the incentives motivate people in higher socioeconomic groups more than those in lower socioeconomic groups, however, they could exacerbate health inequalities.

The view that participants might misuse the rewards or lie to get them was a common objection

Some authors argued that HPFI can encourage individuals to take responsibility for themselves, thereby promoting autonomy

Some concern was raised with providing cash incentives to help control substance misuse as rewards could be used to fund the very behaviour it is designed to prevent

If incentives have a place in smoking cessation, it is perhaps this group who might be seen as the most deserving.

One consistently mentioned issue is that the behavior will revert to baseline once reinforcers are no longer offered

In health promotion, the most frequent misuse of reinforcement occurs when it is applied to outcomes rather than behaviors

Even with other health behaviors, Warner and Murt found that larger rewards had a greater impact than smaller ones, and repeated reinforcement was preferable to one-time incentives.

Firstly, empirical research has found that many socio-economically disadvantaged individuals are not willing to accept government funded HPFI under any circumstances — although reasons for this have not been explored.

Other authors argued that as the target recipients of HPFI are often vulnerable groups, who are most in need of financial resources, the choice to engage is rarely ‘voluntary’.

Some authors of included papers argued that if parties act voluntarily, the mutual benefits of HPFI make them acceptable.

Table 3. Quality assessment of quantitative empirical papers.

Eighty one papers met the inclusion criteria: 22 empirical studies (Allan et al., 2012, Arterburn et al., 2008, Bonevski et al., 2011, Bonevski et al., 2012, Cameron and Ritter, 2007, Ducharme et al., 2010, Hoddinott et al., 2014, Kim et al., 2011, Long et al., 2008, Luyten et al., 2013, Lynagh et al., 2011, Mantzari et al., 2012, Meads et al., 2013, Meredith et al., 2011, Ni Mhurchu et al., 2011, Park et al., 2012, Parke et al., 2013, Promberger et al., 2011, Promberger et al., 2012, Raiff et al., 2013, Ritter and Cameron, 2007, Thomson et al., 2012) and 59 pieces of scholarly writing (see Fig. 1 and Table 1, Table 2) (Ashcroft, 2011, Ashcroft et al., 2008, Aveyard and Bauld, 2011, Axtell-Thompson, 2012, Blacksher, 2008, Blumenthal-Barby and Burroughs, 2012, Burry, 2006, Cawley, 2014, Cookson, 2008, Dreger, 2012, Donatelle et al., 2004, Grant, 2002, Grant, 2006, Goel, 2012, Halpern et al., 2009, Haveman, 2010, Higgins et al., 2012, Horwitz et al., 2013, Kennedy, 2012, Klein, 2012, Klein and Karlawish, 2010, Kowal, 2006, Lawson and Howard, 2012, Lesser and Puhl, 2014, Lewis, 2008, Loeppke, 2012, London et al., 2012, Lunze and Paasche-Orlow, 2013, Luyten et al., 2011, Madison et al., 2011, Malone and Jason, 1990, Marteau et al., 2009, Marteau et al., 2008, Meredith et al., 2014, O'Donnell, 2012, Oliver, 2009, Oliver, 2012, Oliver and Brown, 2012, Pearson and Lieber, 2009, Petry, 2010, Popay, 2008, Robison, 1998, Roozen, 2009, Schmidt, 2008, Schmidt, 2012, Schmidt et al., 2009a, Schmidt et al., 2009b, Schmidt et al., 2012, Serxner, 2013, Sindelar, 2008, Stephens, 2014, ten Have et al., 2013, Terry, 2013, Terry and Anderson, 2011, Voigt, 2012, Volpp and Galvin, 2014, Volpp et al., 2009, Volpp et al., 2011, Wu, 2012).

Synthesis of results

he quality of empirical research papers using qualitative methods was assessed using a tool developed for this purpose

After exclusion of duplicates, one researcher (ELG) screened titles and excluded those definitely not relevant. Next, the same researcher screened remaining titles and abstracts, again excluding those definitely not relevant. Finally, remaining full texts were screened by two researchers independently (ELG & JA) to identify those meeting the inclusion criteria. If in doubt, papers were retained at any stage for inspection by both reviewers, with disagreements resolved by discussion.

Databases searched were: Medline (1946), Embase (1980), Web of Knowledge (1970), Cumulative Index to Nursing and Allied Health Literature (1981), PsycINFO (1806), Applied Social Science Index and Abstracts (1970), Sociological Abstracts (ProQuest, 1952), Scopus (1960), The Philosopher's Index (OVID, 1940), the Cochrane library (Issue 3), Social Science Citation Index (1970) and the International Bibliography for the Social Sciences (1951).

In the United Kingdom (UK), the current government has signalled their interest in using HPFI as part of their ‘nudge’ agenda

In the United States of America (USA), the 2010 Affordable Care Act allowed employers to offer rewards, or impose penalties, for those meeting healthy behaviour targets such as quitting smoking

Although there was consensus that if financial incentives are effective and cost effective they are likely to be considered acceptable, a number of other factors also influenced acceptability.

Five themes were identified: fair exchange, design and delivery, effectiveness and cost-effectiveness, recipients, and impact on individuals and wider society.

Participants may have interpreted grocery vouchers for weight loss to have higher than stated effectiveness, and future studies should attempt to check for this by assessing perceived effectiveness.

the cost of standard treatment may otherwise be neglected compared to that of financial incentives.

On the other hand, acceptance might be higher if the standard vs. incentive effectiveness difference were presented in relative rather than absolute terms (“100% increase in effectiveness”).

As we described them, vouchers for “healthy groceries” may have been seen as instrumental for weight loss, and seen as a treatment rather than a reward.

The higher acceptability of incentives for weight loss may reflect different levels of “moralization” of the two behaviours, a process conferring moral significance to health-related behaviours

All incentives were more acceptable for weight loss than for smoking cessation.

The acceptability of incentive-based treatments increases with effectiveness: most participants were willing to trade off their dislike of incentive treatments against effectiveness.

Separate models including participants' overweight and smoking status and their interaction with incentive type show that both the very overweight and daily smokers are more in favour of incentive-based treatments than those who were never overweight or never smoked

A further 21 (27%) consistently refused at least one incentive type for smoking cessation, most of them cash (14); 21 (21%) did so for weight loss, most of them luxury item vouchers (14).

Results

About 400 participants were invited to Study 1 and were informed that recruitment would close at 80 participants. About 300 participants were invited to Study 2 and informed recruitment would close at 100 participants. Each study closed after about 48 h.

We asked whether participants were overweight (never – formerly – slightly – very) and whether they smoked (never – have quit – not daily – daily).

The patient receivesThe patient receivesvouchers for healthy groceriesstandard medicationworth £50 per month for not smoking.worth £50 per month for stopping smoking.This intervention is proven to helpThis treatment is proven to help20 out of 100 treated.10 out of 100 treated.

£50 per month was the same for all choice pairs, and is an amount typical for such incentive programmes

To keep the number of choice pairs manageable, we dropped the incentive type of luxury item vouchers, as its effect was no different to cash in Studies 1 and 2.

Study 3 built on the online studies to replicate and extend the findings in an offline sample.

Design

Methods

which used large incentives

there is currently less evidence regarding their effectiveness in weight loss programmes

incentives were more acceptable for weight loss than for smoking cessation

the reward seems more likely to be seen as a welcome additional benefit for a behavior for which motivation already exists

However, for most of these behaviors, the prominent conflict of interest is less between different persons and more between different outcomes within the person him- or herself.

From a policy perspective, the motivating effect of a large enough incentive might be worth it, depending on how much beneficial behavior is gained.

These behaviors are most likely primarily motivated by interest in the health benefits to oneself, but may additionally be motivated by considerations for benefits to others.

All three of the potential mechanisms suggested by Bowles (2008) for an undermining effect of incentives in addition to cognitive evaluation theory rely on this conflict of interest: framing of the decision as one where self-interest is the appropriate guiding principle; the incentive system shaping long-term narrowly self-interested preferences; the incentive conveying information, for example about lack of trust in employee work effort.

employees work even when not supervised

behavior is not maximally self-interested

As far as we are aware, no studies in the context of cognitive evaluation theory involve behaviors that challenge impulse control in this way.

An individual must trade these conflicting preferences off against each other when deciding what to do.

Rewards would be offered to individuals with low rather than high initial behavior levels

One possible mechanism suggested by the authors is that money may have enabled the men to purchase risky sex and enabled the women to avoid selling it, which would imply the incentive to have changed, for both men and women, the means to follow through on pre-existing motivation.

the incentivized groups lost more weight than the control group during incentivization, and at 7-month follow-up participants in the previously incentivized groups, unlike those in the control group, weighed less than at baseline.

The group receiving incentives did not differ in motivation from the group receiving no incentives up to 3 months after the incentives had stopped, hence there was no evidence in this study either that the incentives reduced motivation.