Using the Welfare Trade-Off Task (WTT), the authors show MDMA produces context-dependent prosocial effects: 1.0 mg/kg increased generosity toward friends, while 0.5 mg/kg produced a slight increase in generosity toward strangers, particularly in women. Baseline WTT generosity correlated with household income and trait Agreeableness, and the study proposes the WTT as a novel tool to quantify generosity, with underlying neural mechanisms yet to be determined.
Background
3,4-methylenedioxymethamphetamine (MDMA) produces “prosocial” effects that contribute to its recreational use. Few studies have examined the cognitive and behavioral mechanisms by which MDMA produces these effects. Here we examined the effect of MDMA on a specific prosocial effect, i.e. generosity, using a task in which participants make decisions about whether they or another person will receive money (Welfare Trade-Off Task; WTT).
Methods
The project included one study without drug administration and one with MDMA. In Study 1, we administered the WTT to healthy adults ( N = 361) and examined their performance in relation to measures of personality and socioeconomic status. In Study 2, healthy volunteers with MDMA experience ( N = 32) completed the WTT after MDMA administration (0, 0.5, or 1.0 mg/kg).
Results
As expected, in both studies participants were more generous with a close friend than an acquaintance or stranger. In Study 1, WTT generosity was related to household income and trait Agreeableness. In Study 2, MDMA (1.0 mg/kg) increased generosity toward a friend but not a stranger, whereas MDMA (0.5 mg/kg) slightly increased generosity toward a stranger, especially among female participants.
Conclusions
These data indicate that the WTT is a valuable, novel tool to assess a component of prosocial behavior, i.e. generosity to others. The findings support growing evidence that MDMA produces prosocial effects, but, as with oxytocin, these appear to depend on the social proximity of the relationships. The brain mechanisms underlying the construct of generosity, or the effects of MDMA on this measure, remain to be determined.
Papers cited by this study that are also in Blossom
Bedi, G., Hyman, D., De Wit, H. · Biological Psychiatry (2010)
Dumont, G., Sweep, F., van der Steen, R. et al. · Social Neuroscience (2009)
´dric, C., Hysek, M., Schmid, Y. et al. · Social Cognitive and Affective Neuroscience (2013)
Kirkpatrick, M. G., Francis, S. M., Lee, R. et al. · Psychoneuroendocrinology (2014)
MDMA (3,4-methylenedioxymethamphetamine) is reported to produce prosocial or empathogenic effects such as increased sociability and interpersonal closeness, but the cognitive and behavioural mechanisms underlying these effects are not well characterised. Earlier experimental work has been limited by a lack of objective laboratory tasks tailored to measure specific components of prosociality, such as generosity. Prior studies have shown that oxytocin can increase generosity in monetary allocation tasks and that MDMA raises plasma oxytocin, suggesting a possible mechanistic link; a recreational dose of MDMA has also been reported to increase prosocial choices in a resource allocation paradigm. This paper uses a novel Welfare Trade-Off Task (WTT), adapted from social discounting and delay-discounting approaches, to quantify generosity as the willingness to forgo personal monetary gain to benefit another person. Kirkpatrick and colleagues report two studies. Study 1 describes WTT performance in 361 healthy young adults and tests associations with personality (NEO-FFI) and socioeconomic status. Study 2 is a within-subject, double-blind trial in 32 MDMA-experienced volunteers that tested acute effects of two MDMA doses (0.5 and 1.0 mg/kg) versus placebo on WTT generosity toward a close friend and a stranger. The investigators predicted dose-dependent increases in generosity regardless of social proximity.
Study 1 recruited healthy adults aged 18–35 (N = 361; 57% female) as part of a larger genetic study; inclusion required at least a high school education and fluency in English, and participants were restricted to Caucasian ethnicity for that larger project. Screening excluded current psychiatric disorders and medication use. Participants completed a single laboratory visit after abstaining from alcohol, marijuana and other recreational drugs for specified intervals; abstinence was verified by breath and urine tests. Personality was assessed with the NEO-FFI (five traits), and household income was reported on a 21-point scale then grouped into four roughly equal categories. The primary behavioural measure was the computerized Welfare Trade-Off Task (WTT): for a named close friend and a named acquaintance participants made a series of hypothetical choices between receiving money for themselves or giving a larger amount to the other person. Anchor values for the recipient were fixed ($19–$75 across six series), while the self-payoff varied to produce a set of Welfare Trade-Off Ratios (WTRs). The switch point where a participant changed from preferring self to preferring the other person is taken as the WTR; higher WTRs indicate greater generosity. Analysis in Study 1 comprised a paired t-test comparing Friend versus Acquaintance WTRs and two multiple regression models (forced entry) predicting WTR for Friend and for Acquaintance from eight predictors (sex, age, income group, and the five NEO-FFI traits). Stepwise follow-up regressions were used to identify more parsimonious models and verify significant predictors; significance was set at p < 0.05. Study 2 enrolled MDMA-experienced healthy adults aged 18–30 (N = 32; 9 female, 23 male) who had used MDMA between 4 and 80 times. Medical and psychiatric screening (including ECG) excluded individuals with conditions that increased study risk. The design was within-subjects and double-blind: each participant completed three outpatient sessions separated by at least 5 days and received placebo, 0.5 mg/kg MDMA and 1.0 mg/kg MDMA in randomized order; average absolute doses were ~36 mg and ~72 mg given participants’ average weight. Participants provided biological samples to confirm abstinence before each session; capsules contained MDMA powder or lactose placebo. The WTT was completed at approximately 90 minutes after dosing (11:00), and other physiological and subjective measures were collected at multiple time points. For Study 2 the WTT probed generosity toward a named close friend and a designated stranger; post-hoc checks found acquaintance and stranger switch points were comparable across studies. Repeated measures ANOVA tested Drug Dose (placebo, 0.5, 1.0 mg/kg) and Person (Friend, Stranger) as within-subject factors with Participant Sex as a between-subjects factor. Significant main effects or interactions were followed by one-tailed t-tests comparing doses; alpha was 0.05.
Study 1: On the WTT participants were more generous to a close friend than to an acquaintance (mean WTRs 0.65 ± 0.02 versus 0.41 ± 0.02; t(359) = 17.4; p < 0.001). The full regression model predicting Friend WTR was not significant (F[8,360] = 1.5, p = 0.17, R2 = 0.032), but a parsimonious model retained Agreeableness alone (F[1,360] = 8.7, p = 0.003, R2 = 0.024). In that model each one-point increase on the Agreeableness scale was associated with a 0.006 point increase in the Friend switch point (95% CI 0.000–0.012; t(360) = 2.1; p < 0.05). Put differently, using the paper's illustrative conversion, for a friend who could receive $100 a participant at the lowest possible Agreeableness would trade off about $41 whereas a participant at the highest Agreeableness would trade off about $70. For the Acquaintance, the full model was significant (F[8,360] = 2.6, p = 0.009, R2 = 0.056). The best-fitting model included Agreeableness and Income group (F[2,360] = 8.5, p < 0.001, R2 = 0.046). Agreeableness increased Acquaintance switch points by 0.008 per point (95% CI 0.003–0.013; t(360) = 3.0; p < 0.005). Household income was negatively associated with generosity to an Acquaintance (B = -0.038; 95% CI -0.067 to -0.010; t(360) = 2.7; p < 0.01), indicating lower-income participants were more generous to acquaintances (example conversion: lowest income group willing to trade off about $11 more than highest income group when the acquaintance could receive $100). No other demographic or personality traits (including Extraversion) significantly predicted WTRs. Study 2: Across drug conditions participants remained more generous to Friends than Strangers (main effect of Person: F(1,30) = 22.2; p < 0.001). There was a Dose × Person interaction (F(2,60) = 3.5; p < 0.05). Specifically, the higher MDMA dose (1.0 mg/kg) increased WTR switch points toward Friends compared with placebo (comparison between 1.0 mg/kg and placebo: t(31) = 1.8; p < 0.05). The authors illustrate this as a change from trading off about $57 under placebo to about $72 under 1.0 mg/kg MDMA (an increase of roughly $15 for a hypothetical $100 recipient). When the other person was a Stranger, the lower dose (0.5 mg/kg) produced a modest increase in WTRs in women but not in men, as indicated by a Dose × Sex quadratic interaction (F(1,30) = 4.8; p < 0.05). For women the mean difference between 0.5 mg/kg MDMA and placebo on Stranger WTR was 0.25 ± 0.11 (95% CI 0.03–0.48; p < 0.05), whereas for men the mean difference was 0.04 ± 0.07 (95% CI -0.10–0.18; p = 0.56). No other sex differences in drug response reached significance.
Kirkpatrick and colleagues interpret the findings as supporting the WTT as a useful laboratory measure of a facet of prosocial behaviour — generosity — that is sensitive to social proximity and to individual differences. The key descriptive results were that participants were more generous to close friends than to acquaintances or strangers, that higher trait Agreeableness predicted greater generosity to both friends and acquaintances, and that lower household income was associated with greater generosity to acquaintances but not to friends. These patterns align with prior observations that agreeableness relates to prosocial dispositions and that socioeconomic status can inversely relate to generosity in some contexts. Regarding pharmacological effects, the investigators report that MDMA increased generosity in a dose- and relationship-dependent manner: the 1.0 mg/kg dose increased generosity toward a close friend but not a stranger, whereas the lower dose produced a small increase in generosity to strangers primarily in women. The authors relate these findings to prior work on oxytocin and prosocial behaviour, noting that MDMA elevates plasma oxytocin and that intranasal oxytocin has context-dependent effects (for example, preferentially enhancing in-group trust). They suggest the observed dependence on social proximity may be consistent with oxytocin-related mechanisms but acknowledge that serotonin or other non-specific effects may also contribute. The paper notes several limitations acknowledged by the authors. Study 1 used a relatively homogeneous sample (young, Caucasian, screened for psychiatric symptoms), limiting generalisability. Study 2 had a small sample size, which constrains confidence in subgroup findings such as the sex-specific effect for strangers. The mechanism linking MDMA to increased generosity was not directly tested (for example, oxytocin levels were not measured in these data), and the effects of dose order or longer-term persistence of prosocial changes were not established. The authors recommend that future research examine biological mediators (e.g. oxytocin), include larger and more diverse samples, and compare MDMA's effects with other psychoactive drugs. In closing, the investigators conclude that the WTT captures an interpretable component of prosocial behaviour and that MDMA can increase generosity in a manner shaped by social relationship; they suggest the task will be useful for further studies of prosocial effects of MDMA and other substances, and note that such prosocial effects may contribute to MDMA's recreational appeal and abuse potential.
Study 1: Descriptive data using Welfare Trade-Off Task Participants. Healthy adult volunteers aged 18-35 were recruited through online advertisements, flyers on community bulletin boards, and word-of-mouth referrals. Inclusion criteria were at least a high school education, fluency in English, and Caucasian ethnicity because this was part of a larger study designed to examine the genetic basis of a range of mood, cognition, and behaviors. Subjects underwent a brief psychiatric screening interview, and subjects who reported symptoms suggestive of any DSM-IV Axis I disorder, including substance use disorders, were excluded. Subjects currently taking any medications were also excluded. In total, 361 volunteers (57% female) completed Study 1. They were mean ± SD 23.2 ± 3.3 years old and had completed 15.4 ± 1.7 years of formal education. Women were significantly younger than men (22.8 ± 0.2 and 23.7 ± 0.3, respectively: t(359) = 2.3; p < 0.05) but the women and men did not differ in education. Additionally, 344 participants regularly drank alcohol (7.1 ± 5.2 drinks/week), 49 currently smoked marijuana (38 reported at least one occasion per week; 11 reported daily use), and 50 participants were daily tobacco smokers (9.1 ± 9.0 cigarettes/day).
We conducted a paired t-test to test whether individuals were more generous to the Friend than to the Acquaintance. To investigate individual factors that may influence generosity, we conducted two multiple regressions: one for each dependent variable (the welfare trade-off switch points for Friend and for Acquaintance). Because our goal was to investigate the influence of each individual independent variable while controlling for all other variables, all eight predictors were entered into each model using the forced entry method. The predictors were the three demographic variables (Sex, Age, and Income group) and the five trait personality variables (Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness). We also conducted follow-up regression analyses using stepwise methods (both forward and backward) to identify the most parsimonious model for each dependent variable and to verify the significant relationships between the predictors and the outcomes derived from the full models. For all analyses, p values were considered statistically significant at < 0.05.
These analyses with the WTT provided several interesting findings. As expected, participants were more generous with a close friend than an acquaintance or stranger. Interestingly, however, in Study 1 generosity on the WTT was negatively correlated with household income. That is, participants reporting higher incomes were less likely to share money with acquaintances than were participants with lower incomes. Household income was not related to generosity toward a close friend. We found that participants high on the personality trait of Agreeableness were more generous toward both close friends and acquaintances. In Study 2 we found that MDMA (1.0 mg/kg) increased generosity toward a friend, but not toward a stranger, whereas the lower dose (0.5 mg/ kg) did not affect generosity toward the friend but slightly increased generosity toward a stranger. Taken together, the results extend our knowledge of the behavioral trait of generosity, and show that the WTT may be a useful measure in future drug studies of prosocial behavior. In Study 1, generosity was negatively related to household income and positively related to the personality measure of trait Agreeableness. The negative relationship to income may be intuitively paradoxical, but it is consistent with previous reports that individuals with lower socioeconomic status may be more generous than those with higher socioeconomic status. For example,reported that upper-class individuals (defined by parental educational attainment) reported greater psychological entitlement and narcissistic personality tendencies, and the authors concluded that social stratification contributes to basic psychological processes. The positive relationship between generosity and agreeableness is consistent with previous researchand intuitively more direct, and suggests that a positive disposition also has behavioral consequences on willingness to share financial resources. It is also notable that the closely related personality trait of Extraversion did not predict WTRs; this speaks to the specificity of the construct captured by the WTT. One limitation of Study 1, however, is that the sample was relatively homogeneous, consisting of Caucasian young adults with a high school education and no serious psychiatric symptomatology. It is not known whether these findings would generalize to a more heterogeneous population. We also found that MDMA increased generosity, and that this effect depended on the dose and on the relational closeness of the other recipient. When participants had the opportunity to give money to either themselves or a close friend, the larger dose of MDMA increased generosity toward the other person, whereas when the other person was a stranger, the larger dose did not produce this effect. The increase in generosity observed here is consistent with earlier reports that MDMA increases "prosocial" resource allocation and emotional empathy. In our study, the most pronounced increase in generosity was observed after the higher dose of MDMA, and only in relation to a close friend, not a stranger. This finding is strikingly concordant with a recent report on the effects of oxytocin, a neuropeptide involved in social bonding and thought to be related to responses to MDMA. In addition to its actions of serotonin and norepinephrine release, MDMA increases levels of oxytocin, and these effects are thought to contribute to its empathogenic effects. Intranasal oxytocin itself can increase prosocial behavior, dependent upon on the context in which it is experienced (seefor a review). For example, oxytocin promotes trust of people who are members of a social ingroup but does not affect trust of out-group members; see Van IJzendoorn and Bakermans-Kranenburg, 2012 for a review). This relationship is consistent with our observation that MDMA (1.0 mg/kg) increased generosity toward the close friend but not toward the stranger. It is not clear why the drug at the lower dose slightly increased generosity to the stranger. The differences in dose effects may be related to the potency of the drug to release oxytocin or serotonin, which is also involved in prosocial behaviors, or may reflect a relatively nonspecific drug effect at the low dose. Future studies may examine whether the increases in generosity are indeed related to oxytocin levels, or whether these similar effects occur with other drugs, such as the related stimulant amphetamine or the sedative diazepam. It is also possible that the prosocial effects of MDMA endure even after plasma levels return to baseline, as recent clinical studies indicate that MDMA plus psychotherapy produces long-lasting improvements in PTSD symptoms and social relationships. In the present study there was no evidence that drug order (i.e. administration of drug or placebo in a previous session) affected responses, but this remains a subject of interest. Interestingly, the increase in generosity toward the stranger occurred primarily in women. Although this sex difference is consistent with at least one previous report indicating that MDMA may produce more intense psychoactive effects in women compared to men, it is inconsistent with the previous MDMA study showing that increases in monetary resource allocation occurred primarily in men. Differences in dose or task design and our relatively small sample size could explain the apparent discrepancy between the finding of Hysek et al. and ours. Future investigations of the prosocial effects of MDMA may confirm whether these effects differ by sex. Considering that menstrual cycle phase and its associated alterations in sex hormone concentrations are associated with changes in oxytocin levels, stimulant response, and prosocial behaviors such as empathy, it is possible that fluctuations in sex hormone levels substantially alter the prosocial effects of MDMA in women. In conclusion, we demonstrate the value of the WTT in assessing a novel behavioral construct, that of generosity to others. Generosity to friends or strangers plays an important role in social structures and in the functioning of individuals within a group. The fact that generosity was related to both socioeconomic status and to the personality trait of Agreeableness adds to our understanding of individual differences in the measure. The fact that MDMA increased generosity is consistent with some of its anecdotal prosocial effects, and may be related to its purported mechanism via release of oxytocin. The task is likely to be useful in future studies investigating the prosocial effects of MDMA and other drugs of abuse that are used to enhance social behavior. Considering that MDMA's prosocial effects may be an important factor in its abuse potential, this task may help to identify social factors that increase the reward value of abused drugs.
Create a free account to open full-text PDFs.
Kirkpatrick, M. G., Lee, R., Wardle, M. C. et al. · Neuropsychopharmacology (2014)
Liechti, M. E., Gamma, A., Vollenweider, F. X. · Psychopharmacology (2001)
Mithoefer, M. C., Wagner, M. T., Mithoefer, A. T. et al. · Journal of Psychopharmacology (2010)
Wardle, M. C., De Wit, H. · Psychopharmacology (2014)
Papers in Blossom that reference this study
Ramaekers, J. G., Kuypers, K. P. C., Vollenweider, F. X. · Molecular Psychiatry (2026)
Haggarty, C. J., Bershad, A. K., Kumar, M. K. et al. · European Journal of Neurology (2024)
Bershad, A. K., Hsu, D. T., De Wit, H. · Journal of Psychopharmacology (2024)
Lewis, C. R., Tafur, J., Spencer, S. et al. · Frontiers in Psychiatry (2023)
Earp, B. D., Repantis, D., Langlitz, N. et al. · Frontiers in Psychiatry (2021)
Chaliha, D., Mamo, J. C., Albrecht, M. et al. · Current Neuropharmacology (2021)
de Wit, H., Bershad, A. K., Grob, C. S. · Journal of Psychopharmacology (2021)
Feduccia, A. A., Jerome, L., Mithoefer, M. C. et al. · Psychopharmacology (2020)
Greenway, K. T., Garel, N., Jerome, L. et al. · Expert Review of Clinical Pharmacology (2020)
Preller, K. H., Vollenweider, F. X. · Frontiers in Psychiatry (2019)
Holze, F., Vizeli, P., Müller, F. et al. · Neuropsychopharmacology (2019)
Carlyle, M., Stevens, T., Fawaz, L. et al. · Journal of Psychopharmacology (2019)
Jungaberle, H., Thal, S., Zeuch, A. et al. · Neuropharmacology (2018)
Papaseit, E., Torrens, M., Pérez-Mañá, C. et al. · Expert Opinion on Drug Metabolism and Toxicology (2018)
Schmid, Y., Schmidt, A., Müller, F. et al. · International Journal of Neuropsychopharmacology (2017)
Phelps, J. · Journal of Humanistic Psychology (2017)
Dolder, P. C., Müller, F., Schmid, Y. et al. · Psychopharmacology (2017)
Bershad, A. K., Miller, M. A., De Wit, H. · Psychopharmacology (2017)
Bershad, A. K., Miller, M. A., Baggot, M. J. et al. · Journal of Psychopharmacology (2016)
Garcia-Romeu, A., Kersgaard, B., Addy, P. H. · Experimental and Clinical Psychopharmacology (2016)
Amoroso, T., Workman, M. · Journal of Psychopharmacology (2016)
Kamilar-Britt, P., Bedi, G. · Neuroscience and Biobehavioral Reviews (2015)