| 英文摘要: | To the Editor —
The term 'warm glow' was first introduced by economist James Andreoni in an attempt to explain why people sometimes act altruistically (for example, donate to charity)1. The theory suggests that people often act pro-socially because it is rewarding; we derive a positive emotional experience from the act of helping others (that is, a warm glow). In a recent Letter, Taufik, Bolderdijk and Steg2 show that acting green elicits a literal warm glow: people's psychological state directly influences their thermal state. In the experiment, participants who acted in an environmentally friendly way perceived significantly higher temperatures than those who did not. Importantly, the effect on perceived temperature was mediated by a positive self-signal.
What Taufik and colleagues describe in their study is closely related to a psychological concept known as the 'helper's high'3: when doing good actually makes people feel good. Although Taufik et al. did not find any systematic effects on changes in skin temperature, there is a substantial body of research suggesting that when people do good this often results in many observable physiological and psychological benefits4. In fact, the brain's response to increasing body heat (warm glow) is the release of 'feel-good' neurotransmitters such as oxytocin5. By doing good or the 'right' thing, I do not refer to opaque cultural conceptualizations of good and evil, but rather to morality as an evolved capacity. Moral emotions such as empathy and the ability to be compassionate are evolutionarily adaptive traits6. It is therefore not entirely surprising that doing the right thing sends a positive self-signal: the act of helping actually makes people feel good, both physically as well as psychologically.
Yet, while the study by Taufik et al. clearly adds to a growing body of research highlighting that people are motivated by more than just monetary and extrinsic incentives7, 8, 9, I argue that the authors are in fact too modest in stressing the importance of understanding the intrinsic motivational basis of pro-environmental behaviour. One major challenge in behavioural science and psychological research more generally is the (in)ability to sustain (experimental) treatment effects over time10, 11. In the face of many urgent global challenges, whether social (for example, inequality), public health (disease), economic (poverty) or environmental (climate change), understanding how to make positive behaviour change stick is one of the most pressing policy-relevant (but under-researched) questions in social science today. I argue that the answer to this question lies in recognizing both the fundamental limitations of extrinsically- oriented incentives as well as the severely under-leveraged potential of intrinsically motivated behaviour.
To illustrate this principle, I analyse (Fig. 1) the behavioural impact of a campus-wide energy conservation campaign that was recently administered at Princeton University. The 2014 'Do-It-in-the-Dark' campaign (www.wattvision.com/competition/princeton/home) is a perfect example of a popular nationwide energy competition initiative where students across universities (more than 100 took part) are encouraged to reduce their residential energy consumption over the course of a month (the competition period). The so-called campus conservation nationals (CCN; www.competetoreduce.org) is the largest competition of its kind. Usually, several prizes are handed out to the winners of the competition. Using interrupted (change-point) time-series analysis, I assess the (slope-) changes in the energy-usage trend shortly before (pre), during, and after (post) the competition was launched. It is clear from Fig. 1 that the competition noticeably reduced aggregate residential energy consumption across campus. Moving from the baseline (pre) to the competition period, the direction of the trend changes significantly, visibly shifting downward (β change = −111.21, t = −3.88, P < 0.01, 95% CI:−168.69, −53.74). What is particularly interesting, however, is that there is another significant trend change: as soon as the competition ends the positive effect of the intervention is reversed and energy usage bounces right back to the baseline consumption level before the competition was launched (β change = 88.81, t = −5.01, P < 0.01, 95% CI: 53.25, 124.38). |