The Role of Temperature in Moral Decision-Making: Limited Reproducibility. Ryunosuke Sudo et al. Front. Psychol., September 28 2021. https://doi.org/10.3389/fpsyg.2021.681527
Abstract: Temperature is one of the major environmental factors that people are exposed to on a daily basis, often in conditions that do not afford control. It is known that heat and cold can influence a person’s productivity and performance in simple tasks. With respect to social cognition, it has also been suggested that temperature impacts on relatively high-level forms of decision-making. For instance, previous research demonstrated that cold temperature promotes utilitarian judgment in a moral dilemma task. This effect could be due to psychological processing, when a cool temperature primes a set of internal representations (associated with “coldness”). Alternatively, the promotion of utilitarian judgment in cold conditions could be due to physiological interference from temperature, impeding on social cognition. Refuting both explanations of psychological or physiological processing, however, it has been suggested that there may be problems of reproducibility in the literature on temperature modulating complex or abstract information processing. To examine the role of temperature in moral decision-making, we conducted a series of experiments using ambient and haptic temperature with careful manipulation checks and modified task methodology. Experiment 1 manipulated room temperature with cool (21°C), control (24°C) and hot (27°C) conditions and found only a cool temperature effect, promoting utilitarian judgment as in the previous study. Experiment 2 manipulated the intensity of haptic temperature but failed to obtain the cool temperature effect. Experiments 3 and 4 examined the generalizability of the cool ambient temperature effect with another moral judgment task and with manipulation of exposure duration. However, again there were no cool temperature effects, suggesting a lack of reproducibility. Despite successful manipulations of temperature in all four experiments, as measured in body temperature and the participants’ self-reported perception, we found no systematic influence of temperature on moral decision-making. A Bayesian meta-analysis of the four experiments showed that the overall data tended to provide strong support in favor of the null hypothesis. We propose that, at least in the range of temperatures from 21 to 27°C, the cool temperature effect in moral decision-making is not a robust phenomenon.
General Discussion
The present study examined in detail the effect of ambient and haptic temperature on social judgment, focusing on the effect of cold temperature in a moral dilemma task, following on from earlier work by Nakamura et al. (2014). In one of the four experiments here, we found a cool temperature that promoted utilitarian judgment, similar to the previous study. The remaining experiments, however, produced weak effects in the opposite direction or no effect of temperature on moral judgment. This occurred despite the fact that our temperature manipulations elicited reliable differences in perceptions of coldness, feelings of comfort, and physiological measurements of skin temperature.
A meta-analysis of the normalized data from all experiments, using Bayesian testing, provided firm evidence in favor of the null hypothesis. Taken together, our findings trace the limited reproducibility of effects from temperature on moral judgment and thus serve to caution against overinterpretation when psychologizing about the embodied “cold-heartedness” or “cool-headedness.”
One important caveat here is that we worked within a safe range of temperatures, between 21°C and 27°C, in line with the ethical guidelines at the universities where the experiments were carried out. In this setting, we followed temperature studies of social judgments that set cold temperature in the range of approximately between 20°C and 22°C (e.g., Gockel et al., 2014; Wang, 2017). However, the 21°C here reflects a cool temperature within the range used in this study, and could be interpreted as a relatively warm temperature in terms of general temperature. While this range allowed us to effectively elicit both psychological and physiological responses to the temperature conditions, it might not be strong enough to turn temperature into a salient stressor or trigger that could induce an effect on moral judgment. Thus, our findings suggest that the onset of psychological and physiological signatures of temperature does not co-occur with influences on moral judgment. Awareness of cold does not lead to a change in moral judgment. However, it is still possible that influences in the moral dilemma task arise outside the range of 21°C and 27°C, when temperature works as a more salient stressor. Especially, temperatures of less than 21°C should be examined to inspect the relationship between more salient cold temperature and moral judgment.
Hancock et al. (2007) suggested an inverse U-shaped relationship between the effect size and temperature intensity. The effects would be relatively weak in the comfort zone and rapidly become stronger outside this zone. Yeganeh et al. (2018) indicated that the direction of the effect becomes more stable and stronger as the temperature difference increases. From this perspective, the question remains open how an extreme cold temperature would affect performance in the moral dilemma task.
As a limitation of the present experimental procedures, we note that we conducted the manipulation checks several times in each experiment. Moreover, the participants were informed during the initial briefing toward obtaining informed consent that the study related to temperature. One interpretation of the present lack of effects from temperature, then, could be that our participants were on their guard and therefore less susceptible to any effects from temperature on moral judgment. Future studies should consider using deception, as employed by Nakamura et al. (2014), in order to examine how the awareness of temperature may modulate any effect on moral judgment.
The process of moral judgment in moral dilemma situations is explained from dual-process theory (Greene, 2007; Greene, 2009). In this theory, the decision in dilemma could be predicted according to whether automatic emotion or cognitive control predominates. Studies of moral dilemma revealed that manipulations that induce negative emotions like stress lead to the dominance of automatic emotion processing, and this would lead to suppressing utilitarian judgment (Starcke et al., 2012; Youssef et al., 2012). In our study, the cool conditions consistently elicited unpleasant emotions. Nevertheless, to the extent one might discern an effect of cool temperature in certain conditions (our Experiment 1 and the work by Nakamura et al., 2014), the tendency would be for cold to promote utilitarian judgment.
On the other hand, it should be noted that the moral dilemma task involves just one type of moral judgment and arguably a rather unusual case of decision-making in which participants are faced with a choice of life or death for multiple people. In particular, the option to save more people by sacrificing one victim in the moral dilemma task is called utilitarian judgment; however, this does not accurately reflect utilitarian thought in the strict sense. Specifically, it was pointed out that the “the greater good” aspect of the genuine idea of utilitarianism may not be reflected in the tendency to answer utilitarian judgments in the moral dilemma task (Kahane et al., 2015; Crone and Laham, 2017). Two separable dimensions have been identified regarding utilitarian thought in moral psychology (Kahane et al., 2018). One dimension reflects the essence of utilitarianism with impartial concern for “the greater good,” and the other dimension involves permissiveness toward instrumental harm. Strictly speaking, the moral judgments measured in this study may not have reflected a utilitarian tendency, but the acceptability of actively sacrificing victims to save others.