Participants showed greater concern for pain in close others than for their own pain, though this hyperaltruism was steeply discounted with increasing social distance

Social discounting of pain. Giles W. Story  Zeb Kurth‐Nelson  Molly Crockett  Ivo Vlaev  Ara Darzi  Raymond J. Dolan. Journal of the Experimental Analysis of Behavior, October 7 2020. https://doi.org/10.1002/jeab.631

Abstract: Impatience can be formalized as a delay discount rate, describing how the subjective value of reward decreases as it is delayed. By analogy, selfishness can be formalized as a social discount rate, representing how the subjective value of rewarding another person decreases with increasing social distance. Delay and social discount rates for reward are correlated across individuals. However no previous work has examined whether this relationship also holds for aversive outcomes. Neither has previous work described a functional form for social discounting of pain in humans. This is a pertinent question, since preferences over aversive outcomes formally diverge from those for reward. We addressed this issue in an experiment in which healthy adult participants (N = 67) chose the timing and intensity of hypothetical pain for themselves and others. In keeping with previous studies, participants showed a strong preference for immediate over delayed pain. Participants showed greater concern for pain in close others than for their own pain, though this hyperaltruism was steeply discounted with increasing social distance. Impatience for pain and social discounting of pain were weakly correlated across individuals. Our results extend a link between impatience and selfishness to the aversive domain.

Discussion

Here we examined for the first time the relationship between the evaluation of one's own future pain and a sensitivity to pain in others, and whether altruistic responses to another's pain depend on the social distance of the other person. We find support for two novel findings. Firstly, people show greater concern for pain in close others than for their own pain, though this hyperaltruism is steeply discounted (diminishes) with increasing social distance. Secondly, we find a correlation between dread and social discounting, such that people who more strongly prefer immediate pain show steeper social discounting of pain, and thereby tend to be less altruistic overall. In keeping with previous findings, participants chose to speed up the delivery of pain both for themselves or others, even if this entailed an increased intensity of the pain, consistent with an effect of dread (Badia et al., 1966; Berns et al., 2006; Cook & Barnes, 1964; Hare, 1966a; Loewenstein, 1987; Story et al., 2013).

Social Discounting of Pain versus Money

Social discounting is consistent with evolutionary notions of kin altruism, which proposes that altruism towards related others carries an evolutionary advantage (Curry et al., 2013; Madsen et al., 2007; Schaub, 1996). Our finding of social discounting for pain extends previous findings of hyperaltruism towards close others for money (Rachlin & Jones, 2008), whereby some people prefer to assign a hypothetical monetary reward ($75) to their closest friend or relative (Person #1) than to receive a larger sum themselves (e.g. $80). Rachlin and Jones (2008) note that hyperaltruistic behavior is irrational in the monetary context, since participants could take the $80 for themselves and give it to Person #1. The same authors speculated that, in addition to wishing to signal their closeness to Person #1, people may have chosen the hyper‐generous option due to an implicit cost of having to transfer money, or as a self‐control device to prevent them from keeping the money for themselves. That we find hyperaltruism for close others for painful outcomes, which are nontransferrable, supports a more intrinsic charitable motive, in keeping with kin altruism.

We show support for a model of social discounting in which the net degree of altruistic behavior depends on both the degree of discounting over social distance (Ksoc) and an additional ‘altruism factor’ (θ) that is independent of social distance. Those with a high altruism factor and low social discounting (high θ, low Ksoc) would be expected to show charitable or caring behavior even towards distant others, for instance victims of war or famine in other countries. By contrast, those with a high altruism factor but steep social discounting (high θ, high Ksoc) would be expected to be protective of close kin, but to engage in little altruistic behavior directed outside of their social circle. These categories appear to have high face validity. A future line of investigation might be to compare these parameters for pain with those for money. Existing studies directly comparing generosity for pain and money demonstrate more charitable behavior with painful outcomes (Davis et al., 2011; Story et al., 2015), however to our knowledge no studies have examined this across social distance to test whether the effects are attributable to higher θ or lower Ksoc.

Applied Social Discounting of Pain

Further research is also required to establish how social discounting of pain relates to real‐world behavior, either charitable or antisocial. Existing work has linked social discounting of money to a range of real‐world behavior. A recent study has demonstrated lower social discounting of reward in extraordinarily altruistic people who have donated a kidney to a stranger (Vekaria et al., 2017), while steeper social discounting has been demonstrated among boys with externalizing (antisocial) behavioral problems (Sharp et al., 2011). Further applied work in this vein might also examine aversive, as well as monetary, outcomes. The current study illustrates that such preferences can be readily elicited using hypothetical painful scenarios.

Other authors have examined the effect of state‐based changes on the social discount curve for reward. Some such models have also examined the effects on the numerator term in the social discount model, namely, θ. For example, Wu et al. (2019) showed that testosterone administration in males increased social discounting for distant others, but had no effect on generosity towards close others. Strikingly, Margittai et al. (2015) showed that experimentally induced psychosocial stress appeared to have the reverse effect. Stress increased the numerator term, but had no effect on the social discount factor, manifest as greater generosity towards close, but not distant, others; a follow on study (Margittai et al., 2018) demonstrated that oral administration of hydrocortisone had the same effect. Further work is needed to investigate influences on the numerator term, in particular to disentangle effects of the instantaneous utility term from the effect of θ, since these enter multiplicatively into the numerator. Painful stimuli, which allow the form of instantaneous utility to be elicited directly using willingness to pay, offer a route to achieving this.

Positive Correlation between Dread and Social Discounting of Pain

Previous work suggests that the ability to wait for future rewards and the ability to understand the mental states of others are linked. For instance, temporal discounting for reward and altruistic behavior have been shown to be correlated across individuals (Curry et al., 2008; Rachlin & Jones, 2008), and both are impaired in Borderline Personality Disorder (Bateman & Fonagy, 2004). Along these lines a tendency to expedite pain so as to mitigate dread might be conceptualized as a future‐oriented behavior, akin to showing altruism towards one's future self. Indeed, both dread and altruism for pain have been shown to relate to the strength of physiological response to imagined pain: People who show greater anticipatory brain responses to pain are more likely to expedite pain rather than delay it (Berns et al., 2006), and people who show greater skin conductance responses to pain in others are more likely to choose to relieve another's pain (Hein et al., 2011). In keeping with this idea, people with higher trait psychopathy have been shown to be less likely to choose to expedite their own impending pain (Hare, 1966b) and show diminished physiological responses to the anticipation of pain in others (Caes et al., 2012). By this reasoning dread might be associated with lower social discounting of pain. Strikingly however, and contrary to our prediction, we found evidence that ‘higher dreaders’ showed steeper social discounting for pain.

Our data do not permit firm conclusions regarding the reasons for this correlation. However, a possible interpretation is that choice of sooner pain represents more a generic form of impatience than previously thought. We found that preference for sooner pain was best accounted for in terms of waiting cost that scaled with delay, but not with pain intensity. This finding is difficult to reconcile with previous models of dread, which focus on the aversive anticipation of pain, a quantity that would be expected to scale with pain intensity. Imagine for instance that you are contemplating either a trivially painful routine dental check‐up or a considerably more painful dental procedure. Our results suggest that the overall disvalue of the very painful procedure would still be greater than the routine check‐up but that the effect of delay on the disvalue of each would be identical.

The superior fit of a nonscaled model suggests that choices to expedite pain might not solely result from a desire to minimize the anticipation of pain, so much as a desire to reduce a generic cost associated with waiting. Notably reframing dread as impatience does not require any change to the form of our model, since the model does not specify the processes underlying a tendency to expedite pain. It is possible that a similar impatience term also contributes to discounting of reward (see for example Gonçalves & Silva, 2015). Such a reframing would make the observed correlation between impatience for pain and social discounting congruent with the correlation between delay discounting and social discounting seen for reward. There follows a strong prediction that impatience for pain and for reward ought to be correlated, indicating an important direction for further research.

Interactions between Dread and Social Discounting of Pain

A further interesting direction for future work concerns how delay and social discounting of pain interact. A pertinent question, for example, is whether effects of dread and social discounting are multiplicative or whether dread is revealed differently when choosing for others. Here we found that a preference for sooner pain was equivalent whether participants chose regarding their own pain, or that of another person at social distance #50. Notably however, in social discounting choices the mean participant showed neither marked social discounting nor hyperaltruism for a person at social distance #50, therefore further exploration is required to establish whether dread interacts with social discounting effects across a range of social distances. We have examined this in an additional study, submitted to this journal, in which we also elicit choices across both domains, for example pain for oneself now, versus for another person in the future, and vice versa (Story et al., 2020).

Factors in the Valuation of Future Pain

The model described here is challenged to disentangle the effects of discounting and dread within a given individual. We are grateful to a reviewer for the suggestion that measuring temporal preferences for past as well as future pain might offer a means to parse the two effects. Prior research has shown temporal discounting of past events to be lawful and also hyperbolic in form (e.g., Yi et al., 2006). Since dread presumably is not contained within events in the past, measuring discounting of past painful events could help to isolate the contribution of dread.

Finally, there are plausible reasons why choices to expedite pain might depend on factors other than dread of pain. Firstly, in many real‐world situations people choose to endure pain or discomfort so as to obtain an associated reward, for example having an immunization to prevent the possibility of illness, doing exercise to improve overall wellbeing, or working to earn a wage. If the rewards accrue at approximately the same time as the pain and outweigh its disvalue, then discounting of the net benefit could motivate speeding‐up the pain–reward combination. Secondly, it is often the case that painful experiences tend to get worse over time, making it rational to face them sooner: For instance in the real world the timing of a dental appointment might be brought forward to relieve worsening dental pain. Although our scenarios attempt to control for these factors, these prior assumptions may nevertheless influence people's experimental choices. Further experimental work is required to disentangle these possibilites.