NOTE-1279: Activating the legacy motive mitigates intergenerational discounting in the climate game

Contribution

It carries the legacy programme of Wade-Benzoni and colleagues from individual allocation to group cooperation. Earlier work had people split a resource between themselves now and someone later. Here six people must together reach a threshold to protect an endowment, and the endowment goes either to themselves tomorrow or to future generations as trees. The paper replicates Jacquet et al.’s (2013) finding that the second version kills cooperation, and shows that short messages built to activate the legacy motive largely restore it.

Key insight

The obstacle to climate cooperation is two tragedies at once: the commons (each gains by free riding) and the horizon (each generation gains by passing costs on). The legacy motive is offered as a bridge across the second. The authors’ account of why it works is self-interest: “the legacy motive is a mechanism for leveraging short-term self-interest for the long-term best interests of the collective” (p. 9).

Assumptions

• Intergenerational discounting = temporal + social discounting. Benefits to others later are discounted for both delay and social distance (p. 1–2, after Wade-Benzoni 2008).
• Three routes to the legacy motive, taken from the programme (p. 2): mortality salience, the power asymmetry between generations, and intergenerational reciprocity (paying forward what a past generation gave).
• The rational benchmark. For a self-interested player who disregards the endowment to future generations, not investing ($40kept)beatsthecooperativeequilibrium($20 kept), so universal defection is predicted in the tree-planting conditions (p. 6).
• Players believe the trees will be planted. Each received a pledge signed by the chief investigator (pp. 5, 8).

Key results

• Design (pp. 3–6). 180 members of the UWA campus community (92 female, 86 male, 2 unspecified; ages 17–57, mean 21), 30 groups of six, 10 groups per condition. $40operatingfund;eachround$0 (“selfish”), $2("fairshare")or$4 (“altruistic”) into the climate account, which paid for a newspaper advertisement on climate protection. Target $120;ifmissed,each$45 endowment lost with probability .9.

  • Short-delay: saved endowments paid in cash the next day.
  • Intergenerational: saved endowments invested in 72 trees (12 per player) storing CO₂ for 80 years.
  • Legacy induction: as intergenerational, with a roughly 250-word passage before rounds 1, 4 and 7, order counterbalanced (Table 1, p. 4): “Leaving a Positive Legacy” (mortality salience), “The Helpless Nature of Future Generations” (power asymmetry), “We Are One With Future Generations” (reciprocity, using Second World War veterans).

• Reaching the target (p. 6). Short-delay 9/10, intergenerational 2/10, legacy induction 7/10. Logistic regression, intergenerational as reference: short-delay β = 3.58, SE = 1.32, z = 2.72, p = .007; legacy induction β = 2.23, SE = 1.04, z = 2.13, p = .033.

• Group investment (p. 6; mean ± SEM). Short-delay $116.80\pm 8.58,intergenerational$77.20 ± 12.01, legacy induction $105.20 ± 9.38. Condition F(2, 27) = 4.07, p = .029; no effect of round, F(9, 243) = 0.35, and no interaction, F(18, 243) = 1.02.

• Strategies (p. 7; counts of the 60 choices per group). Condition × strategy F(4, 81) = 6.551, p < .001.

  • Short-delay: fair share 38.20, selfish 11.70, altruistic 10.10.
  • Intergenerational: selfish 28.90, fair share 23.60, altruistic 7.50.
  • Legacy induction: fair share 32.00, selfish 17.70, altruistic 10.30.

  Legacy induction did not raise altruistic choices. The authors explain this by conditional cooperation and inequity aversion: once everyone is expected to give $2around,nooneneedstogive$4 (pp. 8–9).

• Each message (p. 8, Fig. 5). Investment over the rounds following each message, against the same rounds in the intergenerational game ($23.16\pm 3.60):mortalitysalience$32.00 ± 3.45, power asymmetry $34.00\pm 3.86,reciprocity$39.20 ± 4.56; F(3, 36) = 2.95, p = .044. The messages were not compared with one another.

• Interpretation (p. 9). “People invest in legacies for largely— although not entirely—selfish reasons.” A legacy offers “a little piece of symbolic immortality” and can be thought of as “a reputation- preserving mechanism” for one’s standing after death. What people get by cooperating with the future is “the opportunity to project their personal life meaning and identity into the future”.

• Limitations the authors state (p. 9). No manipulation checks, chosen to avoid demand; no control message without legacy content; the passages mix mechanisms (the reciprocity passage also builds identification with a past generation); a WEIRD sample.

Claims

id	claim	strength	support
C1	When the benefit of cooperation goes to future generations instead of the players, group cooperation in a threshold climate game collapses	moderate: replicates Jacquet et al. 2013; 10 groups per condition	§3.1–3.2
C2	Messages activating the legacy motive restore much of that cooperation	moderate: 7/10 against 2/10 groups, p = .033; no control message, so any persuasive text might do some of it	§3.1–3.2
C3	Each of mortality salience, power asymmetry and intergenerational reciprocity contributes	weak: each beats a baseline built from other groups; messages not compared, no manipulation checks, passages mix mechanisms	§3.4
C4	Legacy induction works by restoring a fair-share norm, not by raising sacrifice beyond it	moderate as a description of the counts; the inequity-aversion explanation is not tested	§3.3, §4
C5	The legacy motive leverages largely selfish motives for others’ benefit	asserted from the literature (Wade-Benzoni 2019; Wade-Benzoni and Tost 2009); nothing here measures motive	§4

Concepts

• legacy motive: “an inherent drive for one’s life to have a meaningful impact that continues in some symbolic form after one’s physical existence has ceased” (p. 2).
• intergenerational discounting: preferring smaller benefits for oneself now to larger benefits for others later.
• tragedy of the horizon: the conflict between the interests of current and future generations (after Carney 2015).
• collective-risk social dilemma: a threshold public-goods game in which missing the group target risks everyone’s endowment.
• fair-share, selfish, altruistic: investing $2,$0 or $4 in a round.

Connections

• Wade-Benzoni et al. 2012 (LIT-1760), unread. Cited once by name among laboratory studies of intergenerational allocation (p. 2). The claim that “implicitly priming them with thoughts of death” mitigates discounting is attributed to “Wade-Benzoni, 2002” (p. 2), which is the reciprocity paper; it appears to be a miscitation of the 2012 paper. See “Read in its place” for what the APS release adds.
• Wade-Benzoni, Sondak and Galinsky 2010 (NOTE-1279, read the same day). Cited as showing that “activating the legacy motive can promote individual action on climate change” (p. 2). That overstates it: the 2010 paper manipulated benefits versus burdens, not legacy, and its climate studies measured attitudes, not action.
• Zaval et al. 2015 (LIT-1724), unread. Cited with the 2010 paper for individual climate action.
• Gardiner’s perfect moral storm (LIT-1681), unread. The game is a laboratory model of the storm’s intergenerational strand: each generation gains by deferring costs to one that cannot retaliate. This is the record’s inference; the paper does not cite Gardiner.
• Terror management (LIT-1687). The mortality salience message draws on it, but here death awareness is meant to trigger legacy building, not worldview defence.

Bearing on the record

• The legacy question (QUESTION-301). These are inferences.

  • The authors take a side among the question’s readings without testing it. Their legacy is a reputation kept after death, which is Pitcher’s reading (LIT-1740), interests that outlast one, as much as the thesis’s extended self. Either way they call it self-interest.
  • The induction cannot separate the readings. It bundles three routes. One, “We Are One With Future Generations”, with its caption “We are one link in a chain connecting us with past and future generations”, is an identification message: a self extended across generations. One, power asymmetry, is a moral message with no self in it. Only the mortality message is about one’s own legacy. The reciprocity message did best ($39.20against$32.00 and $34.00), but the messages were not compared.
  • The legacy offered is collective. Two passages promise being remembered as a generation (“our descendants remember us as a generation that did not shirk their responsibility”). The paper describes no record of who paid for the trees. If the motive worked, it worked through a “we” that is remembered, not through any individual’s name. That is where the question’s temporal and social extensions of the self meet, and it is the same meeting Castano et al. (LIT-1719) propose.
  • QUESTION-301 asks whether legacy priming raises beneficence where no one will know. Here no future person would know who gave, but the messages told players they would be remembered as a generation. So the study does not answer it.

• What the living owe the future (QUESTION-299). The game makes Gardiner’s storm concrete. Without a legacy frame, 8 of 10 groups let a benefit to future people fail rather than give up $20 each. The remedy that worked appeals to what the living value about themselves, which is the nearest ground to Scheffler’s (LIT-1754), not to the good of future people as such. That is an inference from the message content.

• The owner’s thesis (CLAIM-491). The authors’ interpretation is the thesis in time: cooperation with the future as self-interest on a self projected past death. But nothing measures the self, so the paper is evidence that legacy messages work, not that they work through an extended self.

• The triviality objection (CLAIM-492) and Batson’s objection (CLAIM-494). Not addressed.

• A dated note on QUESTION-301 is proposed. Text:

  Hurlstone et al. (NOTE-1279), read in place of Wade-Benzoni et al. 2012, move legacy to group cooperation. In a climate threshold game, 9 of 10 groups protected an endowment paid to themselves, 2 of 10 one paid to future generations as trees, and 7 of 10 when they also read messages on death, on future generations’ powerlessness and on paying forward a past generation’s sacrifice. The authors call legacy a leveraging of self-interest, but the messages bundle an identification with generations (“one link in a chain”), a moral appeal and one’s own legacy, and promise remembrance as a generation, not as a person. So the result fits the thesis’s reading, Pitcher’s and a group self in time alike, and separates none of them.

• No THEORY filed.

• No ML instruction.

Limitations

• Ten groups per condition. The key contrast is 7 of 10 against 2 of 10.
• No control message and no manipulation checks, as the authors say. Reading any earnest text about climate between rounds might raise investment.
• The per-message test. The three message values come from the same ten groups, but F(3, 36) treats the four sets as independent groups. The windows are of unequal length (rounds 1–3, 4–6, 7–10).
• Stakes are modest and the future is near. A $45 endowment converted to trees; the players lose only what they keep.
• Arithmetic and reporting slips.
  • Intergenerational investment is said to be “36% lower” than legacy induction (p. 6); legacy induction is 36% higher, and the intergenerational figure 27% lower.
  • The messages are said to raise investment by 28%, 32% and 41% over baseline (p. 8). Those are the baseline’s shortfall as a share of each message’s figure. Over baseline the rises are 38%, 47% and 69%.
  • The model test is printed χ²(27) = 11.66, p = .003; that p matches 2 degrees of freedom, not 27.
  • Strategy frequencies are printed with dollar signs (“$38.20 ± 3.97”); they are counts.
• Citation errors.
  • The death-priming result is attributed to Wade-Benzoni 2002 (p. 2).
  • The 2010 paper’s title is given as “the intergenerational allocation of benefits and burdens”.
  • “Millinski” for Milinski in Jacquet et al.; a stray “736” in Henrich et al.’s title “Most people are not WEIRD”.
• The supplementary source. The APS release is a press summary with no numbers, and has “thier” for “their”.

Open questions

• Which message works, and through what? A design that gave each message alone, against a neutral persuasive text, with a measure of felt connection to future generations, would say whether identification carries the effect.
• Does a personal legacy (being remembered by name) move people more or less than a collective one (being remembered as a generation)?