In his target article, Bermúdez makes a case for “rational framing effects,” according to which quasi-cyclical preferences – that violate transitivity – can be normatively correct and appropriate. Here, we (1) argue that a reasonable interpretation of the type of preferences that Bermúdez considers does not imply quasi-cyclical preferences at all, but in fact reflect ordinary economic trade-offs, and (2) refer to previous work to demonstrate framing effects that are not consistent with such trade-offs in the context of inter-group conflict.
Bermúdez treats preferences over actions, outcomes, and outcome categories interchangeably. While a statement such as “Macbeth would like to bravely take the throne” reasonably carries an implicit ceteris paribus (“all other things being equal”), Bermúdez treats it as a coarse partition of the outcome space, in which Macbeth prefers any outcome that involves being the king to any outcome that does not. Quasi-cyclical preferences arise when considering a choice between actions for which different such partitions contrast. In the action space that Macbeth faces, the outcome that involves becoming the king necessarily involves breaching his double duty to Duncan. The cycle disappears when viewing Macbeth's preferences as ceteris paribus rather than categorical preferences.
To illustrate, consider the following statements, which are structurally equivalent to statements (C) and (D) that appear in the target article regarding Macbeth:
(C*) Jose prefers having more money to having less money.
(D*) Jose prefers eating over remaining hungry.
Should Jose buy lunch? Since “having more money,” in this context, means “remaining hungry,” and “having less money” is the same as “eating,” this situation reflects, according to Bermúdez, a choice between a “money frame” and a “food frame.” But what Jose is facing is, in fact, a standard economic decision. Statements C* and D* can be held concurrently as ceteris paribus statements without implying quasi-cyclical preferences. The decision (to buy lunch or not to buy?) does not reflect a choice between frames but results from resolving a trade-off. Such trade-offs can be decomposed in a counterfactual action space. Macbeth would have been happy to become the king without murdering Duncan, just as Jose would be happy with a free lunch. Similarly, it is possible to decompose the timing and reward in self-control problems or the effects on the self and others in the social dilemma examples analyzed in the target article. Indeed, “it is rational to have a complex and multi-faceted response to a complex and multi-faceted situation” (H3; target article, sect. 3.2, para. 6). However, the natural conclusion is not that framing effects are rational but that a complex rational response considers all aspects of the outcome to reach a consistent decision in each given choice task.
Framing can, nonetheless, influence how single-choice tasks are perceived. What does fulfilling Macbeth's double duty to Duncan mean? When viewed from different perspectives, fulfilling Macbeth's double duty can at times mean diverse and even opposing things; the frame can change the perspective. We found evidence for such an effect in the context of inter-group conflict and group identity.
We used the inter-group prisoner's dilemma (Bornstein, Reference Bornstein2003) to examine how framing affects the willingness of individual group members to contribute to their group in the social dilemma that arises in inter-group conflict. The inter-group prisoner's dilemma models conflict between two groups. Each group member chooses between keeping resources for themselves and contributing to their group at a personal cost. Contributions benefit the in-group and simultaneously harm the out-group. In other words, cooperation in one group poses a threat to the other.
We used framing to manipulate the target of the threat, which was either the group as a whole or the individual group members (the two frames are identical in terms of the objective underlying strategic situations, i.e., the mapping from actions to payoffs). These frames lead to opposing reactions, increasing cooperation in the group frame but decreasing cooperation in the individual frame. The unifying principle, in this case, is “help those under threat” (Weisel & Zultan, Reference Weisel and Zultan2016, Reference Weisel and Zultan2021a). In the group frame the perception is that the group is under threat, so group members help the group by increasing their contributions. In the individual frame group members perceive themselves – as individuals – to be under threat, so they help themselves by keeping resources for private use. Thus, the frame does not increase the salience of a particular outcome aspect associated with the action. Instead, framing leads to a reinterpretation of the same aspect.
One may argue that this is also a case of a simple trade-off that can be eliminated by decomposing the outcome, similar to the “I” versus “we” frames analyzed by Bacharach (Reference Bacharach2006). Weisel and Zultan (Reference Weisel and Zultan2021b) tested such decomposition by allowing players to choose whether cooperation harms the out-group or not, keeping the effects of cooperation on the self and the in-group constant across the two options. This decomposition should eliminate the framing effect if frames affect the different narratives' relative salience, such as “cooperation” or “conflict.” However, the framing effect remained strong in the new game: Group members were more likely to harm the out-group under a group frame than under an individual frame.
You have
Access
In his target article, Bermúdez makes a case for “rational framing effects,” according to which quasi-cyclical preferences – that violate transitivity – can be normatively correct and appropriate. Here, we (1) argue that a reasonable interpretation of the type of preferences that Bermúdez considers does not imply quasi-cyclical preferences at all, but in fact reflect ordinary economic trade-offs, and (2) refer to previous work to demonstrate framing effects that are not consistent with such trade-offs in the context of inter-group conflict.
Bermúdez treats preferences over actions, outcomes, and outcome categories interchangeably. While a statement such as “Macbeth would like to bravely take the throne” reasonably carries an implicit ceteris paribus (“all other things being equal”), Bermúdez treats it as a coarse partition of the outcome space, in which Macbeth prefers any outcome that involves being the king to any outcome that does not. Quasi-cyclical preferences arise when considering a choice between actions for which different such partitions contrast. In the action space that Macbeth faces, the outcome that involves becoming the king necessarily involves breaching his double duty to Duncan. The cycle disappears when viewing Macbeth's preferences as ceteris paribus rather than categorical preferences.
To illustrate, consider the following statements, which are structurally equivalent to statements (C) and (D) that appear in the target article regarding Macbeth:
(C*) Jose prefers having more money to having less money.
(D*) Jose prefers eating over remaining hungry.
Should Jose buy lunch? Since “having more money,” in this context, means “remaining hungry,” and “having less money” is the same as “eating,” this situation reflects, according to Bermúdez, a choice between a “money frame” and a “food frame.” But what Jose is facing is, in fact, a standard economic decision. Statements C* and D* can be held concurrently as ceteris paribus statements without implying quasi-cyclical preferences. The decision (to buy lunch or not to buy?) does not reflect a choice between frames but results from resolving a trade-off. Such trade-offs can be decomposed in a counterfactual action space. Macbeth would have been happy to become the king without murdering Duncan, just as Jose would be happy with a free lunch. Similarly, it is possible to decompose the timing and reward in self-control problems or the effects on the self and others in the social dilemma examples analyzed in the target article. Indeed, “it is rational to have a complex and multi-faceted response to a complex and multi-faceted situation” (H3; target article, sect. 3.2, para. 6). However, the natural conclusion is not that framing effects are rational but that a complex rational response considers all aspects of the outcome to reach a consistent decision in each given choice task.
Framing can, nonetheless, influence how single-choice tasks are perceived. What does fulfilling Macbeth's double duty to Duncan mean? When viewed from different perspectives, fulfilling Macbeth's double duty can at times mean diverse and even opposing things; the frame can change the perspective. We found evidence for such an effect in the context of inter-group conflict and group identity.
We used the inter-group prisoner's dilemma (Bornstein, Reference Bornstein2003) to examine how framing affects the willingness of individual group members to contribute to their group in the social dilemma that arises in inter-group conflict. The inter-group prisoner's dilemma models conflict between two groups. Each group member chooses between keeping resources for themselves and contributing to their group at a personal cost. Contributions benefit the in-group and simultaneously harm the out-group. In other words, cooperation in one group poses a threat to the other.
We used framing to manipulate the target of the threat, which was either the group as a whole or the individual group members (the two frames are identical in terms of the objective underlying strategic situations, i.e., the mapping from actions to payoffs). These frames lead to opposing reactions, increasing cooperation in the group frame but decreasing cooperation in the individual frame. The unifying principle, in this case, is “help those under threat” (Weisel & Zultan, Reference Weisel and Zultan2016, Reference Weisel and Zultan2021a). In the group frame the perception is that the group is under threat, so group members help the group by increasing their contributions. In the individual frame group members perceive themselves – as individuals – to be under threat, so they help themselves by keeping resources for private use. Thus, the frame does not increase the salience of a particular outcome aspect associated with the action. Instead, framing leads to a reinterpretation of the same aspect.
One may argue that this is also a case of a simple trade-off that can be eliminated by decomposing the outcome, similar to the “I” versus “we” frames analyzed by Bacharach (Reference Bacharach2006). Weisel and Zultan (Reference Weisel and Zultan2021b) tested such decomposition by allowing players to choose whether cooperation harms the out-group or not, keeping the effects of cooperation on the self and the in-group constant across the two options. This decomposition should eliminate the framing effect if frames affect the different narratives' relative salience, such as “cooperation” or “conflict.” However, the framing effect remained strong in the new game: Group members were more likely to harm the out-group under a group frame than under an individual frame.
Conflict of interest
We declare no conflict of interest.