Making inequality unacceptable: an experiment on fiction reading and prosociality

The public for the pilot is only college students at Universidad de los Andes.

Both the first and second intervention will be done at Universidad de los Andes. The general public intervention will be considered a scaling of the curriculum.

The public for the general intervention are students on their last two years of high-school, early college students at large and members of the general public who frequently visit the public libraries in Bogotá.

2025-04-24

2026-10-31

We want to measure all dimensions of prosociality through economic games.
We will use the following canonical games:
- Dictator game
- Ultimatum game
- Trust game
- Third party game
Instruction for participants on each of these games is provided at the Materials sections.

We will also measure psychological traits with canonical tests in the psychology literature. These include:
- Inclusion of the Other in the Self (IOS) scale.
- Reading the Mind in the Eyes test (RMET). Shortened version.
- Narrative engagement scale
- ⁠UCLA Loneliness scale
- Self-importance of morality scale
- ⁠Attributional complexity scale

We have also created an questionaire that tries to measure a possible mechanism for our result: the value given to the aesthetic experience of the other. We have called this questionaire the Perceived Uniqueness Scale.

For our second intervention (Fall 2026) we will widen the set of psychological scales we will implement. In particular we will use a Perceived Uniqueness Scale and an individuation incentive compatible task we developed. This 10-question scale is designed to measure the degree to wich others are perceived as unique individuals. Statistical analysis has shown that it can be disentangled into two different indexes: Uniqueness Orientation and Attention to Individual Differences. We will use those Indexes in all our causal anaylisis.

We understand prosociality as the share of the personal income that an individual is willing to share with others. This share is different depending on each situation, and that is why we use different economic games. In any case, this is a more objective measure than the ones we find in the existing literature.

Will we think there are many things to improve on the existing literature, we still want to be able to have a dialogue with existing papers. This also allows us to valdiate our findings with economic games as the result of real changes in preferences.

Additionally, we will measure empirical and normative expectations regarding the behavior in those games. These are all outcomes that the intervention can change.

For our second intervention we will also measure as a secondary outcome the individual variance in some of the scales we measure. This is important in as much as what we are modeling is fiction reading as a mean-preserving spread in other-regarding concerns. Scales like the Reading the Mind in the Eyes test (RMET) are designed with the average score as measure, but we will use also the individual variation in responses as a relevant outcome. The same for our Perceived Uniqueness Scale, and its two indexes. The individual variance on these helps us validate the mechanisms we have in mind. We have additionally prepared and incentive-compatible individuation scale that measures both the self-reported motives regarding two actions, and what they perceived others' motives are. We incentivize the prediction of others about what they claimed other would have predicted other motives are, and also, a willingness to pay for knowing what others'motive actually are.

We want to test wether our results are caused by a change on individual preferences and decisions or if they are caused by a change in the empirical or normative expectations about which are the prevailing social norms.
We want to also test if the mean-preserving spread in other-regarding concerns comes from changes in individual preferences or changes in the perception of prevailing social norms. For this we will add incentivized questions on second order beliefs around the importance of perceiving others as unique, as measured by our Perceived Uniqueness Scale and individuation task. We not only expect a larger tendency, for those in the treatment cell, to perceive others as unique, rather than an interchangable abstract, but also, to redeuce the actor-observer gap and to increase the richness of other people's motives

There are two experimental sessions in which economic games are played, and psychological scales are discussed.

The first one occurs on the day of the last session of each course. This is our measure of short term effects.

The second one is performed 4 weeks from the last session of the intervention course. This is our measure of longer term effects.

Not available

Randomization done in the computter. Each participant is randomly allocated to treatment or control.

For the pilot we randomize at the individual level.
We will randomize at the participant level. Depending on the number of participants willing to be included, we will also select some participants directly to the "pure control" arm of our Second Intervention.

No

For the pilot we have 60 enrolled students (30 in each course).
The effective number of participants we have after our first intervenion is N = 227. We aspire to have a similar number for the second intervention that takes place in the fall of 2026.

For the pilot we have 60 enrolled students (30 in each course). The effective number of participants we have after our first intervenion is N = 227. We aspire to have a similar number for the second intervention that takes place in the fall of 2026.

30 students in control and treatment arm for the first instance of the pilot.
100 in each arm during implmentation at Universidad de los Andes.

We will use the dictator game for this calculation. Assuming that on average people send 25 with std dev 14, and we want to detect a change of 7 in the fiction reading group) with an type I-error of 5% and a 1-typeII error of 80, we need a sample of at least 63 in each treatment condition: that is, 126 in total.