Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Preferences as Heuristics

Mohlin, Erik LU and Rigos, Alexandros LU (2026) In Working Papers
Abstract
We present a model of how an individual’s preference for a specific behaviour can
evolve as an adaptation to her social environment. Our key assumptions are that the
decision maker (DM) (i) is boundedly rational in her ability to process information and
evaluate which actions yield the highest payoff and (ii) can develop a subjective prefer-
ence for actions. A stronger preference for an action reduces the probability of not tak-
ing it when it is optimal (omission error), but it also increases the probability of taking
the action when it is suboptimal (commission error). The optimally adapted preference
strikes a balance between these errors and depends on the DM’s environment. DMs are
typically better off... (More)
We present a model of how an individual’s preference for a specific behaviour can
evolve as an adaptation to her social environment. Our key assumptions are that the
decision maker (DM) (i) is boundedly rational in her ability to process information and
evaluate which actions yield the highest payoff and (ii) can develop a subjective prefer-
ence for actions. A stronger preference for an action reduces the probability of not tak-
ing it when it is optimal (omission error), but it also increases the probability of taking
the action when it is suboptimal (commission error). The optimally adapted preference
strikes a balance between these errors and depends on the DM’s environment. DMs are
typically better off (in objective terms) if they are endowed with subjective preferences
that deviate from maximisation of expected objective payoff. Importantly, if DMs can
evaluate actions perfectly, they do not develop such biases. The results extend in an
intuitive manner to n-player, two-strategy supermodular games. Our framework can be
used to interpret preferences for specific actions or strategies (e.g., conditional cooper-
ation, truth-telling, or norm-following) as psychological and cultural consequences of
material incentives and social organisation. (Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Working paper/Preprint
publication status
published
subject
keywords
Preference evolution, Bounded rationality, Culture, Cooperation, Cogni- tive economics, Logit choice, Monotone comparative statics, Supermodular games, C72, C73, D01, D91
in
Working Papers
issue
2026:8
pages
57 pages
language
English
LU publication?
yes
id
5e2789d4-c54c-4456-ae22-c4b0206e1842
date added to LUP
2026-09-09 12:54:33
date last changed
2026-09-09 12:54:33
@misc{5e2789d4-c54c-4456-ae22-c4b0206e1842,
  abstract     = {{We present a model of how an individual’s preference for a specific behaviour can<br/>evolve as an adaptation to her social environment. Our key assumptions are that the<br/>decision maker (DM) (i) is boundedly rational in her ability to process information and<br/>evaluate which actions yield the highest payoff and (ii) can develop a subjective prefer-<br/>ence for actions. A stronger preference for an action reduces the probability of not tak-<br/>ing it when it is optimal (omission error), but it also increases the probability of taking<br/>the action when it is suboptimal (commission error). The optimally adapted preference<br/>strikes a balance between these errors and depends on the DM’s environment. DMs are<br/>typically better off (in objective terms) if they are endowed with subjective preferences<br/>that deviate from maximisation of expected objective payoff. Importantly, if DMs can<br/>evaluate actions perfectly, they do not develop such biases. The results extend in an<br/>intuitive manner to n-player, two-strategy supermodular games. Our framework can be<br/>used to interpret preferences for specific actions or strategies (e.g., conditional cooper-<br/>ation, truth-telling, or norm-following) as psychological and cultural consequences of<br/>material incentives and social organisation.}},
  author       = {{Mohlin, Erik and Rigos, Alexandros}},
  keywords     = {{Preference evolution; Bounded rationality; Culture; Cooperation; Cogni- tive economics; Logit choice; Monotone comparative statics; Supermodular games; C72; C73; D01; D91}},
  language     = {{eng}},
  note         = {{Working Paper}},
  number       = {{2026:8}},
  series       = {{Working Papers}},
  title        = {{Preferences as Heuristics}},
  url          = {{https://lup.lub.lu.se/search/files/260381868/WP26_8_cover_combined.pdf}},
  year         = {{2026}},
}