Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Choice blindness and the (in)transitivity of preferences: A graph-based method for studying preference structures in real-time

Haglund, Erik LU (2021) KOGM20 20211
Cognitive Science
Abstract
The axiom of transitivity states that if a person prefers option
A to B and B to C, he must also prefer A to C. This simple
axiom is often seen as one of the pillars of rational choice,
but whether it should be has created a divide in the research
community. Some state that there cannot be rational choice
or utility without transitivity and that agents who present
intransitive preferences allow themself to be exploited. In
contrast, others hold the opposite position, that intransitive
preferences not only occur but can, in fact, be rational. This
paper aims to study whether intransitive choices occur at
such a high degree as is suggested in the literature and
whether intransitive choices truly reflect intransitive
preferences.... (More)
The axiom of transitivity states that if a person prefers option
A to B and B to C, he must also prefer A to C. This simple
axiom is often seen as one of the pillars of rational choice,
but whether it should be has created a divide in the research
community. Some state that there cannot be rational choice
or utility without transitivity and that agents who present
intransitive preferences allow themself to be exploited. In
contrast, others hold the opposite position, that intransitive
preferences not only occur but can, in fact, be rational. This
paper aims to study whether intransitive choices occur at
such a high degree as is suggested in the literature and
whether intransitive choices truly reflect intransitive
preferences. Taking inspiration from seminal work in
decision-making theory and behavioral economics, an
experimental method was developed to measure preference
transitivity and preference change over time. In this study,
two experiments were conducted via a two-alternative
forced-choice (2AFC) task in which participants selected the
face they found more attractive. During Experiment 1,
intransitive choices could be detected and extracted in
real-time while dynamically utilizing choice blindness to
study its possible impact on intransitive preferences. During
Experiment 2, all stimuli combinations were iterated thrice,
which allowed the preference graphs to be analyzed in their
entirety. To evaluate the intransitive preferences in relation to
those with a transitive pattern, we measured preference
strength, choice consistency, and how the preferences evolved
over time. Our results indicate that although very few
indicators of intransitive choice were found, there was a
significant decrease in intransitivity over time. Almost no
evidence of repeated intransitive choices was found
throughout all experimental phases, and even if we found a
moderate level of choice blindness, this did not lead to
intransitivity in subsequent choices. However, other possibly
exploitable preference structures were found (Less)
Please use this url to cite or link to this publication:
author
Haglund, Erik LU
supervisor
organization
alternative title
Beslutsblindhet och preferensers (in)transitivitet: En grafteoretisk metod för att studera preferensstrukturer i realtid
course
KOGM20 20211
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Cognitive science, Behavioural economics, Choice blindness, money pumps, preferences, cognition, graphs, graph theory, intransitive preferences, transitive, decision-making, transitivity, cyclic preferences
language
English
id
9068149
date added to LUP
2021-12-13 13:02:26
date last changed
2021-12-13 13:02:26
@misc{9068149,
  abstract     = {{The axiom of transitivity states that if a person prefers option
A to B and B to C, he must also prefer A to C. This simple
axiom is often seen as one of the pillars of rational choice,
but whether it should be has created a divide in the research
community. Some state that there cannot be rational choice
or utility without transitivity and that agents who present
intransitive preferences allow themself to be exploited. In
contrast, others hold the opposite position, that intransitive
preferences not only occur but can, in fact, be rational. This
paper aims to study whether intransitive choices occur at
such a high degree as is suggested in the literature and
whether intransitive choices truly reflect intransitive
preferences. Taking inspiration from seminal work in
decision-making theory and behavioral economics, an
experimental method was developed to measure preference
transitivity and preference change over time. In this study,
two experiments were conducted via a two-alternative
forced-choice (2AFC) task in which participants selected the
face they found more attractive. During Experiment 1,
intransitive choices could be detected and extracted in
real-time while dynamically utilizing choice blindness to
study its possible impact on intransitive preferences. During
Experiment 2, all stimuli combinations were iterated thrice,
which allowed the preference graphs to be analyzed in their
entirety. To evaluate the intransitive preferences in relation to
those with a transitive pattern, we measured preference
strength, choice consistency, and how the preferences evolved
over time. Our results indicate that although very few
indicators of intransitive choice were found, there was a
significant decrease in intransitivity over time. Almost no
evidence of repeated intransitive choices was found
throughout all experimental phases, and even if we found a
moderate level of choice blindness, this did not lead to
intransitivity in subsequent choices. However, other possibly
exploitable preference structures were found}},
  author       = {{Haglund, Erik}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Choice blindness and the (in)transitivity of preferences: A graph-based method for studying preference structures in real-time}},
  year         = {{2021}},
}