| Abstract: |
Information systems (IS) research that examines individual user-system
interaction naturally relies on human cognition, judgment, and
decision-making. Accordingly, this branch of IS research has employed
psychological theories since its inception. However, only in recent years have
IS researchers started to investigate how insights from cognitive psychology,
social psychology, and behavioral economics can be leveraged within IS design.
Especially the utilization of cognitive biases in combination with
gamification currently attracts a lot of research attention. Both concepts aim
at designing IS in a way that motivates users towards a target behavior.
Despite calls for IS research on gamification to leverage insights from
behavioral economics and social psychology in the context of gam(bl)ified IS
(i.e., IS that use game/gambling design elements for non-entertainment
purposes), there is only sparse research investigating how information cues
can address cognitive biases to affect outcomes within gam(bl)ified IS. To
advance the emergent research in these connected fields, this thesis
investigates how information cues embedded in user-interface design elements
can account for two important categories of cognitive biases regarding their
influence on user behavior. More specifically, this thesis examines how
information cues that address social influence biases and risk assessment
biases impact user conversion (i.e., the process of turning visitors into
active and/or paying users) within gam(bl)ified IS. Against this backdrop,
five studies were conducted and published across four articles. The first
study lays the foundation for the research context and setting. More
specifically, a taxonomy of gamification elements is developed and extended to
account for the novel gamification-related concept of gamblification. While
gamification is often defined as the use of game design elements in a
non-gaming context, gamblification can be viewed as the use of gambling design
elements in non-gambling contexts. Gamblification is a unique concept that can
be distinguished from gamification because it additionally relies on
chance-based uncertainty and user-system resource transfer. Both gamification
and gamblification provide a suitable setting for examining how cognitive
biases can be accounted for in regard of their impact on user behavior. This
is because gamification is often concerned with social interactions (e.g.,
competition or cooperation) and therefore frequently relies on social
psychology as a theoretical foundation. Gamification is thus predestined to
investigate the role of social influence biases (i.e., cognitive biases based
on social influence theory) in motivating users towards a target behavior.
Likewise, the emergent gamblification research setting is intriguing for
investigating cognitive biases. Due to the inherent characteristics of
gamblification (i.e., chance-based uncertainty and resource transfer)
particularly risk assessment biases (i.e., cognitive biases that are induced
by biased risk assessment) are highly relevant in this environment. The second
article is situated in a gamified IS environment and investigates the role of
social influence biases in fostering user conversion behavior (i.e., user
registration) on an e-commerce platform. Drawing on social influence theory,
two social influence cues (i.e., information cues that account for social
influence biases), namely reciprocity cues and social proof cues, are embedded
in the gamification design element ‘user onboarding’ and investigated in an
e-commerce setting. The article’s findings indicate that both, reciprocity and
social proof, have positive direct effects on user registration. However,
depending on how reciprocity is implemented, the interactive effect of
reciprocity and social proof attenuates or amplifies the positive direct
effect. The third article examines risk assessment cues (i.e., information
cues that aim to account for risk assessment biases) positioned within the
gamblification element of a ‘loot box menu’. By illuminating how these risk
assessment cues can address a group of risk assessment biases, namely
probability evaluation biases (i.e., cognitive biases that result from a
skewed evaluation of probabilities during risk assessment) their role in
affecting product selection in a digital gaming context is investigated. More
specifically, drawing on prospect theory, information cues are designed to
account for two different probability evaluation biases, the certainty effect
and the availability bias. The study’s results demonstrate that offering users
loot box menus with two different probabilities of winning a reward (i.e., the
choice between two different uncertain rewards vs. the choice between a
certain and an uncertain reward) may trigger the certainty effect which
influences user conversion behavior (i.e., product selection). Moreover,
combining these different loot box menu designs with a previous loss
experience causes users to be subject to the availability bias and amplifies
the certainty effect. In contrast, when users are subject to the optimism
bias, the certainty effect is attenuated. The fourth and last article
investigates how risk assessment cues embedded in differently designed product
offerings (i.e., gamblified vs. transaction-based) in the context of a digital
gaming service can address a different group of risk assessment biases (i.e.,
stability biases) and thereby influence user conversion behavior (i.e., user
purchase behavior). More specifically, drawing on literature on
decision-making under risk, information cues are designed to address two
stability biases, namely the risk avoidance effect and the endowment effect.
Moreover, the effect of a general personal trait that is relevant for risk
assessment (i.e., risk aversion) is examined. The results reveal that a
gamblified product offering (i.e., a loot box with an uncertain reward) vs. a
transaction-based product offering (i.e., a loot box with a certain reward)
triggers the risk avoidance effect that influences user conversion. This
effect is amplified when users are either subject to the endowment effect
(i.e., when they experience a previous endowment with a loot box) or when they
are risk-averse. Taken together, this thesis highlights the importance of
considering the role of social influence and risk assessment biases and how
information cues can be designed within gam(bl)ified IS to address these
cognitive biases to motivate users towards a target behavior. Moreover, the
results contribute to IS research by exploring the impact of manifold
information cues that account for social influence and risk assessment biases
as well as context-dependent interacting variables in various gam(bl)ified IS
environments. Alongside these contributions to research, this thesis provides
several interesting and actionable recommendations on how to implement
information cues that take cognitive biases into account. This is primarily
aimed at practitioners tasked with designing gam(bl)ified IS with the goal of
optimizing user conversion behavior. |