University of Southern California Dissertations and Theses

Performance and attention novelty slows hedonic adaptation during habit formation

(USC Thesis Other)

Performance and attention novelty slows hedonic adaptation during habit formation

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT

PERFORMANCE AND ATTENTION NOVELTY SLOWS
HEDONIC ADAPTATION DURING HABIT FORMATION
BY
Lucas M. Carden
______________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

Dissertation Committee
Wendy Wood, PhD, Chair
Norbert Schwarz, PhD
John Monterosso, PhD
Scott Wiltermuth, PhD

December 2019

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT
Acknowledgements
Thank you, Mom and Dad. Thank you for your unconditional love and support during this
journey. I c oul dn’ t have done it without you.
Thank you to my brother, Mike. You helped me maintain the right perspective and always
encouraged me over the years.
Thank you to my advisor, Dr. Wendy Wood. I’l l never forget my first meeting with you many
years ago. I was an aspiring PhD student and you were so generous with your time, giving me
helpful advice and ultimately the opportunity to join your research lab. I’ m grateful that you
challenged me to become a better researcher, writer, and thinker. And, without you, I would have
never had the opportunity to live in Paris, eat multitudes of pastries and baguettes every morning,
and write my dissertation in the cafes of le Marais.
Thank you to my dissertation committee members, Dr. Norbert Schwarz, Dr. John Monterosso,
and Dr. Scott Wiltermuth. The conversations we had changed the way I think about research,
psychology, and life.
Finally, I want to thank all my friends and colleagues that were in my life during graduate
school. This was a formative period in my life and I deeply thank you for the conversations,
laughs, and support.

ii
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT
Abstract
Repetition plays an essential role in pe opl e ’s pursuit of habit formation. Behavior becomes more
automatic with repetition. Yet, repetition has a contrasting effect on our hedonic responses: it
causes hedonic adaptation, the attenuation of our pleasant responses over time. Although the
processes of habit formation and hedonic adaptation have been studied separately in
contemporary psychology for decades, no research has investigated these phenomena jointly.
Hedonic adaptation poses a challenge to habit formation: if people begin repeating an enjoyable
behavior and their enjoyment dissipates too quickly, they may lose the intrinsic motivation to
repeat it, consequently jeopardizing persistence and habit formation. Two experiments (total N =
276) tested an intervention that slows hedonic adaptation during habit formation by introducing
performance and attention novelty during repetition. In two distinct habit formation tasks,
participants who were instructed to modify how they repeated the tasks and where they focused
their attention sustained higher enjoyment during the repetitive experience than a control group.
Critically, the intervention increased enjoyment without inhibiting increases in habit strength.
Process evidence via mediation analyses showed that the effect of the intervention on slowing
hedonic adaptation was mediated by increased attention to the task and decreased perceived
repetitiveness. The results contain novel theoretical implications for research on habit formation
and hedonic adaptation and practical implications for interventions targeting behavior change
and well-being.
Keywords: repetition, habit formation, hedonic adaptation, enjoyment, perception

iii
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT
Table of Contents
Title Page i
Acknowledgements ii
Abstract iii
Table of Contents iv
List of Tables v
List of Figures vi
Performance and Attention Novelty Slows Hedonic Adaptation During Habit Formation 1
Repetition and Habit Formation 4
Hedonic Adaptation, Attention, and Perception 8
How Will Slowing Hedonic Adaptation Influence Habit Formation? 13
Current Research 15
Experiment 1: Repeating a Virtual Sushi Recipe 15
Method 16
Results 25
Discussion 35
Experiment 2: Repeating a Piano Melody 36
Method 36
Results 46
Discussion 60
General Discussion 62
Implications for Research on Habit Formation 64
Implications for Research on Hedonic Adaptation 67
References 70
Appendix A 80
Appendix B 88
Appendix C 92
Appendix D 94
Appendix E 96

iv
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT
List of Tables

Table 1. Pearson Correlation Matrix of Measures from Experiment 1 31
Table 2. Means and Standard Deviations for Dependent Variables from Experiment 1 32
Table 3. Descriptive Statistics for Number of Melody Repetitions Per 90 Second Trial
in Experiment 2
48
Table 4. Pearson Correlation Matrix of Measures from Experiment 2 56
Table 5. Means and Standard Deviations for Dependent Variables from Experiment 2 57

v
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT
List of Figures
Figure 1. Screenshot of sushi game and example instruction 17
Figure 2. Experiment 1 procedure and sequence of tasks and measures 18
Figure 3. Example of how control group (left) and novelty group instructions (right)
were presented before a two minute trial
19
Figure 4. Structure of habit strength context-response association measure 20
Figure 5. Reaction times from habit strength cue-response association test in sushi
game
24
Figure 6. Ratings of perceived behavioral automaticity in sushi game 25
Figure 7. Enjoyment ratings in sushi game 27
Figure 8. Perceptions of repetition ratings of sushi game 28
Figure 9. Double mediation model in sushi game 29
Figure 10. Perceptions of repetition mediation model in sushi game 30
Figure 11. Instructions for where to place fingers to play the virtual piano 37
Figure 12. Feedback when participant pressed the correct number on the keyboard
(left) and incorrect (right)
38
Figure 13. Experiment 2 procedure and sequence of tasks and measures 39
Figure 14. Example of correct reaction time trial in the piano habit strength cue-
response association test
42
Figure 15. Reaction times in habit strength test in piano game 43
Figure 16. Number of completed melody repetitions per 90 second trial 49
Figure 17. Ratings of perceived behavioral automaticity in piano game 51
Figure 18. Enjoyment ratings in sushi game 53
Figure 19. Perceptions of repetition ratings of piano game 54
Figure 20. Double mediation model in piano study 55

vi
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 1

Performance and Attention Novelty Slows Hedonic Adaptation During Habit Formation

“ R out i ne , repetition, tedium, m onot ony … bor e dom , angst, ennui —these are the true he r o’s
enemies, and make no mistake, they are fear-some indeed. For they are r e al ” (Wallace, 2011, p.
231).
—David Foster Wallace

The road to habit formation is paved with repetition. Lots of repetition. In fact, one study
found that it took 91 days on average for exercise to feel like a habit (Lally, Van Jaarsveld, Potts,
& Wardle, 2010). Fortunately, if one stays on the road, repeated behavior becomes more
automatic: it ceases to require much conscious intent, attention, control, and effort (Bargh, 1994;
Evans & Stanovich, 2013; Ouellette & Wood, 1998; Shiffrin & Schneider, 1977; Wood &
Rünger, 2016). Past repetition facilitates future repetition. Unfortunately, long bouts of repetition
are often accompanied by David Foster Wallace ’s “ t r ue he ro’s e ne m i e s ” : feelings of tedium,
monotony, and boredom. In psychological terms, this phenomenon can be captured by the
process of hedonic adaptation. Hedonic adaptation describes the general empirical pattern that
the intensity of our pleasant or unpleasant responses to most stimuli, experiences, and activities
declines over time and with repetition (Frederick & Loewenstein, 1999; Frijda, 1988; Galak &
Redden, 2018). Although hedonic adaptation can be desirable at times —it can transform what
was once unpleasant into something tolerable —it poses a strong challenge to sustaining
enjoyment in life (see hedonic treadmill, Brickman & Campbell, 1971). Specifically, hedonic
adaptation may play a large, yet unexamined role in derailing many of our habit formation
pursuits prematurely. Hedonic adaptation may cause us to stop enjoying a behavior we have been
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 2

repeating, disrupt intrinsic motivation, and make long-term persistence and habit formation
unlikely.
Given the dismal results of most habit formation interventions (Wood & Neal, 2016),
New Year's resolutions (Norcross, Ratzin, & Payne, 1989; Prochaska, DiClemente, & Norcross,
1992), and general behavior change interventions (Marteau, Hollands, & Fletcher, 2012),
psychologists have called for new theoretical and practical approaches to sustaining behavior
change over the long-term. Researchers have questioned the effectiveness of approaching
behavior change via conscious processes or by increasing willpower and instead argued for
putting automatic processes such as habit formation at the center of effective self-regulation
(Galla & Duckworth, 2015; Carden & Wood, 2018; Marteau et al., 2012). Researchers who
study hedonic adaptation have suggested that slowing hedonic adaptation could facilitate
behavior change; yet, this strategy has yet to be explicitly investigated (Galak & Redden, 2018).
Promising recent research has shown that mere enjoyment of an activity (e.g., pleasant taste of
vegetables) is a more powerful predictor of sustained behavior change (e.g., eating more
vegetables) compared with valuing the rewards that come from repeating the behavior (e.g.,
becoming healthier, Woolley & Fishbach, 2017).
The idea that people persist in their behavior depending on current enjoyment levels
aligns with a feelings-as-information approach (Schwarz & Clore, 1983; Schwarz, 1990). For
example, participants in a positive mood in one study were more likely to continue repeating a
task when they were asked whether they were enjoying it vs. finishing it (Martin, Ward, Achée,
& Wyer, 1993). From this perspective, the motivational and behavioral implications of one ’s
feelings depend on which question one is asking oneself. That is, “ pos i t i v e moods tell us to
continue when they reflect our level of enjoyment but tell us to stop when they reflect our level
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 3

of goal a t t a i n m e nt ” (Martin et al., 1993, p. 325). Given that people often use a "how do I feel
about it?" heuristic when performing tasks (Schwarz & Clore, 1988), a better understanding of
how enjoyment fluctuates with repetition in relation to the habit formation process can have
novel implications for how we can persist to achieve long-term goals. Yet, no experimental
research has explicitly examined how to slow hedonic adaptation during habit formation.
The present research identifies hedonic adaptation as a challenge to persistence and thus
to habit formation. Consequently, I test an intervention designed to slow hedonic adaptation as
people repeat a behavior. In addition, I examine mechanisms that may drive this effect —pe op l e ’s
attention to distinct aspects of a task and pe op l e ’s perceptions of repetition —and measure how
the proposed intervention, which employs conscious shifts in performance and attention,
influences habit formation.
Psychology has long traditions of studying the effect of repetition on behavior (e.g., see
reviews on habit formation and automatization, Wood & Rünger, 2016; Shiffrin & Schneider,
1977) and the effect of repetition on hedonic responses (e.g., see reviews of hedonic adaptation
and hedonic decline, Frederick & Loewenstein, 1999; Galak & Redden, 2018). When William
James (1977, p. 12) claimed that “ h a bi t depends on sensations not attended to ,” perhaps he was
implying that habit formation depends on the attenuation of emotional intensity; contemporary
research supports this idea by demonstrating that reports of habitual behavior are often
negatively correlated with emotional intensity (Wood, Quinn, & Kashy, 2002). Yet, no
experimental research that I know of has tested this question. And there is reason to believe that
the trajectory of emotional intensity can be independent of the effects of repetition on behavioral
automatization. Research in cognitive neuroscience suggests that many implicit processes that
control action and emotion (e.g., classical conditioning, priming, and habit learning) have
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 4

separate and specialized neural substrates (e.g., amygdala, prefrontal/temporal cortex, basal
ganglia) that operate somewhat independently (Amodio & Ratner, 2011). Little is known about
the extent to which hedonic adaptation and habit formation processes are related.
Broadly, this research investigates how repetition influences habit formation and
enjoyment. And given the clear practical implications of testing interventions to sustain
enjoyment during repetition, the present research is a seed for future interventions promoting
habit formation. In the following sections, I first explain how habit formation and hedonic
adaptation were defined, measured, and studied in past research. Then, I outline an intervention
to slow hedonic adaptation and describe the mechanisms that explain why the intervention may
work. Throughout the introduction, I will present the hypotheses that are tested in two
experiments.
Repetition and Habit Formation
As people repeatedly perform an action in a stable context, implicit associations form in
memory between context cues and behavioral responses via Hebbian and/or reinforcement
learning (Gardner, 2014; Orbell & Verplanken, 2010; Wood & Rünger, 2016). These
associations are multimodal memory representations that can be activated automatically by
recurring perceptual cues, e.g., people, environments, feelings. Once activated, people are likely
to act on the behavior in mind (perception-to-behavior pathway, Bargh & Chartrand, 1999;
ideomotor principle, James, 1890).
When a behavior becomes more automatic, it relies less heavily on executive control and
can be activated with little conscious awareness or effort (Evans & Stanovich, 2013; Strack &
Deutsch, 2004; Wood & Rünger, 2016). Repetition causes behavioral control to shift from
regions in the prefrontal cortex, which are associated with self-control, to regions in the
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 5

sensorimotor cortices, basal ganglia, and midbrain, which are associated with automatic
processes (Knowlton & Patterson, 2016; Tricomi, Balleine, & O ’D ohe rt y, 2009; Yin &
Knowlton, 2006). Many psychological processes operate automatically, such as priming and
classical conditioning. However, habit is one type of automatic responding that emerges through
repetition, as people learn to associate a context cue with a rewarded response (Amodio &
Ratner, 2011; Wood & Rünger, 2016). Ultimately, perception of the cue triggers thoughts of the
response in mind.
In terms of habit formation, automaticity is conceptualized as the strengthening of
implicit associations between cues and responses (Anderson, 1987). Stronger associations
between cues and response implies that, in the presence of relevant context cues, habit responses
are chronically more accessible (i.e., come to mind more quickly) than non-habit responses.
(Neal, Wood, Labrecque, & Lally, 2012; Moors & De Houwer, 2006). In fact, this chronic
accessibility is due, in part, to the way attention is biased by reward learning. A recent and
growing body of experimental evidence on “ t he attention ha bi t ” (Anderson, 2016) is revising
earlier models of selective attention —traditionally characterized by the distinction between top-
down, motivational factors or stimulus-driven, salience factors —by positing that attention
operates like a habit: it is driven by the same associative, reward learning processes that underlie
habit behavior (Anderson, 2016). This body of evidence suggests that rewards influence attention
directly (as opposed to indirectly via modulating goals and motivation), which augments our
understanding of why habit responses come to mind quickly, bias information search
(Verplanken, Aarts, & van Knippenberg, 1997), and cause people to overlook alternative action
options (Verplanken et al., 1997). In the context of L oga n’s (1988) horse race response
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 6

inhibition model, in which a go and stop process compete in the race to direct action, habit is like
a go response, outrunning the competition thanks to an attentional head start.
One way to measure the strength of a habit is through cognitive association tests using
reaction time indices. For example, one study measured how quickly habitual and nonhabitual
runners responded with thoughts of running to the locations in which they often ran (Neal et al.,
2012). When prompted by locations in which they ran (cue), habitual runners were faster to
recognize words related to running and jogging (response) compared with nonhabitual runners.
That is, the cue activated running thoughts which, in turn, made it easier to process associated
words. Because cognitive association measures directly assess the strength of automaticity
between cues and responses, they are considered the gold standard for measuring habit strength
and automaticity among habit researchers (Gardner, 2014; Labrecque & Wood, 2015; Mazar &
Wood, 2018). Yet, only a few published studies have used this kind of measure (Danner, Aarts,
& de Vries, 2008; Neal et al., 2012; Lin, Wood, & Monterosso, 2016). The main challenge with
such measures is in identifying in advance the cue and response. In other words, researchers can
only measure habit strength this way if they are sure which cues trigger which responses. In the
present research, I employ a paradigm in which participants develop specific cue-response
associations as they repeat a predetermined sequence of steps (i.e., there is a fixed cue-response
pairing known to the researchers, similar to research conducted by Lin et al. (2016). Critically,
both unique tasks used in the experiments also contain tailored reaction time tests to measure the
strength of the specific cue-response associations.
In addition to direct cue-response associations, another measure of habit comes from the
phenomenological experience of an action becoming more automatic (e.g., effortless). The Self-
Report Behavioral Automaticity Index (SRBAI, Gardner, Abraham, Lally, & de Bruijn, 2012), a
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 7

four-item subset of the Self-Report Habit Index (SRHI, Verplanken & Orbell, 2003), measures
pe opl e ’s reports of performing an action automatically, without consciously remembering, and
without thinking (Gardner et al., 2012). In short, it measures how automatic a behavior is
perceived to feel. Researchers have questioned whether the SRHI is an adequate measure of
habit given that it “ c onfl a t e s habit with possible consequences of ha b i t s ” (Sniehotta & Presseau,
2012, p. 139) —i.e., felt automaticity is a consequence of habit, not a determinant of habit (the
cognitive association). Although Sniehotta and Presseau (2012, p. 139) argue that we should
reexamine “ T he Habitual Use of the Self-report Habit Inde x” , they and many habit researchers
argue for multi-pronged assessments of habit (Mazar & Wood, 2018; Orbell & Verplanken,
2018). Consequently, in the present research, I measure the unconscious strengthening of
cognitive associations and the accompanying phenomenological experience of automaticity as
participants repeat a task. I then examine how these measures are correlated, which has not been
done in published research (Mazar & Wood, 2018; Orbell & Verplanken, 2018).
Thus, according to contemporary theory on habit formation, behavioral repetition will
lead to increased cue-response automaticity:
Hypothesis 1a: Repeating a task will lead to stronger cue-response associations.
As a behavior is repeated, subjective ratings of perceived behavioral automaticity will
increase:
Hypothesis 1b: Repeating a task will lead to higher ratings of perceived behavioral
automaticity.
Given that the cue-response association test and self-reported behavioral automaticity
index are measuring the same construct, habit, these measures should be positively correlated:
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 8

Hypothesis 1c: Stronger cue-response associations will be positively correlated with self-
reported behavioral automaticity.
Hedonic Adaptation, Attention, and Perception
Few stimuli we encounter in life escape the pervasive effect of hedonic adaptation, the
gradual decline in the intensity of our pleasant or unpleasant responses over time and repetition
(Frederick & Loewenstein, 1999). What once elicited a strong pleasant response —winning the
lottery (Brickman, Coates, & Janoff-Bulman, 1978), a well-liked song, (Ratner, Kahn, &
Kahneman, 1999), a favorite food (Epstein, Temple, Roemmich, & Bouton, 2009), artwork
(Redden, 2008), close others (Galak, Redden, & Kruger, 2009), or marriage (Lucas & Clark,
2006) —or a strong negative response —tart yogurt (Kahneman & Snell, 1992), unpleasant
sounds (Nelson & Meyvis, 2008), and even incarceration (Zamble, 1992) —over time and with
repetition, elicit a less intense hedonic response. Many theories have sought to explain why and
how this attenuation occurs: habituation (Frijda, 1988; Groves & Thompson, 1970; Thompson &
Spencer, 1966), satiation (Rolls, Rolls, Rowe, & Sweeney, 1981), range-frequency theory
(Parducci, 1995), hedonic adaptation (Frederick & Loewenstein, 1999), and affective adaptation
(Wilson & Gilbert, 2008). Many theories have noted short-term exceptions to hedonic adaptation
such as increases in hedonic responses to novel stimuli via mere exposure (Zajonc, 1968) or via
sensitization (Crolic & Janiszewski, 2016). Eventually, however, even stimuli influenced by
mere exposure succumb to the effect of hedonic adaptation (Galak & Redden, 2018).1 In short,
for better or for worse, strong feelings fade with repetition.

1A contrasting, but related phenomenon to hedonic adaptation is the experience of flow (Csikszentmihalyi, 1990), a
positive state that can accompany repetition. Flow emerges when one ’ s skill level for a task is stretched to meet the
level of difficulty of a task. If a task becomes too easy because skill level is high, boredom ensues. Thus, to continue

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 9

Paradigms for studying hedonic adaptation have either used within-session or across-
sessions approaches. Within-session, participants are repeatedly exposed to a stimulus or repeat
an action. After the first repetition and after the final repetition (e.g., 20 repetitions later),
participants explicitly rate their enjoyment. If final ratings are lower than initial ratings, then
hedonic adaptation has occurred. In a pre-load variation of this paradigm, used primarily with
food stimuli, participants repeatedly eat until they no longer find it enjoyable or desirable and are
then asked whether they want to continue (Griffioen-Roose et al. 2010). Across-sessions
approaches compare two groups of people after repeated exposure to different quantities of a
stimulus. For example, Galak et al. (2009) found that participants who were exposed to their
favorite song 20 times vs. a few times in one lab session were less likely to enjoy that song even
two weeks later in a follow up session. Also, across-sessions approaches allow for comparisons
between a group of people who have experienced a long-lasting stimulus with others who have
not (e.g., lottery vs. non-lottery winners; Brickman, Coates, & Janoff-Bulman, 1978; tenured vs.
non-tenured, Gilbert, Pinel, Wilson, Blumberg, & Wheatley, 1998).
Many factors influence hedonic adaptation (Galak & Redden, 2018): physiological
feedback (e.g., stomach distention, testosterone), perceptual changes (e.g., attention and novelty),
and self-reflection (e.g., perceived past variety, metacognitive theories). Because stimuli vary,
not all factors are relevant to all stimuli (e.g., physiological factors like stomach distention affect
food hedonic adaptation, but not the rate of hedonic adaptation for marriage or winning the

eliciting flow a task needs to increase in difficulty. The set of tasks I am referring to in the present research are fixed
in difficulty, which poses a different kind of challenge to repetition and sustaining enjoyment: how can one continue
to enjoy repeating a task that is unchanging. As mentioned, this is especially relevant for habitual behaviors in which
one must repeat the same task to develop automaticity.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 10

lottery). Since I will be examining the effect of repeating a motor task and forming a habit, I will
not examine physiological feedback factors, which play a predominant role in food stimuli.
Broadly, psychological strategies for slowing hedonic adaptation introduce the
experience of novelty in one of two general forms: novelty-via-breadth or novelty-via-depth
(Sheldon, Boehm, & Lyubomirsky, 2013; Lyubomirsky, Sheldon, & Schkade, 2005; O ’Bri e n,
2019). The former strategy is summarized by the notion that “ va ri e t y is the spice of l i fe .” This
strategy is intuitive —people prefer and choose novel experiences when seeking to maximize
their enjoyment (Kahn & Ratner, 2005; Ratner et al., 1999; Read & Loewenstein, 1995) —and it
works: novel stimuli capture one ’s attention, elicit more intense feelings than familiar stimuli,
and create a more enjoyable immersive experience (Berlyne, 1970; Csikszentmihalyi, 1990;
Killingsworth & Gilbert, 2010). To enact this strategy, people can avoid extended exposure to a
repeat stimulus (Quoidbach & Dunn, 2013) or interleave novel stimuli between repetitions
(Sheldon et al., 2013). In short, avoid the repeated stimulus and approach other novel stimuli.
Yet, if one ’s goal is to maximize habit formation, that is, maximize response repetition, a
strategy of avoiding repetition slows the rate of habit formation. The novelty-via-depth strategy,
in contrast, introduces experiences of novelty within the repeated stimulus itself and thus
encourages response repetition.
The novelty-via-depth strategy exploits the fact that hedonic adaptation is, in part, a
product of where our attention is focused and our perception (Kahneman, Krueger, Schkade,
Schwarz, & Stone, 2006; Galak & Redden, 2018). That is, the rate of hedonic adaptation is not
simply a function of how much time has passed nor how much repetition has occurred, but also
how one perceives the passing of time and numbers of repetitions. For example, in one study,
researchers manipulated particip a nt s ’ perceptions of repetition by drawing p a rt i c i pa nt s ’ attention
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 11

to different features of a repeated stimulus (Redden, 2008). Under the guise of a consumer taste
test study, participants were instructed to eat a set of jelly beans, one by one, and rate their level
of enjoyment every few beans. For some participants, the jelly beans were labeled at a general
level of abstraction (e.g., “ j e l l y bean 1 ” , “ j e l l y bean 2” ), whereas for others, the same jelly beans
were labeled at more detailed, concrete level (e.g., “ c he rry jelly be a n ” , “ l e m on jelly be a n” ).
Subcategory participants did not report experiencing the same stimulus as much as the general
category participants: they paid more attention to differentiating aspects of the jelly beans,2
perceived eating the jelly beans as less repetitive and more novel,3 and consequently enjoyed the
jelly beans more. This study reflects broader principles: pe opl e ’s ongoing hedonic experience
depends on where attention is drawn (see attentional adaptation, Kahneman et al., 2006; O ’Bri e n,
2019) and the level by which people perceive or categorize activities (i.e., concrete to abstract,
Vallacher & Wegner, 1987). Thus, if attention can shift to distinct attributes of a repetitive
experience and alter on e ’s perception of repetition, then experiences of novelty can arise without
introducing completely novel stimuli (i.e., employing the novelty-as-breadth strategy). In short,
the spice of variety can be found in the familiar.
The novelty-as-depth strategy is an overlooked means of combating hedonic adaptation
because people falsely believe that repeating activities is duller than it seems (O ’Bri e n, 2019)
and because people fail to appropriately account for the divergence between in situ enjoyment
and forecasted enjoyment (Schwarz & Xu, 2011). A recent set of studies demonstrated that

2See Redden (2008): Attention manipulation check: flavor discriminability ( “ I could identify the specific flavor of
each jelly be a n,” “ T he flavor of each jelly bean was obvi ous ” ) and flavor salience ( “ I really noticed the specific
flavor of each jelly be a n,” “I did not pay much attention to the different flavors of the jelly beans [reverse c ode d] ,”
“I really noticed the color of each jelly be a n” ) .
3See Redden (2008): Perception of repetition were measured by asking about redundancy ( “ E a ti ng the jelly beans
felt like the same thing over and ove r ,” “ E a ti ng the jelly beans was very bor in g” ) and similarity ( “ T he jelly beans
were very similar to each ot he r ,” “ E a c h jelly bean had aspects that made it different [reverse c ode d] ” ) .
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 12

pe opl e ’s predicted enjoyment of repeat activities (e.g., watching the same movie, visiting the
same museum exhibit) underestimated how much they actually enjoyed repeating those activities
(O ’Bri e n, 2019). Because affective forecasts are often theory driven (Schwarz & Xu, 2011) —
e.g., repetition is dull —people failed to consider that actual enjoyment of repetition would be
driven by in situ attention and perception. In fact, two studies showed that enjoyment from
repetition was mediated by the amount of novelty attended to and perceived within the repeat
activity (see Studies 4 and 5, O ’Bri e n, 2019). Especially for complex, information-rich activities,
repetition can seem novel because attention can be drawn to new attributes of the activity.
To test the novelty-as-depth strategy, psychologists have nudged participants to
experience the same activity in novel ways by adjusting how a repeat activity is performed —for
example, by slightly changing interaction with a stimulus (e.g., using chopsticks vs. hands to eat
popcorn, O ’Bri e n & Smith, 2019) —or by shifting attention —for example, drawing attention to
novel features of a complex stimulus (e.g., variety of flavors in a complex fruit juice, Crolic &
Janiszewski, 2016; Redden, 2008; O’Brien, 2019). Because this novelty-via-depth strategy can
slow hedonic adaptation by changing how people perform and perceive an activity versus
avoiding what activity people repeat, it seems like a promising way to keep people repeating a
target activity and thereby facilitate habit formation. Therefore, the following experiments test an
intervention that slows hedonic adaptation by instructing participants to vary how they perform
and attend to a repetitive task. The intervention should slow hedonic adaptation:
Hypothesis 2a: Varying performance and attention as a task is repeated will slow
hedonic adaptation.
In addition, based on previous research using novelty-via-depth strategies to slow
hedonic adaptation (Crolic & Janiszewski, 2016; Redden, 2007; O ’Bri e n & Smith, 2019), the
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 13

effect of the intervention on slowing hedonic adaptation should increase attention to distinctive
aspects of the repeated task and decrease perceptions of repetition:
Hypothesis 2b: Varying performance and attention as a task is repeated will increase
attention paid to distinct aspects of the task.
Hypothesis 2c: Varying performance and attention as a task is repeated will make the
task seem less repetitive.
Finally, pe op l e ’s attention to distinct aspects of the task and perceptions of repetition
should mediate the effect of the intervention on slowing hedonic adaptation (see similar
mediators: increased attention to food, Crolic & Janiszewski, 2016; O ’Bri e n & Smith, 2019;
Redden, 2008). Specifically, Redden (Study 3, 2008) found a double mediation in which
increased attention to distinct aspects of the task, a proximal mediator, led to lower perceptions
of repetition, a distal mediator, and consequently slowed hedonic adaptation. The double
mediation will also be tested in the present experiments. Thus:
Hypothesis 2d: The performance and attention variation intervention will increase
attention to distinct aspects of a repeated task, making the task seem less repetitive, and
consequently slow hedonic adaptation.
How Will Slowing Hedonic Adaptation Influence Habit Formation?
Given that the proposed intervention to slow hedonic adaptation involves instructing
people to slightly shift performance and attention during a repeated task, there is reason to
believe that this intervention may negatively affect habit formation. Research has shown that
conscious reflection can disrupt habit and skilled performance (Baumeister, 1984: Beilock &
Carr, 2001; Carden, Wood, Neal, & Pascoe, 2017; Masters, 1992; Lewis & Linder, 1997).
Researchers have posited a number of explanations for this effect —often studied in the context
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 14

of “ c hoki ng under pr e s s ure ” —such as self-focus theories (increased attention to step-by-step
control interferes with proceduralized performance, Lewis & Linder, 1997) or distraction
theories (a single proceduralized task becomes a dual-task since attention is directed at task-
irrelevant cues, cognitive demands increase and thereby disrupt automaticity, Beilock & Carr,
2001). Notably, these large literatures have examined only disruptions to habit performance, not
habit learning. Little is known about whether slight variations in performance or attention could
disrupt habit formation. In fact, there is no published research showing that slightly varying
performance and attention during repetition prevents or disrupts the formation of cue-response
cognitive associations. The present research will shed some light on this issue by comparing how
habits form with or without variation.
Finally, the present research may help us better understand how the processes of hedonic
adaptation and habit formation are related in a broader sense. For example, are the processes
correlated? That is, does decreasing enjoyment correlate with increasing habit strength as people
repeat a behavior? Although, a daily diary study conducted by Wood, Quinn, and Kashy (2002)
showed a correlation between increasing habit strength and decreasing emotional intensity, their
studies only used self-report measures of habit. In contrast, the present experiments measure
habit directly (via cognitive association tests) and indirectly (via self-report), allowing for further
testing of the relationship between habit strength and emotion. Preliminary evidence from a pilot
study with the first research paradigm showed that enjoyment was negatively correlated with the
direct measure of habit strength, measured via cognitive associations, but no correlation was
present with self-reported experience of habit, perceived automaticity (see data in Appendix A).
By conducting exploratory analyses on how habit formation and hedonic adaptation are
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 15

correlated in the present experiments, we may deepen our understanding of how repetition, habit
formation, and emotion are related.
Current Research
In the present research, participants repeated an activity and completed measures of
enjoyment and habit strength at two time points. After initial repetition, baseline measurements
of enjoyment and habit were administered. Then, some participants repeated the activity with
instructions to slightly vary performance and attention (novelty group); others repeated the
activity with control instructions (control group). Measures of enjoyment and habit were taken a
second time, in addition to measures of attention to the task and perceptions of repetition.
Between (group) and within (time) subject differences were analyzed. Two different repetitive
tasks were used to test the robustness of the predictions.
Experiment 1: Repeating a Virtual Sushi Recipe
Experiment 1 was conducted in a controlled lab setting. Participants played a computer
game that involved making a sushi roll in a 16-step sequence. Some participants were instructed
to vary how they moved the ingredients with the computer mouse and where to focus their
attention when repeating the game, whereas others, with control instructions, just repeated the
sequence. Slightly varying performance and attention should make the task more enjoyable,
increase attention to its distinct aspects, and feel less repetitive. If the intervention disrupts habit
formation participants in the novelty condition should show weaker habit strength.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 16

Method
Participants. A total of 135 undergraduates (62% female) participated from the USC
psychology participant pool for psychology class credit or $5. Each participant was assigned a
private cubicle with a desktop computer and mouse. Sample size was predetermined for all
studies to comprise at least 50 participants per experimental cell, comparable with past designs
(e.g., Crolic & Janiszewski, 2016; O’Brien, 2019; O ’ Bri e n & Smith, 2019; Redden, 2008). For
this and all experiments, we report all measures, manipulations, and conditions.
Procedure. Participants signed up for a study on making virtual sushi. Originally
designed for children, the sushi video game was tailored by a computer programmer for our
research, allowing us to precisely control how long participants repeated the sushi game
sequence. The sushi game was pilot tested for reliably eliciting hedonic adaptation and showing
increases in habit strength within a 30-minute lab session (see pilot data in Appendix A).
Participants were guided by an avatar on the screen that described which ingredients to use for
each step in the 16-step recipe (see Fig. 1 below). Participants moved the cursor over to the
relevant item on the table, clicked on it, and dragged it to the appropriate location (e.g., drag rice
bowl to pot). The task provided immediate feedback about success by enabling participants to
move to the next step in the recipe (e.g., the rice would empty into the pot). Attempting to click
and move an incorrect recipe item yielded no program response (e.g., putting the spoon in the
vinegar bottle was not possible). Participants could make as many tries as they wished until they
selected the correct item in the sequence.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 17

Figure 1. Screenshot of sushi game and example instruction.

In the first phase of the experiment, all participants completed an untimed trial to get
acquainted with the game (see Fig. 2 below for procedure outline). Then, they repeated the recipe
as many times as they could within two minutes. A clock on the lower right corner of the game
counted down from two minutes in seconds. After two minutes, participants completed a
measure of habit strength of the recipe sequence, a cognitive association reaction time test (see
below). Then, they completed Time 1 self-report measures of enjoyment, perceived automaticity,
boredom, and motivation (described below).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 18

Figure 2. Experiment 1 procedure and sequence of tasks and measures.

In the second phase of the experiment, all participants were told that they would repeat
the sushi game four more times in two-minute trials and that they would receive instructions
before each two minute trial. The control group participants were instructed to “ c on t i nue to play
the game as you have been playing i t ” before each two minute trial. The novelty group
participants received four different kinds of performance and attention instructions before each
two minute trial (see Fig. 3 below): “ M ove quickly. Focus on how quickly you move the
i ngre di e n t s ,” “ M ove smoothly. Focus on how smoothly you move the i ngre d i e nt s ,” “ M ove
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 19

sharply. Focus on how sharply you move the i ngre di e nt s ,” and “ M ove slowly. Focus on how
slowly you move the i ngre di e nt s .”

Figure 3. Example of how control group (left) and novelty group instructions (right) were
presented before a two minute trial.

At the end of the four trials (each two minutes), all participants repeated the same habit
strength test and self-report measures in the first phase of the experiment. In addition,
participants completed measures of attention, perceptions of repetition, and demographics.
Measures.
Habit strength. To measure habit formation directly, a reaction time test assessed the
strength of learned cue-response associations from the sushi game sequence (this exact test was
used previously in Labrecque, 2015; Labrecque, Lee, & Wood, unpublished; similar habit
strength measures were also used in Neal et al., 2012 and Danner et al., 2008). In the test,
participants were primed with one step in the sushi-making sequence (cue) and then instructed to
choose the correct next step in the recipe (response) from two choices (see Fig. 4 below). If
participants developed stronger cue-response associations in memory, they should pick the
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 20

correct next step in the recipe faster because it is more cognitively accessible (Wood & Rünger,
2016).

Figure 4. Structure of habit strength context-response association measure. Participants were
instructed to choose the ingredient that follows the prime in the recipe as quickly as possible.

Specifically, each reaction time trial consisted of a fixation cross (2 seconds), a prime
representing a prior step in the recipe (e.g., vinegar bottle, 1 second), and then a choice of two
possible next steps in the recipe (one correct, one incorrect, 3 seconds). Participants were
randomly assigned 30 trials. Strength of cognitive associations was assessed from mean reaction
times (RT) for all correct trials (Average percentage of correct trials for novelty group
participants was 88% at Time 1 and 95% at Time 2. Average percentage of correct trials for
control group participants was 92% at Time 1 and 95% at Time 2. No group differences in
number of correct trials were observed at Time 1, t(133) = -1.88, p = .062, and Time 2, t(133) = -
.02, p > .25.)
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 21

Enjoyment. Participants rated their agreement with the statement, “I am enjoying making
s us hi ,” from 1 (strongly disagree) to 7 (strongly agree).
Motivation. Participants rated their agreement with the statement, “I am motivated to
play the sushi ga m e ,” from 1 (strongly disagree) to 7 (strongly agree).
Boredom. Participants rated their agreement with the statement, “ T he sushi game is
bori ng,” from 1 (strongly disagree) to 7 (strongly agree).
Perceived behavioral automaticity. Perceived behavioral automaticity was measured
using the Self-Report Behavioral Automaticity Index (Gardner et al., 2012), a four-item subset of
the Self-Report Habit Index (SRHI, Verplanken & Orbell, 2003). This measure is considered a
self-report measure of habit formation since it assesses how automatic an action is perceived to
feel; the more habitual the behavior, the more automatic it should feel (Mazar & Wood, 2018).
Participants indicated from 1 (strongly disagree) to 7 (strongly agree), “Each step in the sushi
recipe is s om e t hi ng... ” “I do a ut o m a t i c a l l y; ” “I do without having to consciously r e m e m be r; ” “I
do without t hi nk i ng; ” and “I start doing before I realize I’m doing i t .” Reliability was good
(alpha = .90).
Perceptions of repetition. Adapted from Redden (2008), perceptions of repetition was
measured by asking about redundancy (“Playing the sushi game felt like the same thing over and
ove r” ) and similarity (“ E a c h round making sushi seemed very similar to each o t he r,” “ E a c h
round making sushi had aspects that made it d i ffe re n t ,” reverse coded). Participants indicated
agreement to items from 1 (strongly disagree) to 7 (strongly agree). Reliability across all
measures was good (alpha = .69).
Attention to distinct aspects of task. Adapted from Redden (2008), attention was
measured by asking about discriminability (“ D i ffe re nt ways to move the ingredients when
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 22

making sushi were obvi ous ,” “I could identify different ways to move the ingredients when
making s us hi ” ) and salience (“ I really noticed the different ways I moved the ingredients when
making s us hi ,” “I did not pay much attention to the different ways I moved the ingredients when
making s us hi ” reverse coded). Participants indicated agreement to items from 1 (strongly
disagree) to 7 (strongly agree). Reliability across all measures was acceptable (alpha = .66).
Retrospective judgments of variety. Adapted from Redden (2008; see also Broniarczyk,
Hoyer, & McAlister, 1998), retrospective judgments of variety were measured with two items
(“ T he re was a lot of variety in the sushi ga m e .” , “ T h e re were many different ways to play the
sushi ga m e .” ). Participants indicated agreement to items from 1 (strongly disagree) to 7 (strongly
agree). Reliability was good (alpha = .78).
In Redden (Study 3, 2008) the items from the above three indexes —perceptions of
repetition, attention to distinct aspects of task, and retrospective judgments of variety —were
factor analyzed and yielded three separate constructs with items loading most highly on the
intended factors. I also conducted a principal components analysis on all items and results
indicated the same three factor structure with items loading highest on the intended factors (see
factor analysis in Appendix B).
Willingness to continue playing. As a proxy for measuring participants ’ desire to
continue repeating the sushi game, participants were asked, “ If you were paid to continue playing
this game for 30 more minutes, how much money would you request from us? From 0 - 100 in
dol l a rs .”4

4Additional measures of personality were assessed to test an unrelated hypothesis and will not be discussed further.
Participants were asked the degree to which certain adjectives described them from 1 (strongly disagree) to 7
(strongly agree using the Ten Item Personality Inventory (TIPI) by Gosling, Rentfrow & Swann, 2003).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 23

Results
My analysis plan is as follows: First, I present statistical tests for my hypotheses
(Hypotheses 1a/b/c and Hypotheses 2a/b/c/d). I describe the type of statistical test used at the
beginning of each section. Second, I present exploratory analyses. Means and standard
deviations for all dependent variables in Experiment 1 are shown in Appendix C.
Note, given that some dependent variables were measured at two time points (Time
1/Time 2), a variety of methods for analyzing data using baseline test-posttest (Time 1-Time 2)
type research designs could be used. I used the most exhaustive and conservative test, a repeated
measures ANOVA, since it provides simultaneous tests of differences between novelty and
control instruction conditions (between subjects factor), differences between Time 1 and Time 2
(within subjects factor), and the interaction between the two factors. Staticians also recommend
using ANCOVAs for baseline test-posttest designs in some cases because the F statistic results
provided by repeated measures ANOVAs for main effect tests are conservative and sometimes
difficult to interpret, given that baseline/Time 1 measure differences are not affected by the
treatment (Dimitrov & Rumrill, 2003). Thus, to aid in the interpretation of repeated measures
ANOVA results I sometimes add ANCOVAs to test for Time 2 condition differences while
controlling for random variance in baseline/Time 1 differences.
Increasing habit strength (hypothesis 1a). A Condition (between-subjects factor:
Novelty, Control) x Time (within-subjects factor: reaction times on cue-response association test
at Time 1, reaction times on cue-response association test at Time 2) repeated measures ANOVA
on the RT measure of habit strength revealed the predicted main effect of time such that
repeating the sushi game led to significantly faster reaction times (in milliseconds) at Time 2 (M
= 648, SD = 147) vs. Time 1 (M = 850, SD = 191), F(1, 133) = 190.76, p < .001, partial η2 = .589
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 24

(see Fig. 5 below).5 There was no main effect of condition, F(1, 133) = 2.79, p = .097, partial η2
= .021, nor a significant interaction, F(1, 133) = .980, p > .25, partial η2 = .007, suggesting that
the novelty intervention did not influence increases in habit strength.

Figure 5. Reaction times from habit strength cue-response association test in sushi game.

Increasing perceived behavioral automaticity (hypothesis 1b). A Condition (between-
subjects factor: Novelty, Control) x Time (within-subjects factor: perceived behavioral
automaticity at Time 1, perceived behavioral automaticity at Time 2) repeated measures
ANOVA on behavioral automaticity revealed the predicted main effect of Time such that

5Reaction time data were also analyzed after being log transformed. Results were the same, so raw scores are
presented for ease of interpretation. Also, as recommended from research on analyzing RT data, a cut off of 200 ms
was used when including responses for analysis (Whelan, 2008).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 25

repeating the sushi game led to significantly higher ratings of perceived behavioral automaticity
at Time 2 (M = 5.13, SD = 0.11) vs. Time 1 (M = 3.99, SD = 0.11), F(1, 133) = 157.41, p < .001,
partial η2 = .542 (see Fig. 6 below). There was no main effect of condition, F(1, 133) = 0.01, p >
.25, partial η2 = .0, nor a significant interaction, F(1, 133) = 2.61, p = .108, partial η2 = .019,
suggesting that the novelty intervention did not influence increases in self-reported behavioral
automaticity.

Figure 6. Ratings of perceived behavioral automaticity in sushi game.

Correlations between habit measures (hypothesis 1c). The reaction time measure of
habit from the cognitive association test and ratings of perceived behavioral automaticity were
uncorrelated at Time 1, r(135) = -.12, p = .151, and at Time 2, r(135) = -.135, p = .117. Although
both measures showed strengthened habits as participants repeated the sushi game, the increases
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 26

in stronger cognitive associations and increases in experienced automaticity were unrelated; in
other words, this indicates that some participants increased strength through faster reaction times
whereas others increased strength through greater experienced automaticity.
Slowing hedonic adaptation (hypothesis 2a). A Condition (between-subjects factor:
Novelty, Control) x Time (within-subjects factor: enjoyment at Time 1, enjoyment at Time 2)
repeated measures ANOVA on enjoyment revealed the predicted main effect of Time such that
repeating the sushi game led to significantly lower enjoyment ratings at Time 2 (M = 4.84, SD =
0.13) vs. Time 1 (M = 5.49, SD = 0.11), F(1, 133) = 34.72, p < 0.001, partial η2 = .207 (see Fig.
7 below). There was also a main effect of condition such that novelty group participants enjoyed
playing the sushi game more (M = 5.47, SD = 0.15) than control group participants (M = 4.86,
SD = 0.15), F(1, 133) = 8.54, p < 0.01, partial η2 = .06. However, these main effects were
qualified by the predicted interaction between Condition x Time on enjoyment, F(1, 133) = 3.91,
p = 0.05, partial η2 = .207, such that enjoyment ratings decreased less between Time 1 and Time
2 for the novelty group (Time 1 M = 5.69, SD = 1.23, Time 2 M = 5.26, SD = 1.42) vs. control
group (Time 1 M = 5.29, SD = 1.21, Time 2 M = 4.43, SD = 1.58).

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 27

Figure 7. Enjoyment ratings in sushi game.

Given the marginal differences in enjoyment at Time 1 between groups (Novelty group,
M = 5.69, SD = 1.23, Control group, M = 5.29, SD = 1.21, t(133) = 1.87, p = .064), I also
conducted an ANCOVA to control for enjoyment ratings at Time 1 and test whether enjoyment
ratings at Time 2 differed significantly between groups. As predicted, Time 2 enjoyment ratings
were significantly greater in the novelty group vs. control group when controlling for Time 1
enjoyment, Mdiff = 0.55, F(1, 134) = 6.51, p = .012, partial η2 = .047.
Overall, these results suggest that the novelty intervention sustained higher enjoyment
with repetition and thus slowed hedonic adaptation.
Novelty intervention causes no difference in attention to distinct aspects of task
(hypothesis 2b). An independent-samples t test revealed no difference in attention to distinct
aspects of the sushi game participants in the novelty group (M = 4.5, SD = 1.12) vs. control
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 28

participants (M = 4.8, SD = 0.94) , t(133) = -1.57, p = .119). Thus, the novel performance and
attention instructions did not cause people to report paying more attention to the details of the
sushi game (e.g., “I really noticed the different ways I moved the ingredients when making
s us hi ” ).
Novelty intervention lowers perceptions of repetition (hypothesis 2c). An
independent-samples t test revealed that participants in the novelty group perceived playing the
sushi game as less repetitive (M = 4.99, SD = 1.17) than did control participants (M = 5.56, SD =
0.92) , t(133) = 2.73, p = .002, see Fig. 8 below), suggesting that the novel performance and
attention instructions decreased pa r t i c i pa nt s ’ perceptions of repetition.

Figure 8. Perceptions of repetition ratings of sushi game.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 29

Perceptions of repetition mediate the relationship between novelty instructions and
enjoyment (hypothesis 2d). To investigate whether the effect of the intervention on slowing
hedonic adaptation (i.e., lesser decrease in enjoyment between Time 1 and Time 2) was driven
by corresponding differences in perceptions of repetition, a double mediation analysis was
conducted using condition (Novelty, Control instructions) as the independent variable, change in
enjoyment as the dependent variable, and attention to distinct aspects of the task and perceptions
of repetition as the mediators. The indirect effect was tested using percentile bootstrap estimation
approach with 10000 samples (Shrout & Bolger, 2002), implemented with SPSS PROCESS
Model 6 (Hayes, 2017). The results did not support the double mediation since the indirect effect
of condition on enjoyment, via attention and perceptions of repetition, was not significant,
Indirect Effect = 0.17, SE = .016; 95% CIbootstrapping [-0.004, -0.06], see Fig. 9 below.

Figure 9. Double mediation model in sushi game. Standardized regression coefficients for the
relationship between condition and enjoyment as mediated by perceptions of repetition. The
standardized regression coefficient representing the total effect relationship between condition
and enjoyment is in parentheses. *p < .05, **p < .01, ***p < .001.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 30

However, the indirect effect from a single mediation model with perceptions of repetition
mediating the effect of condition on enjoyment was significant, Indirect Effect = -0.16, SE =
.077; 95% CIbootstrapping [-0.33, -0.03], see Fig. 10, suggesting that the effect of condition on lesser
decreases in enjoyment or slower hedonic adaptation was explained by lowered perceptions of
repetition (see Fig. 10 below for standardized regression coefficients of the model). This indirect
effect was tested using percentile bootstrap estimation approach with 10000 samples (Shrout &
Bolger, 2002), implemented with SPSS PROCESS Model 4 (Hayes, 2017).

Figure 10. Perceptions of repetition mediation model in sushi game. Standardized regression
coefficients for the relationship between condition and enjoyment as mediated by perceptions of
repetition. The standardized regression coefficient representing the total effect relationship
between condition and enjoyment is in parentheses. *p < .05, **p < .01.

Exploratory analyses.
Correlations between measures. Pearson's product-moment correlations were computed
with all measures (see Table 1 below for correlations and, for reference, Table 2 below for
descriptive data).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 31

Table 1
Pearson Correlation Matrix of Measures in Experiment 1

Note. T1: Measurement at Time 1 (Baseline measure after 2 minutes of Sushi Game repetition).
T2: Measurement at Time 2 (Post intervention measure after 10 minutes of Sushi Game
repetition). Habit RT: Reaction times on cognitive association task (Lower scores, stronger
habits). Automaticity: Self-report perceived behavior automaticity index (Higher scores,
increased felt automaticity). Enjoyment: Higher scores, increased reported enjoyment.
Motivation: Higher scores, increased reported motivation. Boredom: Higher scores, increased
reported boredom. Repetitiveness: Higher scores, increased perceived repetitiveness of Sushi
Game (e.g., Playing the sushi game felt like the same thing over and over). Attention: Higher
scores, increased reported attention to details of Sushi Game (e.g., I really noticed the different
ways I moved the ingredients when making sushi). Variety: Higher scores, increased perceived
variety in repeating the sushi game (e.g., There was a lot of variety in the sushi game.). Willing
to pay: “ If you were paid to continue playing this game for 30 more minutes, how much money
would you request from us? From 0 - 100 in do l l a rs . ”
*p < 0.05. **p < 0.01
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1. T1 Habit RT − -.12 -.07 .04 -.07 .52** .0 .02 .01 .04 -.13 .03 .09 .01
2. T1 Automaticity

− .21* .03 .01 -.16 .65** -.02 .11 -.12 -.01 -.02 -.01 .04
3. T1 Enjoyment

− -.55* .48* -.22** .33* .6* -.36** .29* .01 -.13 -.09 .07
4. T1 Boredom

− -.55** .14 -.11 -.52** .73** -.44** .24** .05 -.11 .18*
5. T1 Motivation

− -.12 .15 .48** -.52** .63 -.14 .07 .14 -.17
6. T2 Habit RT

− -.14 -.16 .11 -.07 -.01 -.02 -.02 .04
7. T2 Automaticity

− .19* .03 .07 -.05 .0 .03 -.04
8. T2 Enjoyment

− -.58** .53** -.21* .02 .15 -.13
9. T2 Boredom

− -.71** .27** -.13 -.25** .26**
10. T2 Motivation

− -.32** .2* .33* -.21*
11. Repetitiveness

− -.2* -.71** .1
12. Attention

− .83** -.03
13. Variety

− -.08
14. Willing to Play

−
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 32

Table 2
Means and Standard Deviations for Dependent Variables from Experiment 1
Control Group (n = 65) Novelty Group (n = 70)
Statistic Time 1 Time 2 Time 1 Time 2
Enjoyment
M
SD
5.29
1.21
4.43
1.58
5.69
1.23
5.26
1.42
Motivation
M
SD
4.68
1.32
4.18
1.43
4.49
1.43
4.14
1.57
Boredom
M
SD
3.78
1.57
4.43
1.71
3.51
1.41
4.41
1.44
Perceived Automaticity
M
SD
4.06
1.39
5.05
1.25
3.93
1.2
5.21
1.23
Reaction Time in ms
M
SD
820
193
633
137
878
187
661
155
Perception of Repetition (only Time 2)
M
SD
5.56
0.92
4.99
1.17
Attention to Task (only Time 2)
M
SD
4.78
0.94
4.5
1.12
Perceived Variety (only Time 2)
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 33

M
SD
3.78
0.69
3.86
0.93
Pay Requested to Continue Playing in dollars (only Time 2)
M
SD
30.13
30.57
33.16
34.22

Note. Time 1: Baseline measure after 2 minutes of Sushi Game repetition. Time 2: Post
intervention measure after 10 minutes of Sushi Game repetition. Enjoyment: Higher scores,
increased reported enjoyment; scale 1-7. Motivation: Higher scores, increased reported
motivation; scale 1-7. Boredom: Higher scores, increased reported boredom; scale 1-7. Perceived
Automaticity: Self-report perceived behavior automaticity index (Higher scores, increased felt
automaticity); scale 1-7. Reaction Time: Average reaction times from habit strength cognitive
association task in milliseconds (Lower scores, stronger habits). Perceptions of Repetition:
Higher scores, increased perceived repetitiveness of Sushi Game (e.g., Playing the sushi game
felt like the same thing over and over); scale 1-7. Attention to Task: Higher scores, increased
reported attention to details of Sushi Game (e.g., I really noticed the different ways I moved the
ingredients when making sushi); scale 1-7. Perceived Variety: Higher scores, increased
perceived variety in repeating the sushi game (e.g., There was a lot of variety in the sushi game.);
scale 1-7. Pay Requested to Continue Playing: “ If you were paid to continue playing this game
for 30 more minutes, how much money would you request from us? From 0 - 100 in dollars.

Correlations between habit and enjoyment measures. Habit strength, measured via
reaction times, was uncorrelated with enjoyment at either time points in the study. However,
perceived automaticity was modestly correlated with enjoyment at both time points, Time 1,
r(135) = .209, p = .015, Time 2, r(135) = .188, p = .029. This shows that, although stronger
cognitive associations were formed and enjoyment declined with repetition, these effects were
unrelated. However, higher enjoyment was related to higher perceived automaticity, which
suggests a possibility that self-reports of felt automaticity may be drawing on the same
information as judgments of feeling —if it feels good, it feels automatic (e.g., feelings-as-
information framework, Schwarz, 1990). This possibility will be discussed in the general
discussion.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 34

Decreasing motivation with repetition. A Condition (between-subjects factor: Novelty,
Control) x Time (within-subjects factor: motivation at Time 1, motivation at Time 2) repeated
measures ANOVA on motivation revealed a main effect of time such that repeating the sushi
game led to significantly declining motivation between Time 1 (M = 4.58, SD = 0.12) and Time
2 (M = 4.16, SD = 0.14), F(1, 133) = 13.96, p < .001, partial η2 = .095. There was no main effect
of Condition, F(1, 133) = .249, p > .25, partial η2 = .0, nor a significant interaction, F(1, 133) =
.447, p > .25, partial η2 = .0, suggesting that the novelty intervention did not affect motivation.
Increasing boredom with repetition. A Condition (between-subjects factor: Novelty,
Control) x Time (within-subjects factor: boredom at Time 1, boredom at Time 2) repeated
measures ANOVA on boredom revealed repeating the sushi game significantly increased
boredom between Time 1 (M = 3.65, SD = 0.13) and Time 2 (M = 4.42, SD = 0.4), F(1, 133) =
64.11, p < .001, partial η2 = .325. There was no effect of condition, F(1, 133) = 0.34, p > .25,
partial η2 = .0, nor a significant interaction, F(1, 133) = 1.73, p = .191, partial η2 = .013,
suggesting that the novelty intervention did not affect boredom.
No difference in retrospective judgments of variety. An independent-samples t test on
pa rt i c i pa nt s ’ retrospective judgments of variety in the sushi game yielded no effects, t(133) =
.54, p > .25, suggesting that intervention did not change perceived variety (e.g., “There was a lot
of variety in the sushi ga m e ” ).
No difference in willingness to continue playing. An independent-samples t test on how
much money p a rt i c i pa n t s ’ requested to continue playing the sushi game for 30 more minutes at
the end of the study revealed no differences (Novelty group average request: $30.13, Control
group average request: $33.16), t(133) = .54, p = .18, suggesting that the intervention did not
cause people to want to continue playing the sushi game.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 35

Discussion
This study accomplished three main objectives. First, it demonstrated that introducing
novelty by slightly modifying how people perform and attend to a repetitive task can slow
hedonic adaptation. The intervention caused participants to sustain higher enjoyment as they
repeated the task. Second, it demonstrated that an intervention of this kind did not inhibit habit
formation at a cognitive or experiential level, suggesting that the trajectory of hedonic adaptation
can be slowed without influencing habit formation. Participants in the novelty group increased
habit strength and enjoyed the experience more than control group participants. Third, although
the results did not support the double mediation hypothesis, the results did provide process
evidence linking the effect of the intervention to one of the proposed mechanisms, decreased
perception of repetition. Slightly varying performance and attention made the task seem less
repetitive and consequently more enjoyable.
Before accepting this account, it is important to note a potential confound. It is possible
that the slower hedonic adaptation of participants in the novelty group was observed because
they repeated the sushi game sequence more slowly and thus fewer times than control
participants. If the intervention caused participants to repeat the sequence less often, then this
lesser practice could be responsible for lowered perceptions of repetition and consequently
higher enjoyment. Although the number of repetitions was not assessed in the first study, such an
account is inconsistent with the findings for habit strength. Fewer repetitions would have likely
resulted in weaker cognitive associations and thus lead to slower responses in the reaction time
test. However, participants in the novelty group had similar reaction time scores compared with
control participants. Nonetheless, Experiment 2 addressed this possibility by tracking the exact
amount of repetitions.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 36

The second experiment also addressed a potential criticism of the RT habit strength
measure. Ideally, the mapping between the perception of the cue and the behavioral response
would be tested a way as close as possible to the actual task. Habit representations in the mind
are multimodal memory representations in which the cue can be presented via different
modalities (e.g., visual, auditory, tactile, gustatory, interoceptive, Wood & Rünger, 2016).
Perception of the cue then leads to a behavioral response. The habit strength RT measure in
Study 1 only assessed thought of response, but not the actual behavior. Experiment 2’s habit
strength measures will test associations by perceiving a cue and responding behaviorally to that
cue.
Thus, Experiment 2 was designed to address the potential confound with number of
repetitions, to improve the internal validity of the habit strength measure, and to further test the
robustness of this novelty intervention on hedonic adaptation and habit formation.
Experiment 2: Repeating a Piano Melody
In Experiment 2, participants repeated a piano melody with a virtual piano using their
computer. Some participants were instructed to vary how they pressed the keys and where to
focus their attention when playing, whereas others repeated the sequence with control
instructions. We hypothesized that introducing performance and attention novelty would make
repeating the melody more enjoyable, focus more attention on the task, feel less repetitive, and,
at the same time, not impede habit formation.
Method
Participants. Participants (N = 141; 52% female; Mage = 35.97, SDage = 10.92) were
recruited on A m a z on.c o m ’s M-Turk for nominal payment. Given that it was important for
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 37

participants to attend to the task without distraction, participants were asked at the end of the
study whether they were multitasking during the experiment. Four participants answered
affirmatively and were thus excluded from the analyses.
Procedure. Participants signed up for a study on testing a virtual piano. They were
instructed to make the browser full size and ensure that they could hear the sound of the piano.
The virtual piano operated like a real piano. Participants placed their fingers on keyboard keys
that corresponded with unique piano sounds (see Fig. 11 below and Fig. 12 below).

Figure 11. Instructions for where to place fingers to play the virtual piano.

To play, participants pressed numerical keyboard keys and received visual and auditory
feedback (see Fig. 12 below). The piano provided immediate feedback about the success of
pa rt i c i pa nt s ’ keyboard press through three indicators: the corresponding number that matched the
melody, the ability to continue to play the next note in the melody sequence, and the blue
highlighted color on the keyboard —incorrect responses generated a red highlight. Participants
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 38

could only progress through the melody if they pressed the correct key. All participants played
and repeated the same seven note melody throughout the experiment (3-7-4-8-9-1-2, this melody
was pre-tested for difficulty and how pleasant it sounded).

Figure 12. Feedback when participant pressed the correct number on the keyboard (left) and
incorrect (right). Hands with assigned fingers reminded participants to only use those fingers.

The structure of the experiment resembled the sushi game procedure in Experiment 1 (see
Fig. 13 below). In the first phase, all participants practiced the melody twice untimed to get
acquainted with the game. Then, they played the melody as many times as they possible in 90
seconds. As shown in Figure 12 above, participants would see the seven-note melody sequence
throughout the 90 seconds, and a counter in the upper right corner of the screen counted down
from 90 to 0. After 90 seconds, participants completed the cue-response association test to
measure habit strength (see below). Then, they completed Time 1 self-report measures of
enjoyment, perceived automaticity, boredom, and motivation (see below).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 39

Figure 13. Experiment 2 procedure and sequence of tasks and measures.

In the second phase of the experiment, all participants repeated the melody for 90
seconds in four separate trials. Before each trial, control participants were instructed to “ c ont i nu e
to play as you have been playing the piano.” Participants in the novelty group received a
different instruction before each 90 second trial: “ pr e s s the keys delicately and focus on how
your fingers feel when you do t hi s ,” “ pre s s the keys firmly and focus on how your fingers feel
when you do t hi s ,” “ pre s s the keys softly and focus on how your fingers feel when you do t h i s ,”
and “ pre s s the keys precisely and focus on how your fingers feel when you do t hi s .”
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 40

At the end of the four trials, all participants repeated the same habit strength test and
similar self-report measures as in the first phase of the experiment.
Measures.
Habit strength. A tailored reaction time test was developed to measure habit strength of
the piano melody. As participants repeated the melody, stronger cue-response associations
between the steps of the melody sequence should be formed. Critically, participants played each
of the keyboard keys of the virtual piano with the same fingers (e.g, to play key 7, use right index
finger) —this ensured that all participants learned the same cue-response associations. In this
context, a unique “cue ” can be operationalized a number from the melody (e.g., “ 7 ” ), the finger
assigned to that number (e.g., right index finger is assigned to “ 7” ), or the sound heard when that
key is pressed (e.g., sound of “ 7”). A unique “ r e s pons e ” can be operationalized as the next
number in the sequence (“ 4” comes after “ 7” in the sequence), the next finger pressed in the
sequence (left index finger assigned to “ 4” is pressed after the right index finger press of “ 7” ),
and the next sound heard in the sequence (sound of “ 4” comes after the sound of “ 7” ).
Given these different cue-response association, participants are learning many
overlapping associations (number → finger, number → sound, number → next number in the
sequence, finger → next finger press, etc.). When testing the habit strength of the melody, we
chose to structure the test so that participants would perform one part of the sequence, play note
from the melody (cue), and then perform the next part of the sequence, play the next note in the
melody (response), as quickly as possible. If participants developed stronger cue-response
associations in memory of the melody sequence, they should press the next note of the melody
more quickly with more practice because it is more cognitively accessible (Wood & Rünger,
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 41

2016). The measure had the following procedure (see visual summary of procedure in Fig. 14
below):
1. A fixation cross was presented (2 seconds)
2. A random number from the melody sequence was presented for two seconds (e.g.,
“7 ”)
3. Participants pressed the corresponding keyboard key with the correct finger (e.g.,
they would press the “7 ” keyboard key with right index finger). They also heard
the sound of the number. If participants did not press the number within two
seconds, they would be presented with a “ T O O S L O W ” and a new trial would
begin (see Fig. 15 below). Or, if they pressed an incorrect key, a new trial would
begin (see Figure in Appendix E).
4. Then, participants had three seconds to press the number the correct subsequent
number from the melody sequence as quickly as possible (indicated by an arrow
and question mark, e.g., 7 → ?, see Fig. 14 below).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 42

Figure 14. Example of correct reaction time trial in the piano habit strength cue-response
association test.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 43

Figure 15. Example of incorrect “ T oo S l ow ” trial.

Participants performed twelve reaction time trials (given the seven part melody sequence,
3-7-4-8-9-1-2, each part was randomly presented two times, with the exception of the last part of
the sequence, “ 2,” which was not presented since there is no part that comes after “ 2” ). Key press
responses were recorded in milliseconds. Strength of cue-response associations was assessed
from mean reaction times (RT) to the next number prompt “ → ?” for all correct trials (Average
percentage of correct trials for novelty group participants was 83% at Time 1 and 93% at Time 2.
Average percentage of correct trials for control group participants was 83% at Time 1 and 94%
at Time 2. No group differences in number of correct trials were observed at Time 1, t(139) =
.14, p > .25, and Time 2, t(139) = -.45, p > .25.).
Enjoyment. Participants rated their agreement with the statement, “I am enjoying playing
the melody on the piano, ” from 1 (strongly disagree) to 7 (strongly agree).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 44

Motivation. Participants rated their agreement with the statement, “I am motivated to
play the melody on the piano, ” from 1 (strongly disagree) to 7 (strongly agree).
Boredom. Participants rated their agreement with the statement, “Playing the melody on
the piano is boring, ” from 1 (strongly disagree) to 7 (strongly agree).
Perceived behavioral automaticity. Following Study 1 (Gardner et al., 2012), participants
indicated from 1 (strongly disagree) to 7 (strongly agree), “ E a c h step in the melody is
s om e t hi ng...” “I do a ut o m a t i c a l l y; ” “I do without having to consciously remember; ” “I do
without t hi nk i ng; ” and “I start doing before I realize I’m doing i t . ” Reliability was good (alpha =
.90).
Declarative knowledge of melody. Participants were asked directly “ W ha t was the
melody sequence? It is ok if you don't know the exact sequence. Type what you know. (e.g.,
1234567890).”
Perception of repetition. Perception of repetition was measured as in Experiment 1:
redundancy (“Playing the melody on the piano each 90 second round felt like the same thing
over and ove r” ) and similarity ( “ E a c h 90 second round playing the melody on the piano seemed
very similar to each ot he r, ” “ E a c h 90 second round playing the melody on the piano had aspects
that made it di ffe re n t ,” reverse coded). Reliability was good (alpha = .77).
Attention to variety within task. Attention to variety was measured as in Experiment 1:
discriminability (“ D i ff e re nt ways to experience playing the melody on the piano during each 90
second round were obvi ous ,” “I could identify different ways to play the melody on the piano
during each 90 second round” ) and salience ( “ I really noticed the different ways I could play the
melody on the piano during each 90 second round,” “I did not pay much attention to the different
ways I played the melody on the piano during each 90 second round” reverse coded).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 45

Participants indicated agreement to items from 1 (strongly disagree) to 7 (strongly agree).
Reliability was good (alpha = .88).
Retrospective judgments of variety. Retrospective judgments of variety were measured
as in Experiment 1 ( “ T he re were many different ways to play the melody on the piano during
each 90 second round” , “ T he r e was a lot of variety in playing the melody on the piano during
each 90 second round.”). Participants indicated agreement to items from 1 (strongly disagree) to
7 (strongly agree). Reliability was good (alpha = .86).
Results from principal components analysis on all items from the above three indexes
—perceptions of repetition, attention to distinct aspects of task, and retrospective judgments of
variety —revealed the same three factor structure with items loading highest on the intended
factors (see analysis in Appendix B).
Number of melody repetitions per trial. The program tallied the number of times a
participant completed the full seven-note melody per each 90 second trial.
Behavioral identification. To investigate whether the intervention caused participants to
think more concretely vs. abstractly about the task, five items were adapted from Vallacher and
W e gne r’s (1989) behavioral identification scale. (Similar adaptations were used in past research,
see Labrecque, 2015). Participants chose between two descriptions of an action (e.g., Making a
list: Getting organized vs. Writing things down). Analysis was conducted on the target item:
“ P l a yi ng the piano melody: Learning to play the piano vs. Pressing keyboard keys .”
Willingness to continue playing. As a proxy for measuring pa rt i c i pa nt s ’ desire to
continue repeating the piano game, participants reported, “ If you were paid to continue playing
this game for 10 more minutes, how much money would you request from us? From 0 - 10 in
dol l a rs .”
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 46

Piano playing experience. Participants indicated their level of piano playing experience
(Never learned to play piano, Beginner, Average, Advanced).
MTurk attention check questions. To understand MTurk pa rt i c i pa n t s ’ environment,
questions were asked about how quiet their environment was, the device used (desktop
computer, laptop), and their level of distraction.
Results
First, I present statistical tests for the hypotheses. Second, I present exploratory analyses.
Means and standard deviations for all dependent variables in Experiment 2 are shown in
Appendix D.
Increasing habit strength (hypothesis 1a). A Condition (between-subjects factor:
Novelty, Control) x Time (within-subjects factor: Time 1 vs. Time 2) repeated measures
ANOVA on reaction times in the cue-response habit strength test revealed a main effect of time
such that repeating the piano melody led to significantly faster reaction times (in milliseconds) at
Time 2 (M = 637, SD = 23) vs. Time 1 (M = 1022, SD = 20), F(1, 139) = 209.51, p < .001,
partial η2 = .601 (see Fig. 15 below).6 This suggests that repeating the melody did indeed cause
stronger habits to form.
Unexpectedly, there was a main effect of condition such that participants in the control
group had faster reaction times (M = 787, SD = 24) compared with novelty group participants (M
= 873, SD = 23), F(1, 139) = 6.8, p = .01, partial η2 = .047, but no significant interaction, F(1,
139) = 1.74, p = .189, partial η2 = .012. To better understand the condition effect, an ANCOVA

6 Reaction time data were also analyzed after being log transformed. Results were the same, so raw scores are
presented for ease of interpretation. Also, as recommended from research on analyzing RT data, a cut off of 200 ms
was used when including responses for analysis (Whelan, 2008).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 47

was conducted to test group differences in Time 2 while controlling for variation in Time 1.
ANCOVA results showed that participants in the novelty group indeed had slower Time 2 RT
scores than control participants, after adjusting for Time 1 RT scores, Mdiff = 108 ms, F(1, 138) =
5.94, p = .016, partial η2 = .041. This suggests that the intervention unexpectedly slowed habit
formation.

Figure 15. Reaction times in habit strength test in piano game.

One plausible explanation is that the novelty intervention caused participants to repeat the
melody fewer times than control group participants. Since the frequency of repetition increases
habit strength, participants who repeated the melody fewer times would be expected to have
slower reaction times on the habit strength test.
Table 3 and Figure 16 below show descriptive data for the number of times participants
completed the full seven-note melody per each 90 second trial. As expected, there were no group
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 48

differences in the number of melody repetitions at baseline (Time 1), before the intervention.
However, there were significant differences between groups for all 90 second trials and for total
completed trials, with control group participants repeating more total melody repetitions than
novelty group participants.

Table 3
Descriptive Statistics for Number of Melody Repetitions Per 90 Second Trial in Experiment 2
Control group Novelty Group
Statistic Mean # of Melody Repetitions Mean # of Melody Repetitions
1st 90 second trial (before intervention, baseline)
M
SD
21.81
8.05
20.29
8.69
2nd 90 second trial**
M
SD
27.82
10.25
23.09
8.00
3rd 90 second trial**
M
SD
30.16
9.56
25.78
8.39
4th 90 second trial**
M
SD
30.95
10.44
25.14
9.59
5th 90 second trial*
M
SD
29.55
10.44
26.09
9.0
Total completed trials**
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 49

M
SD
140.3
43.84
120.39
35.74

Note. Significant difference in repetitions between groups, *p < .05. **p < .01

Figure 16. Number of completed melody repetitions per 90 second trial.

If the number of melody repetitions influenced the significant differences in reaction
times between groups, then adding the total number of repetitions as a covariate to the ANCOVA
would assess whether group differences in reaction times would remain after controlling for
melody repetitions and baseline Time 1 RT differences. Results from the ANCOVA showed that
participants in the novelty group still had slower Time 2 RT scores than control participants,
after controlling for total melody repetitions and Time 1 RT scores, Mdiff = 110 ms, F(1, 137) =
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 50

5.77, p = .018, partial η2 = .04. These results suggest that the intervention caused participants in
the novelty group to form slightly weaker cognitive associations in spite how often they repeated
the piano melody. Notably, the intervention did not prevent habits from forming —novelty
participants still developed stronger cognitive associations with repetition —but the performance
and attention instructions somehow slowed habit formation. Possible explanations are provided
in the discussion.
Increasing perceived behavioral automaticity (hypothesis 1b). A Condition (between-
subjects factor: Novelty, Control) X Time (within-subjects factor: Time 1 vs.Time 2) repeated
measures ANOVA on perceived automaticity revealed a main effect of Time such that repeating
the piano melody led to significantly higher automaticity at Time 2 (M = 4.68, SD = 1.47) vs.
Time 1 (M = 3.53, SD = 1.46), F(1, 139) = 133.6, p < .001, partial η2 = .49 (see Fig. 17 below).
There was no main effect of condition, F(1, 139) = 0.00, p > .25, partial η2 = .0, but there was a
significant interaction, F(1, 139) = 3.24, p = .033, partial η2 = .032, suggesting that the increases
in perceived behavioral automaticity between Time 1 and Time 2 differed between groups. For
the novelty group, perceived behavioral automaticity increased less between Time 2 and Time 1
(Mdiff = 0.94, SE = 0.13, p < .001) compared with the control group (Mdiff = 1.37, SE = 0.15, p <
.001).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 51

Figure 17. Ratings of perceived behavioral automaticity in piano game.

Before drawing conclusions from this interaction effect, I conducted an ANCOVA to
discern whether the interaction is being driven, in part, by random differences in perceived
automaticity ratings at Time 1. Results from the ANCOVA showed that participants in the
control group had marginally higher ratings of perceived automaticity at Time 2 than novelty
participants, after controlling for Time 1 ratings, Mdiff = .36, F(1, 138) = 3.78, p = .054, partial η2
= .027.
Perhaps like the RT scores, automaticity ratings were influenced by the number of
melody repetitions: more repetitions, higher perceived automaticity. Suggesting that this was
true, results from an ANCOVA controlling for total melody repetitions and Time 1 perceived
automaticity ratings, showed no differences in ratings of perceived automaticity in groups at
Time 2, Mdiff = .32, F(1, 137) = 2.73, p = .101, partial η2 = .02.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 52

Taken together, these results suggests perceived automaticity increased in both groups
with repetition; pa rt i c i p a nt s ’ automaticity ratings in the novelty group, however, increased
slightly less due, in part, to fewer melody repetitions than control participants.
Correlations between habit measures (hypothesis 1c). Echoing Experiment 1, reaction
times from the cognitive association test and ratings of perceived behavioral automaticity were
essentially uncorrelated at Time 1, r(141) = -.15, p = .076. However, they revealed a significant
correlation at Time 2, r(141) = -.236, p = .005 (the correlation is negative because higher
automaticity is correlated with lower reaction times on the cognitive association test). This result
shows that participants who increased habit strength in terms of stronger cognitive associations
also increased in perceived automaticity, providing some support for the idea that stronger
cognitive associations and increases in perceived behavioral automaticity are related.
Slowing hedonic adaptation (hypothesis 2a). A Condition (between-subjects factor:
Novelty, Control) X Time (within-subjects factor: enjoyment at Time 1, enjoyment at Time 2)
Repeated Measures ANOVA on enjoyment yielded a main effect of time such that repeating the
piano melody led to significantly lower enjoyment ratings at Time 2 (M = 4.95, SD = 1.75) vs.
Time 1 (M = 5.52, SD = 1.43), F(1, 139) = 21.56, p < 0.001, partial η2 = .134 (see Fig. 18
below). A main effect of condition reflected that novelty group participants enjoyed playing the
piano melody more (M = 5.46, SD = 0.16) than control group participants (M = 4.99, SD = 0.17),
F(1, 139) = 3.92, p = .05, partial η2 = .027. The predicted Condition x Time interaction on
enjoyment was marginally significant, F(1, 139) = 3.92, p = .062, partial η2 = .025, such that
enjoyment ratings decreased slightly less between Time 1 and Time 2 for the novelty group
(Time 1 M = 5.63, SD = 1.34, Time 2 M = 5.29, SD = 1.5) vs. control group (Time 1 M = 5.4, SD
= 1.52, Time 2 M = 4.59, SD = 1.92).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 53

Figure 18. Enjoyment ratings in sushi game.

To control for random variation in enjoyment ratings at Time 1, an ANCOVA on
enjoyment ratings at Time 2 revealed stronger support for the predicted slowing of hedonic
decline. That is, Time 2 enjoyment ratings were significantly greater in the novelty group vs.
control group after controlling for Time 1 enjoyment, Mdiff = 0.54, F(1, 138) = 5.1, p = .025,
partial η2 = .036.
Overall, these results provide modest support for the hypothesis that the novelty
intervention would sustain higher enjoyment with repetition and thus slow hedonic adaptation.
Novelty intervention increases attention to variety in the task (hypothesis 2b). An
independent-samples t test on attention to variety in the task revealed that the novelty group
reported attending to the variety within the piano game more (M = 4.49, SD = 1.12) vs. control
participants (M = 3.9, SD = 1.49) , t(139) = 2.54, p = .012). The novel performance and attention
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 54

instructions thus caused a discovery of novel elements within the repetitive experience (e.g., “I
really noticed the different ways I could play the melody on the piano during each 90 second
round” ).
Novelty intervention lowers perception of repetition (hypothesis 2c). An independent-
samples t test on perceptions of repetition revealed that participants in the novelty group
perceived playing the piano melody as less repetitive (M = 5.11, SD = 1.08) than did control
participants (M = 6.04, SD = 1.08) , t(139) = -5.1, p < .001, see Fig. 19 below). Thus, the novel
performance and attention instructions decreased pa r t i c i p a nt s ’ perception of repetition.

Figure 19. Perceptions of repetition ratings of piano game.

Double mediation: Intervention increased attention to distinct aspects of task and
decreased perceptions of repetition, which, in turn, slowed hedonic adaptation (hypothesis
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 55

2d). To investigate whether the effect of the intervention on slowing hedonic adaptation (i.e.,
lesser decrease in enjoyment between Time 1 and Time 2) was driven by corresponding
differences in attention to the task and perceptions of repetition, a mediation analysis was
conducted using condition (Novelty, Control instructions) as the independent variable, change in
enjoyment as the dependent variable, and attention to distinct aspects in task as the proximal
mediator and perception of repetition as the distal mediator. The indirect effect was tested using
percentile bootstrap estimation approach with 10000 samples (Shrout & Bolger, 2002),
implemented with SPSS PROCESS Model 6 (Hayes, 2017). The results confirmed that the
indirect effect of condition on enjoyment, via attention to task and perceptions of repetition, was
significant, Indirect Effect = -.06, SE = .04, 95% CIbootstrapping [-0.15, -0.01], suggesting that the
intervention focused attention on distinct aspects of the task, making the task seem less repetitive
and consequently more enjoyable (see Fig. 20 below for standardized regression coefficients of
the model).7

Figure 20. Double mediation model in piano study. Standardized regression coefficients for the
relationship between condition and enjoyment as mediated by attention to distinct aspects of task

7 The results also replicated the single mediation observed in Experiment 1 in which the indirect effect of condition
on enjoyment, via perceptions of repetition, was significant, Indirect Effect = -.35, SE = .13, 95% CIbootstrapping [-0.64,
-0.15], The indirect effect was tested using percentile bootstrap estimation approach with 10000 samples (Shrout &
Bolger, 2002), implemented with SPSS PROCESS Model 4 (Hayes, 2017).
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 56

and perception of repetition. The standardized regression coefficient representing the total
relationship between condition and enjoyment is in parentheses. *p < .05, **p < .01, ***p <
.001.
Exploratory analyses.
Correlations between measures. Pearson's product-moment correlations were conducted
for all measures (see Table 4 below and, for reference, Table 5 below for descriptive data).
Table 4
Pearson Correlation Matrix of Measures from Experiment 2

Note. T1: Measurement at Time 1 (Baseline measure after 90 seconds of playing piano melody).
T2: Measurement at Time 2 (Post intervention measure after 7.5 minutes of playing piano
melody). Habit RT: Reaction times on cognitive association task (Lower scores, stronger habits).
Automaticity: Self-report perceived behavior automaticity index (Higher scores, increased felt
automaticity). Enjoyment: Higher scores, increased reported enjoyment. Motivation: Higher
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1. T1 Habit RT − -.08 -11 .04 -.15 .24** -.17* .0 -.1 .07 -.1 -.58** .13 -.02
2. T1 Automaticity

− .4** .37** .4** -.15 .67** .26** -.17* .25** -.03 .19* -.01 .31**
3. T1 Enjoyment

− -.76** .69** -.13 .33** .58** -.51** .5** -.1 .14 -.04 .26**
4. T1 Boredom

− -.71** .09 -.29** -.61** .61** -.59** .2* -.09 .08 -.24**
5. T1 Motivation

− -.15 .28** .55** -.43** .64** -.12 .19* -.1 .32**
6. T2 Habit RT

− -.24** .0 -.04 -.03 -.05 -.16 .0 .02
7. T2 Automaticity

− .26** -.2* .23* .09 .21* -.09 .17*
8. T2 Enjoyment

− -.85** .82** -.35* .04 -.25** .35*
9. T2 Boredom

− -.77** .35** .13 .25** -.29
10. T2 Motivation

− -.31** .02 -.26** .35**
11. Repetitiveness

− .21* .12 -.42**
12. Melody Reps

− -.07 .0
13. Willing to Play

− -.06
14. Attention

-
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 57

scores, increased reported motivation. Boredom: Higher scores, increased reported boredom.
Repetitiveness: Higher scores, increased perceived repetitiveness of playing piano melody (e.g.,
Playing the piano melody felt like the same thing over and over). Melody Reps: Average number
of completed melody repetitions. Willing to Play: “ If you were paid to continue playing this
game for 10 more minutes, how much money would you request from us? From 0 - 10 in
dol l a rs .” Attention: Higher scores, increased reported attention to details of Piano Game (e.g., I
really noticed the different ways to play the melody).
*p < 0.05. **p < 0.01

Table 5

Means and Standard Deviations for Dependent Variables from Experiment 2
Control Group (n = 68) Novelty Group (n = 73)
Statistic Time 1 Time 2 Time 1 Time 2
Enjoyment
M
SD
5.4
1.52
4.59
1.92
5.63
1.34
5.29
1.5
Motivation
M
SD
5.49
1.45
4.85
1.81
5.68
1.42
5.45
1.35
Boredom
M
SD
2.91
1.8
3.99
2.19
2.58
1.35
3.18
1.68
Perceived Automaticity
M
SD
3.42
1.47
4.79
1.34
3.64
1.46
4.58
1.59
Reaction Time in ms
M
SD
996
232
576
233
1048
236
697
297
Perception of Repetition (only Time 2)
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 58

M
SD
6.04
1.08
5.11
1.08
Attention to Task (only Time 2)
M
SD
3.79
1.61
4.11
1.58
Perceived Variety (only Time 2)
M
SD
3.79
1.85
4.18
1.65
Pay Requested to Continue Playing in dollars (only Time 2)
M
SD
4.71
3.44
5.08
3.18

Note. Time 1: Baseline measure after 90 seconds of repeating piano melody. Time 2: Post
intervention measure after 7.5 minutes of piano melody repetition. Enjoyment: Higher scores,
increased reported enjoyment; scale 1-7. Motivation: Higher scores, increased reported
motivation; scale 1-7. Boredom: Higher scores, increased reported boredom; scale 1-7. Perceived
Automaticity: Self-report perceived behavior automaticity index (Higher scores, increased felt
automaticity); scale 1-7. Reaction Time: Average reaction times from habit strength cognitive
association task in milliseconds (Lower scores, stronger habits). Perception of Repetition: Higher
scores, increased perceived repetitiveness of piano melody (e.g., Playing the melody on the piano
each 90 second round felt like the same thing over and over); scale 1-7. Attention to Task: Higher
scores, increased reported attention to details of piano melody (e.g., I really noticed the different
ways I could play the melody); scale 1-7. Perceived Variety: Higher scores, increased perceived
variety in repeating the piano melody (e.g., There were many different ways to play the melody
on the piano during each 90 second round); scale 1-7. Pay Requested to Continue Playing: “ If
you were paid to continue playing this game for 10 more minutes, how much money would you
request from us? From 0 - 10 in dollars.

Correlations between habit and enjoyment measures. Habit strength, measured via
reaction times, was uncorrelated with enjoyment at either time points in the study. However,
perceived automaticity was positively correlated with enjoyment at both time points, Time 1,
r(135) = .21, p = .015, Time 2, r(135) = .19, p = .029. These results replicate the results from
Experiment 1. An explanation of these effects is presented in the general discussion.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 59

Intervention slowed decrease in motivation with repetition. A Condition (between-
subjects factor: Novelty, Control) x Time (within-subjects factor: Time 1 vsTime 2) Repeated
Measures ANOVA on motivation revealed a main effect of time such that repeating the piano
game led to significantly declining motivation between Time 1 (M = 5.59, SD = 0.12) and Time
2 (M = 5.15, SD = 0.14), F(1, 139) = 15.2, p < .001, partial η2 = .099. There was no main effect
of condition, F(1, 139) = 2.9, p = .091, partial η2 = .02, nor a significant interaction, F(1, 139) =
3.24, p = .074, partial η2 = .023. However, after controlling for random variation in motivation at
Time 1, an ANCOVA showed that novelty group participants were more motivated to play the
piano melody at Time 2 compared with control participants, Mdiff = 0.45, F(1, 139) = 4.48, p =
.036, partial η2 = .031, suggesting that the intervention prevented motivation from declining for
the novelty group as much as the control group (see means and standard deviations in Appendix
D), which contrasts with the results from Experiment 1 (no group differences were observed in
motivation change).
Intervention slowed decrease in boredom with repetition. A Condition (between-
subjects factor: Novelty, Control) x Time (within-subjects factor: Time 1 vs. Time 2) repeated
measures ANOVA on boredom revealed a main effect of Time such that repeating the piano
game led to significantly greater boredom between Time 1 (M = 2.74, SD = 0.13) and Time 2 (M
= 3.58, SD = 0.16), F(1, 139) = 38.18, p < .001, partial η2 = .22. However, in contrast to
Experiment 1 results, a main effect of condition showed that participants in the control group
were more bored (M = 3.5, SD = .19) than the novelty group participants (M = 2.88, SD = .19),
F(1, 139) = 4.62, p = .033, partial η2 = .03. There was no significant interaction, F(1, 139) =
3.01, p = .085, partial η2 = .02. To ensure that the main effect of condition was not driven by
random variation in boredom at Time 1, an ANCOVA was conducted and showed that novelty
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 60

group participants were indeed less bored with playing the piano melody at Time 2 compared
with control participants, Mdiff = 0.56, F(1, 139) = 4.5, p = .036, partial η2 = .032, suggesting that
the intervention decreased boredom for the novelty group participants (see means and standard
deviations in Appendix D), which contrasts with the results from Experiment 1 (no group
differences were observed in boredom change).
No difference in retrospective judgments of variety. An independent-samples t test on
pa rt i c i pa nt s ’ retrospective judgments of variety in the piano game (e.g., “There was a lot of
variety in playing the melody on the piano during each 90 second round.” ) revealed no
differences, t(139) = 1.76, p = .08.
No difference in the amount of pay requested to continue playing. An independent-
samples t test on payment required to continue playing the piano melody revealed no effects
(Novelty group average request: $5.08, Control group average request: $4.71; participants could
choose from 0-10 dollars), t(133) = 0.68, p > .25, suggesting that the intervention did not cause
people to request less money to continue playing the piano melody.
No difference in behavioral identification. An independent-samples t test on level of
categorization of the piano task (Participants chose one of two options: “ L e a rn i ng to play the
piano vs. Pressing keyboard ke ys .” ) revealed no differences between groups, t(139) = -1.73, p =
.087, suggesting that the intervention did not cause people to differ in the way they categorized
the task.
Discussion
The results from Experiment 2 demonstrate the robustness of the proposed intervention to
slow hedonic adaptation via instructing people to slightly vary how they perform and attend to a
repetitive task. As in Experiment 1, the intervention made repetition more enjoyable. In contrast
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 61

to the results in Experiment 1, the intervention also made repetition less boring and curbed
decreases in motivation. Experiment 2 provides further process evidence directly linking the
effect of varying performance and attention to the proposed mechanisms, attention to distinct
aspects of the task and perceptions of repetition. The significant double mediation suggests that
the intervention caused participants to focus their attention on distinct aspects of the task, making
playing the melody seemed less repetitive, and consequently more enjoyable.
However, unlike Experiment 1, the intervention had a more complex influence on the
direct measure of habit strength, the cue-response association test. All participants developed
stronger cue-response associations of the piano melody between Time 1 and Time 2, but
participants in the novelty group showed significantly slower reaction times in the cue-response
association test compared with control group participants. Notably, the intervention did not
inhibit habit formation, but somehow slowed the rate of habit formation; that is, cue-response
associations significantly strengthened with repetition but were less strong for novelty group
participants. The likely candidate explanation for this difference was that novelty group
participants repeated the melody less often and therefore formed weaker cue-response
associations. Analyses were conducted to test this possibility, but the results showed that number
of repetitions did not account for differences in habit strength.
What are other possible explanations for the slightly weaker cue-response associations in
the novelty group? Perhaps by having participants vary how they played the melody and (e.g.,
press the keys softly) where they focused their attention (e.g., focus on how your fingers feel),
participants increased the amount they deliberated and consciously tried to learn the melody
sequence. Some research has shown that deliberation can impede habit formation because it
keeps action controlled by a model of the world (model-based learning) as opposed to stimulus
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 62

driven control (i.e., model-free/habit, Gillan, Otto, Phelps, & Daw, 2015). Research in our lab
has also found that asking people to deliberate when repeating a task (e.g., remember the steps in
the sushi game sequence) can lead to decreases in habit formation (Labrecque, Lee, & Wood,
unpub). Finally, as noted in the introduction, research has suggested that increased attention to
step-by-step control can interfere with automaticity performance (Beilock & Carr, 2001; Lewis
& Linder, 1997), but no published research showed such an effect for automaticity development.
The results from Experiment 2 point to a case in which this can happen.
It is important to note that although participants in the novelty group did not make the
same gains in habit strength compared with controls, they did enjoy repetition more, were less
bored, and were more motivated to continue repeating the piano melody. Moreover, the
intervention did not impede habit formation, rather, it slowed its progress. Thus, insofar as
intrinsic motivation is important for behavioral persistence and long-term goal achievement
(Woolley & Fishbach, 2017), the tradeoff of slightly slowing habit formation may be worth it if
it means people are more likely to repeat an action due to sustained intrinsic motivation.
General Discussion
The goal of the present research was to examine how hedonic adaptation and habit
formation are related. Specifically, I argued that hedonic adaptation poses a challenge to forming
habits since habit formation requires many repetitions; and, with repetition, one cannot ignore
the effect of hedonic adaptation, i.e., how enjoyment tends to fade in intensity over time. Based
on an examination of factors that slow the rate of hedonic adaptation, a distinction was drawn
between strategies that slow hedonic adaptation by avoiding repeating an activity —a novelty-as-
breadth approach —and strategies that slow hedonic adaptation by changing the way people
perform and attend to repeat activities —a novelty-as-depth approach. An intervention was
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 63

developed that was inspired by the latter strategy because it allows for continued repetition,
which increases the rate of habit formation. This strategy rests on the general principle that what
we attend to and perceive in the moment influences how we feel (see research on focusing
illusion, Kahneman et al., 2006; affect misattribution, Schwarz & Clore, 1983; action
identification theory, Vallacher & Wegner, 1987). The intervention was tested in two distinct
tasks and its effect on hedonic adaptation and habit formation was analyzed.
The results from both experiments support three main findings. First, participants formed
stronger habits with increased repetition, indicated by faster reaction times on the cue-response
association tests (hypothesis 1a) and higher ratings of self-reported behavioral automaticity
(hypothesis 1b). Critically, the instructions to vary performance and attention during repetition
did not prevent gains in habit strength, providing evidence that conscious attention and variation
to a repeat task does not prevent habit learning and automaticity development. Second, the
intervention did indeed slow hedonic adaptation as people repeated a sushi video game and a
piano melody (hypothesis 2a). After initial repetition, participants enjoyed both tasks; however,
after prolonged repetition only those participants who were instructed to slightly alter how they
moved their mouse in the sushi game, how they pressed the keys in the piano task, and how they
attended to those adjustments continued to enjoy the tasks at a higher level than participants told
to just continue to repeat the tasks. Finally, results from both experiments provided process
evidence that the effect of the intervention on sustaining enjoyment was driven by the proposed
mechanisms: increased attention to distinct aspects of the task (hypothesis 2b) and decreased
perceptions of repetition (hypothesis 2c). Experiment 1 supported a single mediation model
(intervention lowered perceptions of repetition and, in turn, sustained enjoyment), whereas
Experiment 2 supported the hypothesized double mediation model: the intervention focused
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 64

attention on distinct aspects of the task, making the task seem less repetitive and consequently
more enjoyable. Overall, the present experimental results support the idea that slightly varying
performance and attention can make a repeat task more enjoyable as people form habits.
Although many of the results from the experiments provide support for hypotheses posed
in the research, new insights can be gleaned from analyzing why some of the hypotheses were
not fully supported. For example, in contrast to the prediction that both habit measures would be
correlated (hypothesis 1c), why were the cognitive and self-report measures of habit sometimes
uncorrelated? I discuss how my findings bear on this topic of habit measurement in the next
section. Then, I discuss the implications of the results on hedonic adaptation research. In both
sections, I describe the limitations of the research and discuss future directions.
Implications for Research on Habit Formation
Habit is one of ps yc hol ogy’s most capacious constructs. Like emotion, behavior, or
attention, habit must be carefully defined before it is studied scientifically —without a clear
definition, measurement of habit is challenging, making the construct hard to employ when
explaining and predicting phenomena in the world. “ E ve ryone knows what attention i s ,” William
James famously quipped (1890, p. 403).” Similarly, everyone seems to know what habit is.
Contemporary researchers of habit, however, start from a specific conceptual definition (Mazar
& Wood, 2019, p. 2): habits are “ c ue-response associations in memory that are acquired slowly
through repetition of an action in a stable c i rc u m s t a n c e ” (Gardner, 2015; Orbell & Verplanken,
2010; Wood & Rünger, 2016). This definition posits that a cognitive system —specifically, an
associative network of cues and responses stored in memory —plays a mediating role in why we
respond a certain way when we encounter certain cues in the world. Given this essential role in
determining habit behavior, we would expect that most contemporary habit research to include
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 65

measures of cognitive associations. However, very little research has attempted to do this (e.g.,
Lin et al., 2016; instead, researchers have relied on self-report measures of habit (for review see
Mazar & Wood, 2018). The present research, which employed qualitatively different tasks, —
learning to move a set of objects or press a set of keys in a specific sequence —demonstrated that
strengthening cognitive associations did not always correlate with self-reported behavioral
automaticity. Why might this be the case?
Interestingly, in both experiments, cognitive associations became significantly stronger
and felt automaticity significantly increased with repetition, yet these measures were sometimes
uncorrelated. It seems reasonable to expect that as people repeat a response in a specific context,
learning processes operate unconsciously in the background —i.e., the mind is tracking and
storing regularities between environmental cues, behavior, and rewarding experiences in
memory —while learned responses consciously feel more and more automatic in the presence of
the specific context. However, the predictive validity and reliability of the self-reported
behavioral automaticity index (Gardner et al., 2012) has been doubted (see review, Labrecque &
Wood, 2015); people c a n’ t reliably introspect on unconscious processes (Nisbett & Wilson,
1977) and when people are asked provide felt automaticity judgments, researchers have argued
that these judgments are based on conscious inferences that do not tap habit processes (Mazar &
Wood, 2018; Sniehotta & Presseau, 2012). Perhaps, felt automaticity may be simply an
epiphenomenon that draws on the same sources of information that elicit feelings of enjoyment
rather than reflecting strengthening cognitive associations.
As the correlations from the present experiments show, felt automaticity was significantly
positively correlated with enjoyment at both time points in both tasks. Moreover, neither
enjoyment nor automaticity was strongly correlated with reaction times from the cognitive
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 66

association tests. To self-report on felt automaticity, perhaps people simply ask themselves “ how
am I fe e l i ng? ” and if it feels good, it feels automatic (this aligns with a feelings-as-information
framework, Schwarz, 1990). If this is true, interpretations of the results from multitudes of
research using the self-report habit index should be revised in light of this possibility. This may
explain, in part, why self-report automaticity measures are sometimes unreliable. Just as self-
reporting on feelings is prone to misattribution effects (Schwarz & Clore, 1983), automaticity
judgments may be prone to how we are feeling in the moment, rather than reflecting the
unconscious strengthening of cognitive associations, as the present research shows.
The results of the present experiments also shed light on what does or does not disrupt
habit formation. In contrast to research that suggests that automaticity is disrupted by conscious
monitoring (Beilock & Carr, 2001), the results show that one can slightly alter how people repeat
a task or where they focus their attention without strongly disrupting habit formation. I
emphasize slightly because it hints at a potential future area of research: how much performance
and/or attentional variance can be introduced to task repetition without disrupting habit
formation? For example, if I would have instructed the experimental group participants to play
the melody backwards instead of simply changing the way they pressed the keys (e.g., “ pre s s
fi rm l y” ), it would have likely disrupted previously learned cue-response associations. But would
playing the melody with one ’s eyes closed or with gloves affect association strengthening?
Perhaps, like the research on performing repetitive tasks in unconventional contexts to curb
hedonic adaptation (eating popcorn with chopsticks vs. hands, O ’Bri e n & Smith, 2019), we can
continue to test the limits of making repetition less dull via variation without disrupting habit
formation.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 67

Regarding attention, perhaps I could have instructed participants to pay attention to
something irrelevant to the task (to daydream or fantasize, Wilson, Westgate, Buttrick, &
Gilbert, 2019) rather than pointing attention toward attributes of the task one is repeating; but
would this distract participants and disrupt habit formation? Future research should tease apart
the contribution of performance and/or attention variation in shaping perceptions of repetition. It
is easier to argue that someone would perceive their behavior as novel when they are asked to
perform a different behavior, even if it is slightly different; in contrast, causing someone to
perceive a task as more novel just by guiding attention requires stronger empirical evidence.
Implications for Research on Hedonic Adaptation
In a paper on “ T he Hedonic Adaptation Prevention M ode l ,” Sheldon et al. (2011, p. 2)
argued that “ va r i e t y is not only the spice of life, but the spice of happiness as w e l l .” They
presented two studies to test the HAP model (one correlational, in which reported variety was
positively correlated with sustained happiness and an experimental one, in which engaging in
varied kindness activities sustained higher well-being that less variation) and inferred that variety
can effectively overcome hedonic adaptation (Sheldon et al., 2011). The present research goes
beyond asking whether variety slows hedonic adaptation, and asks a set of more challenging
questions: why and how does variety have this effect? And, more broadly, what exactly
constitutes variety?
The present experiments show that introducing variation to a repeat task slows hedonic
adaptation, in part, because it changes where attention is focused and the way repetition is
perceived. Specifically, when participants in the studies were instructed to vary how they moved
virtual sushi ingredients (e.g., sharply, quickly), how they pressed keyboard piano keys (e.g.,
firmly, softly), and how they paid attention to these variations, they reported noticing more
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 68

variety within the tasks and perceiving the tasks as less repetitive; in turn, they reported more
enjoyment than participants that just repeated the tasks without variation. This finding is
consistent with research showing that variety slows hedonic adaptation because it increases
attention to distinct elements of a task (O ’Bri e n, 2019; Redden, 2008), increases a sense of
immersion in the task (O ’Bri e n & Smith, 2019), and changes perceptions of repetition (Redden,
2008). The spice of variety can be experienced not only by doing novel activities, but by
discovering novel ways to engage with what is familiar.
Yet, in a sense, the results from the experiments can be explained by the simple
possibility that participants were in fact no longer repeating the same tasks; after all, participants
were presumably pressing the keyboard piano keys in one manner before any instructions, and
then told to do something novel (press the keys in a different manner). This explains why
disrupting repetition by performing novel tasks —the novelty-as-breadth strategy mentioned in
the introduction —is effective at slowing hedonic adaptation. The present research does not
disentangle to what extent the intervention worked because (a) people did slightly novel actions,
(b) people attended to and perceived more novelty within the task, or (c) a combination of
performance and attention novelty. A stronger test of the novelty-as-depth strategy to slow
hedonic adaptation would be to solely manipulate attention without altering performance in
future research.
Whether an experience is repetitive or novel is, in large part, in the eye of the beholder.
As William James (1984, p. 156) put it, “what is called our 'experience' is almost entirely
determined by our habits of attention ” and as O ’Bri e n (2019, p. 519) concisely expressed, “ l i t e ra l
repetition is surprisingly f i c t i ona l . ” Insofar as our perception of an activity can change as our
attention shifts with repetition, drawing the line between literal repetition and perceived
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 69

repetition becomes a challenge. In fact, people can strategically regulate their motivation and
interest by changing where they place their attention and how they perceive a task (Sansone,
Weir, Harpster, & Morgan, 1992). As David Foster Wallace (2011) eloquently describes this
strategy in action:
“It turns out that bliss [...] lies on the other side of crushing, crushing boredom. Pay
attention to the most tedious thing you can find (Tax Returns, Televised Golf) and, in
waves, a boredom like you’ve never known will wash over you and just about kill you.
Ride these out, and i t ’s like stepping from black and white into color. Like water after
days in the desert. Instant bliss in every atom. ” (p. 546)

Fortunately, most of the activities we want to form into habits like cooking, taking an
afternoon walk, or practicing the piano are more enjoyable than completing tax returns.
Unfortunately, like completing tax returns, these too are not immune to hedonic adaptation. The
present research demonstrates one method by which we can keep the spectre of hedonic
adaptation at bay while making the long road to habit formation a bit more enjoyable.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 70

References
Amodio, D. M. & Ratner, K. G. (2011) A memory systems model of implicit social cognition.
Current Directions in Psychological Science, 20, 143 –148.
Anderson, J. R. (1987). Skill acquisition: Compilation of weak-method problem solutions.
Psychological Review, 94, 192-210.
Anderson, B. A. (2016). The attention habit: How reward learning shapes attentional selection.
Annals of the New York Academy of Sciences, 1369(1), 24-39.
Bargh, J. A. (1994). The four horsemen of automaticity: Awareness, intention, efficiency, and
control in social cognition. In R. S., Wyer, Jr., & T. K. Srull (Eds.), Handbook of social
cognition (2nd ed., pp. 1–40). Hillsdale, NJ: Erlbaum.
Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being. American
Psychologist, 54, 462–479.
Baumeister, R. F. (1984). Choking under pressure: Self-consciousness and paradoxical effects of
incentives on skillful performance. Journal of Personality and Social Psychology, 46,
610–620.
Beilock, S.L., & Carr, T.H. (2001). On the fragility of skilled performance: What governs
choking under pressure? Journal of Experimental Psychology: General, 130, 701–725.
Berlyne, D. E. (1970). Novelty, complexity, and hedonic value. Perception & Psychophysics, 8,
279-286.
Brickman, P., & Campbell, D. T. (1971). Hedonic relativism and planning the good society. In
M. H. Appley (Ed.), Adaptation level theory: A symposium (pp. 287–302). New York:
Academic Press.
Brickman, Coates, & Janoff-Bulman. (1978). Lottery winners and accident victims: is
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 71

happiness relative? Journal of Personality and Social Psychology, 36, no. 8, 917-927.
Broniarczyk, S. M., Hoyer, W. D., & McAlister, L. (1998). Cons um e rs ’ perceptions of the
assortment offered in a grocery category: The impact of item reduction. Journal of
Marketing Research, 35(2), 166-176.
Carden, L., Wood W., Neal, D., T., & Pascoe, A. (2017). Incentives activate a control mind-set:
Good for deliberate behaviors, bad for habit pe rform a nc e ,” Journal of the Association for
Consumer Research, 2(3).
Carden, L., & Wood, W. (2018). Habit formation and change. Current Opinion in Behavioral
Sciences, 20, 117-122.
Crolic, C., & Janiszewski, C., (2016). Hedonic escalation: When food just tastes better and
better. Journal of Consumer Research, 43, 388–406.
Csikszentmihalyi, M. (1990). Flow. New York, NY: Harper Perennial.
Danner, U. N., Aarts, H., & de Vries, N. K. (2008). Habit vs. intention in the prediction of future
behaviour: The role of frequency, context stability and mental accessibility of past
behaviour. British Journal of Social Psychology, 47(2), 245–265.
Dezfouli, A., & Balleine, B.W. (2012). Habits, action sequences and reinforcement learning.
European Journal of Neuroscience, 35(7), 1036-1051.
Dimitrov, D. M., & Rumrill Jr, P. D. (2003). Pretest-posttest designs and measurement of
change. Work, 20(2), 159-165.
Epstein, L. H., Temple, J. L., Roemmich, J. N., & Bouton, M. E. (2009). Habituation as a
determinant of human food intake. Psychological Review, 116(2), 384–407.
Evans, J., & Stanovich, K. E. (2013). Dual-process theories of higher cognition:
Advancing the debate. Perspectives on Psychological Science, 8, 223–241.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 72

Frederick, S., & Loewenstein, G. (1999). Hedonic adaptation. In In D. Kahneman, E. Diener, &
N. Schwarz (Eds.), Well-being: The foundations of hedonic psychology (pp. 302-329).
New York: Russell Sage Foundation.
Frijda, N. H. (1988). The laws of emotion. American Psychologist, 43, 349-358.
Galak, J., Redden, J. P., & Kruger, J. (2009). Variety amnesia: Recalling past variety can
accelerate recovery from satiation. Journal of Consumer Research, 36(4), 575–584.
Galak, J., & Redden, J. P. (2018). The properties and antecedents of hedonic decline. Annual
Review of Psychology, 69, 1–25.
Galla, B. M., & Duckworth, A. L. (2015). More than resisting temptation: Beneficial habits
mediate the relationship between self-control and positive life outcomes. Journal of
Personality and Social Psychology, 109, 508–525.
Gardner, B. (2014). A review and analysis of the use of ‘habit ’ in understanding, predicting and
influencing health-related behaviour. Health Psychology Review, 9(3), 277 –295.
Gardner, B., Abraham, C., Lally, P., & de Bruijn, G. J. (2012). Towards parsimony in habit
measurement: Testing the convergent and predictive validity of an automaticity subscale
of the self-report habit index. International Journal of Behavioral Nutrition and Physical
Activity, 9, 102.
Gilbert, D. T., Pinel, E. C., Wilson, T. D., Blumberg, S. J., & Wheatley, T. P. (1998). Immune
neglect: A source of durability bias in affective forecasting. Journal of Personality and
Social Psychology, 75, 617–638.
Gillan, C. M., Otto, A. R., Phelps, E. A., & Daw, N. D. (2015). Model-based learning protects
against forming habits. Cognitive, Affective, & Behavioral Neuroscience, 15, 523-536.
Gosling, S. D., Rentfrow, P. J., & Swann Jr, W. B. (2003). A very brief measure of the Big-Five
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 73

personality domains. Journal of Research in Personality, 37(6), 504-528.
Griffioen-Roose, S., Finlayson, G., Mars, M., Blundell, J. E., & de Graaf, C. (2010). Measuring
food reward and the transfer effect of sensory specific satiety. Appetite, 55, 648-655.
Groves, P. M., & Thompson, R. F. (1970). Habituation: A dual-process theory. Psychological
Review, 77(5), 419–450.
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process analysis: A
regression-based approach. Guilford Publications.
James, W., (1984). Psychology: The Briefer Course. Cambridge, MA: Harvard University Press
James, W., (1977). The Writings of William James: A Comprehensive Edition. Edited by
James J. McDermott. Chicago: University of Chicago Press.
James, W. (1890). Principles of psychology. New York: Henry Holt.
Kahn, B. E., & Ratner, R. K. (2005). Variety for the sake of variety? Diversification motives in
consumer choice. In S. Ratneshwar & D. G. Mick (Eds.), Inside Consumption: Consumer
Motives, Goals, And Desires (pp. 102 –121). Hove, UK: Routledge.
Kahneman, D. (2011). Thinking Fast and Slow. New York, NY: Macmillan.
Kahneman, D., Krueger, A. B., Schkade, D., Schwarz, N., & Stone, A. A. (2006). Would you be
happier if you were richer? A focusing illusion. Science, 312(5782), 1908-1910.
Kahneman, D., & Snell, J. (1992). Predicting a changing taste: Do people know what they will
like? Journal of Behavioral Decision Making, 5, 187 –200.
Killingsworth, M. A., & Gilbert, D. T. (2010). A wandering mind is an unhappy mind. Science,
330(6006), 932-932.
Knowlton, B. J., & Patterson, T. K. (2016). Habit formation and the striatum. Behavioral
Neuroscience of Learning and Memory, 37, 275.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 74

Labrecque, J. (2015). Habits as Friend and Foe to Self-Control: The Dual Roles of Routinized
Behavior (Doctoral dissertation). University of Southern California, Los Angeles, CA.
Labrecque, J. S., & Wood, W. (2015). What measures of habit strength to use? Comment on
Gardner (2015). Health Psychology Review, 9, 303-310.
Labrecque, J. S., Lee, & Wood, W. (2018). Thinking of forming a habit? Manuscript under
review.
Lally, P., Van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed:
Modelling habit formation in the real world. European Journal of Social Psychology, 40,
998–1009.
Lewis, B., & Linder, D. (1997). Thinking about choking? Attentional processes and paradoxical
performance. Personality and Social Psychology Bulletin, 23, 937–944.
Lin, P. Y., Wood, W., & Monterosso, J. (2016). Healthy eating habits protect against
temptations. Appetite, 103, 432-440.
Logan, G. D. (1988). Toward an instance theory of automatization. Psychological Review, 95,
492–527.
Lucas, R. E., & Clark, A. E. (2006). Do people really adapt to marriage? Journal of Happiness
Studies, 7(4), 405-426.
Lyubomirsky, S., Sheldon, K. M., & Schkade, D. (2005). Pursuing happiness: The architecture
of sustainable change. Review of General Psychology, 9(2), 111-131.
Marteau, T. M., Hollands, G. J., & Fletcher, P. C. (2012). Changing human behavior to prevent
disease – The importance of targeting automatic processes. Science, 337, 1492 –1495.
Martin, L. L., Ward, D. W., Ach é e, J. W., & Wyer, R. S. (1993). Mood as input: People have to
interpret the motivational implications of their moods. Journal of Personality and Social
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 75

Psychology, 64, 317-326.
Masters, R. S. W. (1992). Knowledge, knerves and know-how: The role of explicit versus
implicit knowledge in the breakdown of a complex motor skill under pressure. British
Journal of Psychology, 83, 343–358.
Mazar, A., & Wood, W. (2018). Defining habit in psychology. The Psychology of Habit (pp.
13-29). Springer, Cham.
Moors, A., & De Houwer, J. (2006). Automaticity: a theoretical and conceptual analysis.
Psychological Bulletin, 132(2), 297.
Nelson, L. D., & Meyvis, T. (2008). Interrupted consumption: Disrupting adaptation to hedonic
experiences. Journal of Marketing Research, 45, 654 –664.
Neal, D. T., Wood, W., Labrecque, J., & Lally, P. (2012). How do habits guide behavior?
Perceived and actual triggers of habits in daily life. Journal of Experimental Social
Psychology, 48, 492-498.
Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on
mental processes. Psychological Review, 84, 231 –259.
Norcross, J. C., Ratzin, A. C., & Payne, D. (1989). Ringing in the New Year: The change
processes and reported outcomes of resolutions. Addictive Behaviors, 14(2), 205-212.
O ’Bri e n, E. (2019). Enjoy it again: Repeat experiences are less repetitive than people think.
Journal of Personality and Social Psychology, 116(4), 519.
O ’Bri e n, E., & Smith, R. W. (2019). Unconventional consumption methods and enjoying things
consumed: Recapturing the “ f i rs t- t i m e ” experience. Personality and Social Psychology
Bulletin, 45(1), 67-80.
Orbell, S., & Verplanken, B. (2018). Progress and Prospects in Habit Research. In The
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 76

Psychology of Habit (pp. 397-409). Springer, Cham.
Orbell, S., & Verplanken, B. (2010). The automatic component of habit in health behavior: Habit
as cue-contingent automaticity. Health Psychology, 29(4), 374.
Ouellette, J., & Wood, W. (1998). Habit and intention in everyday life: The multiple processes
by which past behavior predicts future behavior. Psychological Bulletin, 124, 54-74.
Parducci, A. (1995). Happiness, pleasure, and judgment: The contextual theory and its
applications. Hillsdale, NJ, England: Lawrence Erlbaum Associates, Inc.
Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change:
applications to addictive behaviors. American Psychologist, 47(9), 1102.
Quoidbach, J., & Dunn, E. W. (2013). Give it up: A strategy for combating hedonic adaptation.
Social Psychological and Personality Science, 4, 563-568.
Ratner, R. K., Kahn, B. E., & Kahneman, D. (1999). Choosing less-preferred experiences for the
sake of variety. Journal of Consumer Research, 26, 1–15.
Read, D., & Loewenstein, G. (1995). Diversification bias: Explaining the discrepancy in variety
seeking between combined and separated choices. Journal of Experimental Psychology:
Applied, 1(1), 34.
Redden, J. P., (2008). Reducing Satiation: The Role of Categorization Level. Journal of
Consumer Research, 34, 624–34.
Rolls, B. J., Rolls, E. T., Rowe, E. A., & Sweeney, K. (1981). Sensory specific satiety in man.
Physiology and Behavior, 27, 137–142.
Sansone, C., Weir, C., Harpster, L., & Morgan, C. (1992). Once a boring task, always a boring
task? Interest as a self-regulatory mechanism. Journal of Personality and Social
Psychology, 63, 379-390.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 77

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: new
procedures and recommendations. Psychological Methods, 7(4), 422.
Schwarz, N. (1990). Feelings as information: Informational and motivational functions of
affective states. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of Motivation and
Cognition (Vol. 2, pp. 521–561). New York: Guilford.
Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being:
Informative and directive functions of affective states. Journal of Personality and Social
Psychology, 45(3), 513-523.
Schwarz, N., & Clore, G. L. (1988). How do I feel about it? The information function of
affective states. In K. Fiedler & J. P. Forgas (Eds.), Affect, cognition and social behavior:
New evidence and integrative attempts (pp. 44–63). Toronto: C.J. Hogrefe.
Schwarz, N., & Xu, J. (2011). Why don't we learn from poor choices? The consistency of
expectation, choice, and memory clouds the lesson of experience. Journal of Consumer
Psychology, 21, 142–145.
Sheldon, K. M., Boehm, J., & Lyubomirsky, S. (2013). Variety is the spice of happiness: The
hedonic adaptation prevention model. Oxford handbook of happiness, 901-914.
Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human information
processing: II. Perceptual learning, automatic attending and a general theory.
Psychological Review, 84, 127–190.
Sniehotta, F. F., & Presseau, J. (2012). The Habitual Use of the Self-report Habit Index. Annals
of Behavioral Medicine; New York, 43(1), 139–140.
Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior.
Personality and Social Psychology Review, 8, 220 –247.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 78

Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for the study of
neural substrates of behavior. Psychological Review, 73, 16–43.
Tricomi, E., Balleine, B. W., & O ’D ohe rt y, J. P. (2009). A specific role for posterior dorsolateral
striatum in human habit learning. European Journal of Neuroscience, 29(11), 2225-2232.
Vallacher, R. R., & Wegner, D. M. (1987). What do people think t he y’r e doing? Action
identification and human behavior. Psychological Review, 94, 3–15.
Vallacher, R., & Wegner, D. (1989). Levels of personal agency: Individual variation in
action identification. Journal of Personality and Social Psychology, 57, 660-671.
Verplanken, B., Aarts, H., & Van Knippenberg, A. (1997). Habit, information acquisition, and
the process of making travel mode choices. European Journal of Social Psychology,
27(5), 539-560.
Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: A self-report index of habit
strength. Journal of Applied Social Psychology, 33, 1313–1330.
Wallace, D. F. (2011). The Pale King: An Unfinished Novel. New York, NY: Little, Brown, and
Co.
Wilson, T.D., & Gilbert, D.T. (2008) Explaining away: a model of affective adaptation.
Perspectives in Psychological Science. 3, 370–86.
Wilson, T. D., Westgate, E. C., Buttrick, N. R., & Gilbert, D. T. (2019). The mind is its own
place: The difficulties and benefits of thinking for pleasure. Advances in Experimental
Social Psychology, 20, 1.
Whelan, R. (2008). Effective analysis of reaction time data. The Psychological Record, 58(3),
475-482.
Wood, W., & Neal, D. T. (2009). The habitual consumer. Journal of Consumer Psychology, 19,
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 79

579-592.
Wood, W., & Neal, D. T. (2016). Healthy through habit: Interventions for initiating and
maintaining health behavior change. Behavioral Science & Policy, 2, 71-83.
Wood, W., Quinn, J. M., & Kashy, D. A. (2002). Habits in everyday life: Thought, emotion,
and action. Journal of Personality and Social Psychology, 83, 1281-1297.
Wood, W., & Rünger, D. (2016). Psychology of habit. Annual Review of Psychology, 67,
289-314.
Woolley, K., & Fishbach, A. (2017). Immediate rewards predict adherence to long-term goals.
Personality and Social Psychology Bulletin, 43, 151 –162.
Yin, H. H., & Knowlton, B. J. (2006). The role of the basal ganglia in habit formation. Nature
Reviews Neuroscience, 7(6), 464.
Zajonc, R. B. (1968). The attitudinal effects of mere exposure. Journal of Personality and Social
Psychology, 9, 1–27.
Zamble, E. (1992). Behavior and adaptation in long-term prison inmates: Descriptive
longitudinal results. Criminal Justice and Behavior, 19, 409-425.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 80

Appendix A
Pilot Study: Testing Sushi Game Paradigm for Hedonic Adaptation and Habit Formation
To test the hypotheses proposed in this paper, a lab task was needed that could reliably
induce hedonic adaptation and habit formation. That is, a task that is enjoyable at first, but
decreases in enjoyment when repeated. And a task that shows increases in habit strength when
repeated in the same time frame. Then, it would be possible to test an intervention that slows
hedonic adaptation and measure its effect habit formation.
We used a computer sushi game in which participants move ingredients on a computer
screen to make a roll of sushi in a 16-step sequence. Originally designed for children, the video
game was tailored by a programmer for our research so that we could manipulate different
parameters within the game and so that we could measure how habitual the task was becoming.
The game is ideal to study habit formation because it (a) can be repeated varying numbers of
times, (b) involves a sequence of actions, which is important in habit formation (Dezfouli &
Balleine, 2012), and (c) provides discrete contexts (prior step in sequence of recipe) and
responses (current step) that enable tests of the strength of habit associations in memory (Wood
& Rünger, 2016).
In the pilot study, some participants repeated the sequence only a few times (2 trials),
whereas others practiced it many times (10 trials). If playing the sushi game many times formed
stronger habits, then we would expect both habit strength measures, cognitive association
reaction time and self-reported automaticity, to be higher in the many times group. If playing the
sushi game many times was less enjoyable than playing it a few times, then we would expect
higher enjoyment ratings from the few times group compared to the many times group. Measures
of fluency, ease, and motivation were also used to better understand how participants were
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 81

experiencing the sushi game. For example, showing that participants in both groups were
motivated to the same degree to complete all the task trials could rule out whether decreases in
ratings of enjoyment were due to decreases in motivation.
Method
Participants were 107 university students who completed the sushi computer game task in
individual experimental rooms.
Experimental design. Participants were guided by avatar on the screen that described
each of the 16 steps in the recipe (see Fig. 1 below).

Figure 1. Screenshot of sushi game.
Participants moved the cursor over to the relevant item on the table, clicked on the item,
and dragged it to the appropriate location (e.g., dragged the rice into the pot). The task provided
immediate feedback about the success of pa rt i c i pa nt s ’ choice by completing that step in the
recipe. Attempting to click and move an incorrect recipe item yielded no program response.
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 82

Participants could make as many tries as they wished until they selected the correct item in the
sequence.
Measures After Trials
Habit strength. To test habit strength, i.e., strength of cognitive associations, participants
were primed with one step in the sushi-making sequence before responding with the next step
(see Fig. 2). The 56 trials each consisted of a 2s fixation cross, a 1s prime representing a prior
step in the recipe (e.g., vinegar), and then 3s of a choice of two possible next steps in the recipe
(one correct, one incorrect). Strength of (habit) associations between contexts and responses was
assessed from mean reaction times (RT) for all correct trials (RT were analyzed in raw form and
log-transformed; both showed the same results).

Figure 2. Screenshot of context-response associations test.

Self-report behavioral automaticity. (Gardner et al., 2012). Participants indicated on 4
scales from 1 (strongly disagree) to 5 (strongly agree), “ Be gi nni ng the next step in making sushi
is s om e t hi ng...” “I do automatic a l l y; ” “I do without having to consciously re m e m be r ; ” “I do
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 83

without t hi nk i ng; ” and “I start doing before I realize I’m doing i t . ” Reliability was good (alpha =
.69).
Task enjoyment. Participants reported on a 5-point scale from 1 (disliked considerably)
to 5 (liked considerably) how much they enjoyed making sushi.
Task fluency. Participants reported on a 5-point scale from 1 (strongly disagree) to 5
(strongly agree) whether the “ s us hi-making process felt fluent, like it fl ow e d” during training.
Task motivation. Participants indicated on a scale anchored by 1 (strongly disagree) and
5 (strongly agree) whether they intended to make sushi correctly in the task.
Task ease/difficulty. Participants indicated their ease of completing the task successfully
from 1 (extremely difficult) to 7 (extremely easy).
Results
To evaluate the effects of the experimental manipulations, two-way mixed ANOVAs
were computed with between-subjects factor as the number of sushi game trials practices (2 vs.
10) and the within-subjects factors as the type of habit measure (reaction time scores and self-
reported automaticity ratings) and enjoyment ratings. To test for main effects, t-tests were
computed. Means and standard deviations for the study are in Table 1.
Table 1
Means and Standard Deviations
2 trial group
(n = 60)
10 trial group
(n = 47)
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 84

Habit strength (Raw RT
reported - log transformed
show same results)
.96 (.25) .79 (.28)
Perceived automaticity
(SRBAI)
2.99 (.957) 3.62 (1.143)
Task enjoyment 4.12 (.90) 3.47 (1.12)
Fluency 4.15 (1.07) 4.51 (1.14)
Task motivation 4.33 (.79) 4.40 (.80)
Task ease 5.55 (1.27) 6.11 (.94)

Note. Habit strength as assessed by reaction times in the context-response association test;
perceived automaticity as assessed by a mean of 4 scales from 1 (strongly disagree) to 5
(strongly agree); task fluency from 1 (strongly disagree) to 5 (strongly agree); task enjoyment
from 1 (disliked considerably) making sushi to 5 (liked considerably); task motivation to make
sushi correctly from 1 (strongly disagree) to 5 (strongly agree); task ease from 1 (extremely
difficult) to 7 (extremely easy).
Habit and enjoyment measures. Two-way mixed ANOVAs were conducted to test
interactions between the type of measure (i.e., habit strength reaction time scores/self-reported
automaticity ratings vs. emotion ratings, within-subjects factor) and the number of sushi game
trials (i.e., 2 vs 10, between-subjects factor).
First, the ANOVA showed that mean self-report automaticity and enjoyment ratings were
differentially influenced by the amount of trials practiced, F(1, 105) = 25.18, p < .001. That is,
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 85

participants that repeatedly practiced the game many times were significantly more likely to have
different self-report automaticity and enjoyment ratings compared with participants that
practiced the game a few times. Likewise, that habit strength reaction time scores and enjoyment
ratings were differentially influenced by the amount of trials practiced, F(1, 105) = 23.85, p <
.001. That is, participants that repeatedly practiced the game many times were significantly more
likely to have different reaction time scores and enjoyment ratings compared with participants
that practiced the game a few times. Given the significant interaction results in the omnibus tests,
the following analyses investigate the direction of the effects for each measure.
Increasing habit strength. Participants in the many trials group were quicker to
recognize the next step in the sushi sequence when primed with the previous step (M = 0.79, SD
= 0.28) compared with participants in the few trials group (M = 0.96, SD = 0.25); t(105) = 3.31, p
< .001.
Participants in the many trials group reported higher ratings of experienced automaticity
(M = 3.62, SD = 1.14) compared with participants in the few trials group (M = 2.99, SD = 0.96);
t(105) = 3.10, p = .002. More repetitions of the sushi game led to higher self-reported
automaticity and faster responses in cue-response cognitive association task.
Decreasing enjoyment. Repetition lead to hedonic adaptation, participants in the many
trials group reported lower ratings of enjoyment (M = 3.47, SD = 1.12) compared with
participants in the few trials group (M = 4.12, SD = 0.90); t(105) = 3.33, p < .001.
Motivation and ease. Participants in the many trials group reported equivalent
motivation (M = 4.40, SD = 0.80) compared with participants in the few trials group (M = 4.33,
SD = 0.79); t(105) = 0.45, p > .250. Participants in the many trials group reported higher ratings
of task ease (M = 6.11, SD = 0.94) compared with participants in the few trials group (M = 5.55,
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 86

SD = 1.27); t(105) = 2.53, p = .013. This is an interesting result since it would be reasonable to
attribute the decline in enjoyment to the possibility that participants in the many trials group
simply lost motivation as they repeated the game. However, given the results, the decline in
enjoyment ratings cannot be attributed to a loss of motivation or challenge. This makes the sushi
game a good paradigm in which to study habit formation and hedonic adaptation without the
confound of changing motivation.
Other Ratings
There was no significant difference in ratings of fluency ratings between participants in
the many trials group (M = 4.51, SD = 1.14) compared with participants in the few trials group
(M = 4.15, SD = 1.07); t(105) = 1.68, p = .96.
Intercorrelations. Enjoyment ratings and habit automaticity scores, as measured by
reaction time to cognitive associations, were negatively correlated, r(105) = -.19, p < .01 (see
Table 2). However, there was no significant correlation between enjoyment ratings and self-
reported automaticity scores, r(105) = .08, p > .250, suggesting that enjoyment ratings and self-
reported experiences of automaticity are not increasing or decreasing at similar rates, even
though they are both increase and decrease significantly as a result of the amount of trials as
shown in the main effects results. Moreover, both measures of habit, reaction times and self-
reported automaticity, were uncorrelated, r(105) = .001, p > .250, suggesting that self-reported
experiences of automaticity are not tracking implicit automaticity via strengthening of cognitive
associations.
Table 2
Correlation Matrix
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 87

SRBAI Fluency
Task
enjoyment
Task
motivation
Task
ease
Habit
strength
(RT)
Fluency

.24**

-

Task
enjoyment

.08

.18**

-

Task
motivation

.20**

.29**

.05

-

Task ease

.26**

.28**

.06
.40**
-

Habit
Strength
(RT)

.00

.01
-
.19**

.03

.12

-

Note. Task fluency from 1 (strongly disagree) to 5 (strongly agree); task enjoyment from 1
(disliked considerably) making sushi to 5 (liked considerably); task motivation to make sushi
correctly from 1 (strongly disagree) to 5 (strongly agree); task ease from 1 (extremely difficult)
to 7 (extremely easy); intention to add the newly chosen ingredient from 1 (strongly disagree)
and 5 (strongly agree); Reaction times in the context-response association test. * p < 0.05, ** p <
0.01

Discussion
The purpose of this pilot study was to validate an experimental method to demonstrate
two effects, increased habit strength and hedonic adaptation. Results indicated that repeating the
16-step sushi game many times compared with a few times led to increased habit strength (using
the cognitive association test and using the subjective automaticity measures) and decreased
enjoyment ratings. In short, participants formed stronger habits and enjoyed the task less over
time. This empirical evidence allows for testing the proposed novelty intervention to slow
hedonic adaptation and measure its effect on habit strength.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 88

Appendix B
A principal components analysis (PCA) was run on all items from three indexes —
perceptions of repetition, attention to distinct aspects of task, and retrospective judgments of
variety —to determine whether items did indeed load on the intended constructs (items and
constructs were adapted from Study 3, Redden, 2008).
For Experiment 1, PCA revealed three components which explained 65.3% of the total
variance. A Varimax orthogonal rotation was employed. The interpretation of the data was
consistent with items loading highest on the intended constructs: perceptions of repetition items
on Component 1, attention to distinct aspects of task items on Component 2, and retrospective
judgments of variety items on Component 3. Component loadings and communalities of the
rotated solution are presented in Table 1.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 89

Table 1
Factor Loadings from Experiment 1 Principal Component Analysis with Varimax
Rotation
Component Loading
Items 1 2 3
Each round making sushi seemed very similar to
each other.
.743 .012 -.210
Playing the sushi game felt like the same thing
over and over.
.737 .119 -.372
Each round making sushi had aspects that made
it different. (reverse coded)
.626 -.139 -.345
Different ways to move the ingredients when
making sushi were obvious.
.179 .805 .190
I could identify different ways to move the
ingredients when making sushi.
.042 .734 .044
I really noticed the different ways I moved the
ingredients when making sushi.
-.331 .601 -.089
I did not pay much attention to the different
ways I moved the ingredients when making
sushi. (reverse coded)
-.563 .571 -.35
There was a lot of variety in the sushi game. -.246 .105 .857
There were many different ways to play the
sushi game.
-.303 -.018 .807
Component Loadings and Communalities from Principal Components Analysis for 9
items from Experiment 1

For Experiment 2, PCA also supported a three component structure which explained
78.2% of the total variance. A Varimax orthogonal rotation was employed. Component loadings
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 90

and communalities of the rotated solution are presented in Table 2. The interpretation of the data
was consistent with items loading highest on the intended constructs: perceptions of repetition
items on Component 1, attention to distinct aspects of task items on Component 2, and
retrospective judgments of variety items on Component 3.
Table 2
Component Loadings and Communalities from Principal Components Analysis for 9 items from
Experiment 2
Component Loading
Items 1 2 3
Each 90 second round playing the melody on
the piano seemed very similar to each other.
.886 -.197 -.096
Playing the melody on the piano each 90 second
round felt like the same thing over and over.
.896 -.043 -.117
Each 90 second round playing the melody on
the piano had aspects that made it different.
(reverse coded)
.626 -.293 -.361
Different ways to experience playing the
melody on the piano during each 90 second
round were obvious.
-.023 .629 .574
I could identify different ways to play the
melody on the piano during each 90 second
round.
-.012 .650 .639
I really noticed the different ways I could play
the melody on the piano during each 90 second
round.
-.227 .642 .572
I did not pay much attention to the different
ways I played the melody on the piano during
each 90 second round. (reverse coded)
-.197 .869 .096
There was a lot of variety in playing the melody
on the piano during each 90 second round.
-.325 .142 .831
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 91

There were many different ways to play the
melody on the piano during each 90 second
round.
-.168 .246 .865

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 92

Appendix C
Means and Standard Deviations for Dependent Variables in Experiment 1
Control Group (n = 65) Novelty Group (n = 70)
Statistic Time 1 Time 2 Time 1 Time 2
Enjoyment
M
SD
5.29
1.21
4.43
1.58
5.69
1.23
5.26
1.42
Motivation
M
SD
4.68
1.32
4.18
1.43
4.49
1.43
4.14
1.57
Boredom
M
SD
3.78
1.57
4.43
1.71
3.51
1.41
4.41
1.44
Perceived Automaticity
M
SD
4.06
1.39
5.05
1.25
3.93
1.2
5.21
1.23
Reaction Time in ms
M
SD
820
193
633
137
878
187
661
155
Perception of Repetition (only Time 2)
M
SD
5.56
0.92
4.99
1.17
Attention to Task (only Time 2)
M
SD
4.78
0.94
4.5
1.12
Perceived Variety (only Time 2)
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 93

M
SD
3.78
0.69
3.86
0.93
Pay Requested to Continue Playing in dollars (only Time 2)
M
SD
30.13
30.57
33.16
34.22

Note. Time 1: Baseline measure after 2 minutes of Sushi Game repetition. Time 2: Post
intervention measure after 10 minutes of Sushi Game repetition. Enjoyment: Higher scores,
increased reported enjoyment; scale 1-7. Motivation: Higher scores, increased reported
motivation; scale 1-7. Boredom: Higher scores, increased reported boredom; scale 1-7. Perceived
Automaticity: Self-report perceived behavior automaticity index (Higher scores, increased felt
automaticity); scale 1-7. Reaction Time: Average reaction times from habit strength cognitive
association task in milliseconds (Lower scores, stronger habits). Perception of Repetition: Higher
scores, increased perceived repetitiveness of Sushi Game (e.g., Playing the sushi game felt like
the same thing over and over); scale 1-7. Attention to Task: Higher scores, increased reported
attention to details of Sushi Game (e.g., I really noticed the different ways I moved the
ingredients when making sushi); scale 1-7. Perceived Variety: Higher scores, increased
perceived variety in repeating the sushi game (e.g., There was a lot of variety in the sushi game.);
scale 1-7. Pay Requested to Continue Playing: “ If you were paid to continue playing this game
for 30 more minutes, how much money would you request from us? From 0 - 100 in dollars.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 94

Appendix D
Means and Standard Deviations for Dependent Variables in Experiment 2
Control Group (n = 68) Novelty Group (n = 73)
Statistic Time 1 Time 2 Time 1 Time 2
Enjoyment
M
SD
5.4
1.52
4.59
1.92
5.63
1.34
5.29
1.5
Motivation
M
SD
5.49
1.45
4.85
1.81
5.68
1.42
5.45
1.35
Boredom
M
SD
2.91
1.8
3.99
2.19
2.58
1.35
3.18
1.68
Perceived Automaticity
M
SD
3.42
1.47
4.79
1.34
3.64
1.46
4.58
1.59
Reaction Time in ms
M
SD
996
232
576
233
1048
236
697
297
Perception of Repetition (only Time 2)
M
SD
6.04
1.08
5.11
1.08
Attention to Task (only Time 2)
M
SD
3.79
1.61
4.11
1.58
Perceived Variety (only Time 2)
EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 95

M
SD
3.79
1.85
4.18
1.65
Pay Requested to Continue Playing in dollars (only Time 2)
M
SD
4.71
3.44
5.08
3.18

Note. Time 1: Baseline measure after 90 seconds of repeating piano melody. Time 2: Post
intervention measure after 7.5 minutes of piano melody repetition. Enjoyment: Higher scores,
increased reported enjoyment; scale 1-7. Motivation: Higher scores, increased reported
motivation; scale 1-7. Boredom: Higher scores, increased reported boredom; scale 1-7. Perceived
Automaticity: Self-report perceived behavior automaticity index (Higher scores, increased felt
automaticity); scale 1-7. Reaction Time: Average reaction times from habit strength cognitive
association task in milliseconds (Lower scores, stronger habits). Perception of Repetition: Higher
scores, increased perceived repetitiveness of piano melody (e.g., Playing the melody on the piano
each 90 second round felt like the same thing over and over); scale 1-7. Attention to Task: Higher
scores, increased reported attention to details of piano melody (e.g., I really noticed the different
ways I could play the melody); scale 1-7. Perceived Variety: Higher scores, increased perceived
variety in repeating the piano melody (e.g., There were many different ways to play the melody
on the piano during each 90 second round); scale 1-7. Pay Requested to Continue Playing: “ If
you were paid to continue playing this game for 10 more minutes, how much money would you
request from us? From 0 - 10 in dollars.

EFFECT OF NOVELTY ON HEDONIC ADAPTATION AND HABIT 96

Appendix E

Figure 1. Example of incorrect reaction time response in experiment 2.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Illusory motives, elusive habits: lay theories of everyday behavior

PDF

The role of good habits in facilitating long-term goals

PDF

Only half of what I’ll tell you is true: how experimental procedures lead to an underestimation of the truth effect

PDF

Spatial and temporal patterns of brain activity associated with emotions in music

PDF

Better now than later: the cost of victims' delayed accusations

PDF

Selflessness takes time: altruistic (but not cooperative) prosocial behavior increases with decision time

PDF

How perceived moral congruence shapes propensities to engage in pro-group behaviors

PDF

Novelty versus familiarity: divergent effects of low predictability and low personal influence

PDF

Black sheep or devil protection?: the effects of cooperative and competitive intergroup goal structures on group members reactions to an offensive act when committed by an ingroup or an outgroup ...

PDF

The effect of present moment awareness and value intervention of ACT on impulsive decision-making and impulsive disinhibition

PDF

A roadmap for changing student roadmaps: designing interventions that use future “me” to change academic outcomes

PDF

When photos backfire: truthiness and falsiness effects in comparative judgements

PDF

Meta-accuracy as a moderator of the association between social experience and emotional adjustment during childhood

PDF

The footprint of pain: investigating persistence of altered trunk control in recurrent low back pain

PDF

Visual and audio priming of emotional stimuli and their relationship to intertemporal preference shifts

PDF

Using virtual humans during clinical interviews: examining pathways to self-disclosure

PDF

The structure of kinematic variability during hopping: a window into motor control factors in overuse injury

PDF

Adaptive patterning during skin formation: assembly of vasculature and adipose tissue

PDF

Young children’s demand for reciprocal engagement: evidence from typically-developing and autistic preschoolers

PDF

Politics is something we do together: identity and institutions in U.S. elections

Asset Metadata

Creator Carden, Lucas M. (author)

Core Title Performance and attention novelty slows hedonic adaptation during habit formation

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 11/15/2019

Defense Date 08/20/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag attention,enjoyment,habit formation,hedonic adaptation,novelty,OAI-PMH Harvest,perception,repetition

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Wood, Wendy (committee chair), Monterosso, John (committee member), Schwarz, Norbert (committee member), Wiltermuth, Scott (committee member)

Creator Email lcarden@usc.edu,lucascarden@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-233722

Unique identifier UC11674802

Identifier etd-CardenLuca-7927.pdf (filename),usctheses-c89-233722 (legacy record id)

Legacy Identifier etd-CardenLuca-7927.pdf

Dmrecord 233722

Document Type Dissertation

Rights Carden, Lucas M.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA