See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/261066543

A Habit-Goal Framework of Depressive Rumination

Article  in  Journal of Abnormal Psychology · February 2014

DOI: 10.1037/a0035540 · Source: PubMed

CITATIONS READS
72 1,514

2 authors, including:

Edward R Watkins
University of Exeter
136 PUBLICATIONS   10,518 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

RuMeChange: Mechanisms of Rumination Change in Adolescents with remitted MDD View project

PREVENT View project

All content following this page was uploaded by Edward R Watkins on 03 April 2014.

The user has requested enhancement of the downloaded file.

 Journal of Abnormal Psychology © 2014 American Psychological Association
2014, Vol. 123, No. 1, 24 –34 0021-843X/14/$12.00 DOI: 10.1037/a0035540

A Habit-Goal Framework of Depressive Rumination

Edward R. Watkins Susan Nolen-Hoeksema

University of Exeter and University of Western Australia Yale University

Rumination has been robustly implicated in the onset and maintenance of depression. However, despite
empirically well-supported theories of the consequences of trait rumination (response styles theory;
Nolen-Hoeksema, 1991), and of the processes underlying state episodes of goal-oriented repetitive
thought (control theory; Martin & Tesser, 1989, 1996), the relationship between these theories remains
unresolved. Further, less theoretical and clinical attention has been paid to the maintenance and treatment
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

of trait depressive rumination. We propose that conceptualizing rumination as a mental habit (Hertel,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

2004) helps to address these issues. Elaborating on this account, we propose a framework linking the
response styles and control theories via a theoretical approach to the relationship between habits and
goals (Wood & Neal, 2007). In this model, with repetition in the same context, episodes of self-focused
repetitive thought triggered by goal discrepancies can become habitual, through a process of automatic
association between the behavioral response (i.e., repetitive thinking) and any context that occurs
repeatedly with performance of the behavior (e.g., physical location, mood), and in which the repetitive
thought is contingent on the stimulus context. When the contingent response involves a passive focus on
negative content and abstract construal, the habit of depressive rumination is acquired. Such habitual
rumination is cued by context independent of goals and is resistant to change. This habit framework has
clear treatment implications and generates novel testable predictions.

Keywords: rumination, control theory, response styles, goal, habit

Depressive rumination is the tendency to repetitively analyze
oneself and one’s problems, concerns, and feelings of distress and
depressed mood (Nolen-Hoeksema, 1991; Watkins, 2008). Pro-
spective longitudinal and experimental studies have implicated
depressive rumination in the onset and maintenance of symptoms
and diagnosis of major depression (Nolen-Hoeksema, 2000;
Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008; Watkins, 2008),
as well as anxiety disorders, eating disorders, and substance/
alcohol abuse, leading to the hypothesis that it is a transdiagnostic
pathological process (Harvey, Watkins, Mansell, & Shafran, 2004;
Ehring & Watkins, 2008; Nolen-Hoeksema & Watkins, 2011).
The principal theory concerning depressive rumination is the
response styles theory (RST; Nolen-Hoeksema, 1991), which hy-
pothesizes that depressive rumination is a stable, enduring, and
habitual trait-like tendency to engage in repetitive self-focus in
response to depressed mood (see Nolen-Hoeksema et al., 2008).
RST argues that depressive rumination leads to depression by
enhancing negative mood-congruent thinking, impairing problem
solving, and interfering with instrumental behavior, with these
mechanisms confirmed in experimental studies (Nolen-Hoeksema,
1991; Nolen-Hoeksema et al., 2008).
Edward R. Watkins, Mood Disorders Centre, School of Psychology,
University of Exeter, Exeter, United Kingdom, and School of Psychology,
University of Western Australia, Perth, Australia; Susan Nolen-Hoeksema,
Department of Psychology, Yale University.
Correspondence concerning this article should be addressed to Edward R.
Watkins, Sir Henry Wellcome Building for Mood Disorders Research, University
of Exeter, Perth, Australia, EX4 4GG. E-mail: e.r.watkins@exeter.ac.uk
24
A distinct alternative view of rumination is provided by
control theory accounts, which define it as “a class of conscious
thoughts that revolve around a common instrumental theme and
that recur in the absence of immediate environmental demands
requiring the thoughts” (Martin & Tesser, 1996, p. 7). This account
explains the conditions leading to state episodes of ruminative
thought. It hypothesizes that state rumination is initiated by, and
focused on, a perceived discrepancy between one’s goals and one’s
current state and continues until the goal is either attained or
abandoned (Martin & Tesser, 1989, 1996). Unlike RST, this ac-
count encompasses both normal and constructive versus abnormal
and unconstructive forms of repetitive thought: Goal-oriented ru-
mination can be helpful if focus on the unresolved goal facilitates
progress, but unhelpful if it only serves to increase salience of the
discrepancy. Consistent findings confirm that unattained person-
ally important goals predict repetitive thought in real-world studies
(see Watkins, 2008).
Watkins (2008) examined what processes determine whether
repetitive thought has unconstructive versus constructive conse-
quences. During rumination, negative thought content, negative
context (e.g., negative mood), and abstract-evaluative thinking
focused on the meaning and implications of negative events (e.g.,
asking “Why?”) relative to concrete thinking, focused on the
details, process, and context of negative events (e.g., asking
“How?”), contribute to unconstructive outcomes. In control theory,
goals, events, and actions can be represented at different levels of
abstraction within a hierarchy ranging from abstract representa-
tions capturing their decontextualized gist and characterizing why
they occurred or are performed, down to concrete representations
of their contextual and situational details characterizing how they
occurred or are to be performed. Effective self-regulation requires
 HABIT FRAMEWORK OF DEPRESSIVE RUMINATION
flexible movement through the hierarchy to match circumstances.
Overly abstract processing of negative information was hypothe-
sized and found to be unhelpful because it increases the personal
importance of events and exacerbates emotional reactivity (Wat-
kins, Moberly, & Moulds, 2008) and impairs problem-solving by
limiting the availability of detailed alternative plans (Watkins &
Moulds, 2005). This elaborated control theory account (Watkins,
2008) provides a partial reconciliation between RST and control
theory because it identifies abstract-evaluative thinking focused on
negative content as pathological, consistent with the RST charac-
terization of depressive rumination as focusing on the causes and
implications of depressive mood.
Despite the successes of these theories, key questions such as
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
“What are the developmental antecedents of individual differences
This document is copyrighted by the American Psychological Association or one of its allied publishers.
in rumination?” and “What makes it so difficult to break free of
rumination once it has begun?” remain unanswered (Nolen-
Hoeksema et al., 2008, p. 418). Moreover, the relationship be-
tween state episodes of goal-oriented repetitive thought and the trait
tendency toward depressive rumination is unresolved. Are they sepa-
rate and distinct processes? Or do some individuals transition from
state episodes of goal-oriented rumination into more habitual de-
pressive rumination?
A major but relatively unelaborated implication within RST is
that depressive rumination is a habitual cognitive response to
negative mood (Nolen-Hoeksema et al., 2008). We propose that
more explicitly considering rumination as a mental habit (see
Hertel, 2004) has value in generating potential insights and testable
hypotheses relevant to these issues. Habits are behaviors that are
performed frequently in stable contexts (Ji & Wood, 2007). In
classical conditioning and learning theory (e.g., Hull, 1943), habits
are learned behavioral responses to situational cues that have
become established through a history of systematic repetition and
reinforcement. A stimulus–response (S-R) habit is learnt when a
particular behavioral response is contingent on the stimulus (Dick-
inson, 1985; Mackintosh, 1983; Rescorla & Wagner, 1972). For
example, with overtraining or training with interval schedules of
reward contingency, animals learn to perform a response habitu-
ally to set cues, even when the reward is devalued, for example, by
making animals mildly ill with a lithium chloride injection after
consuming sucrose pellets (Dickinson, 1985).
Wood and Neal (2007, p. 843) defined habits as “learned dis-
positions to repeat past responses. They are triggered by features of
the context that have covaried frequently with past performance,
including performance locations, preceding actions in a sequence,
and particular people.” Habitual behavior typically involves some
automaticity (Verplanken, 2006; Verplanken & Orbell, 2003;
Wood, Tam, & Witt, 2005). A behavior can be conceptualized as
automatic on several distinct dimensions: lack of conscious aware-
ness, not requiring extensive resources to be performed (e.g.,
performed equally well under cognitive load vs. not under load),
lack of control, and lack of conscious intent (Bargh, 1994). Ver-
planken, Friborg, Wang, Trafimow, and Woolf (2007, p. 526)
proposed that a habit is “a behavior that has a history of repetition,
is characterized by a lack of awareness and conscious intent, is
mentally efficient, and is sometimes difficult to control.”
As a response that occurs frequently, unintentionally, and repet-
itively to the same emotional context (depressed mood), depressive
rumination fulfils all of these conceptualizations of habit. Its gold
standard measure, the Response Styles Questionnaire (Nolen-
25
Hoeksema & Morrow, 1991), assesses the frequency of repeated
ruminative behaviors in response to the stable internal context of
feeling sad, down, or depressed. Depressive ruminators report that
rumination occurs without conscious intent, and that they are
unable to control it (Watkins & Baracaia, 2001). A self-reported
index of habitual negative thinking capturing frequency, lack of
conscious awareness, lack of conscious intent, mental efficiency,
and difficulty in control was positively correlated with both trait
and state rumination (Verplanken et al., 2007).
More explicitly, Hertel (2004) hypothesizes that rumination is a
habit of thought, noting that the initiation of an episode of rumi-
nation often occurs automatically, without conscious awareness or
effort. Hertel argues that impairments in cognitive control may
enable the emergence of such mental habits. Situations in which
stimulus and task dimensions do not inherently constrain and guide
attention (e.g., tasks with lax external control) afford habitual
rumination in patients with depression (Hertel, 1998), whereas
situations constrained to force participants to sustain attention on
particular task stimuli do not (Hertel & Rude, 1991). Outlining an
argument further developed here, Hertel (2004, p. 208) proposes
that “the best antidote to maladaptive habits is a new set of habits”
formed through repeated practice at more controlled responses.
In this article, we focus on analyzing depressive rumination as
a habit, consistent with Hertel (2004), while adding Wood and
Neal’s (2007) empirically supported theoretical framework. We
propose that applying this habit-goal framework provides a means
to understand the relationship between trait depressive rumination
(RST) and goal-oriented state repetitive thought (control theory),
helps to explain why rumination may be hard to change, and
suggests novel therapeutic implications and testable predictions.
A Habit-Goal Framework of Depressive Rumination
Wood and Neal (2007) argue that repeated performance of a
behavior in a consistent context can lead to the development of
habitual responding via (a) associative learning and (b) instrumen-
tal learning. They identify three interrelated principles that sum-
marize the development and performance of habits: (a) habits are
cued by context; (b) habit context–response associations are not
mediated by goals; and (c) because habits are acquired slowly with
experience, they are inherently conservative and tend to persist
even in the face of antagonistic current goals or intentional behav-
ior. These principles reflect properties within conditioning models
(Mackintosh, 1983; Rescorla & Wagner, 1972), in which (a) habits
are acquired when responses are contingent on stimulus contexts;
(b) S-R habit strength is not mediated by goals; (c) because habits
are acquired slowly, they extinguish or countercondition slowly.
We propose that these principles provide a useful framework to
consider how the consequences of habits may play out in depres-
sive rumination, with implications for understanding and treating
depressive rumination.
First Principle: Habits are Directly Cued by Context
as a Result of Frequent Covariation of the Behavior
With the Cueing Context
Wood and Neal (2007, p. 844) propose that “habits typically are
the residue of past goal pursuit: they arise when people repeatedly
use a particular behavioral means in particular contexts to pursue
 26 WATKINS AND NOLEN-HOEKSEMA
their goals.” Such context cues become automatic triggers for the
behavioral response, such that the behavior is controlled by the
presence or absence of the cue, rather than mediated by an implicit
or explicit goal. Thus, a goal-based response that occurs frequently
and contingently to a particular context could result in the devel-
opment of a habitual response to that context, consistent with
classic S-R theories of learning (Mackintosh, 1983). Context in-
cludes physical setting (e.g., bedroom), time of day, the people
(e.g., alone), behaviors performed prior to, and internal state (e.g.,
mood, fatigue; Ji & Wood, 2007), and any combination thereof in
which the behavior typically occurs.
The formation of associative links between the habitual re-
sponse and a potential cueing context requires the repeated coacti-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
vation of the mental representations of both response and context,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
as a consequence of contingent pairings between the response and
the cueing context across multiple episodes in the environment.
The learning of a habit (S-R association) requires the context and
the response to occur close together in time (temporal contiguity),
and for the potential habitual response (e.g., rumination) to be
contingent on the stimulus context. With repetition across multiple
episodes, incremental changes are hypothesized to occur in pro-
cedural memory, such that the common features of response and
context become learnt and encoded (Gupta & Cohen, 2002). Im-
portantly, multiple different behaviors occurring in a particular
context will weaken any reliable covariation and contingency
between any one particular habitual behavior and the contextual
cue: “The greater the numbers of behaviors linked to a given
context, the lesser the capacity for the context to cue any one
behavior directly” (Wood & Neal, 2007, p. 858).
Implications for depressive rumination. This habit-goal
framework suggests that depressive rumination will become a
habitual trait if state episodes of ruminative thought are contingent
on the context of negative mood, such that the negative mood state
becomes an automatic cue for ruminative thought. This account
provides a mechanism by which state episodes of repetitive
thought initiated by goal discrepancy (consistent with control
theory) can become habitual if such thinking is contingent on a
co-occurring context (negative affect; consistent with RST). With
respect to goal-discrepancy accounts of rumination (Martin &
Tesser, 1989, 1996; Watkins, 2008), the identification of an unre-
solved goal (or of insufficient goal progress) can directly generate
repetitive thought as a function of increased priming and accessi-
bility of goal-relevant information, and the perseverance of goal-
related thoughts (Brunstein & Gollwitzer, 1996; Goschke & Kuhl,
1993; Martin & Tesser, 1989; Zeigarnik, 1938).
Control theory accounts (Carver & Scheier, 1982) and goal
models of emotion (Oatley & Johnson-Laird, 1987) also predict
that insufficient goal progress can generate negative affect (e.g.,
perceived loss of valued goal produces sadness). These dual effects
of unresolved goals afford the opportunity for sad mood (or other
mood states) to co-occur with repetitive thought, making it a
potential mechanism for developing habitual repetitive thought.
Building on Wood and Neal (2007), this goal-habit framework
suggests that habitual rumination can emerge as an unintended
residue of goal-oriented repetitive thought.
However, the repetitive thought produced by unresolved goals is
not necessarily pathological: As noted earlier, goal-oriented repet-
itive thought can be constructive (Watkins, 2008). Thus, the habit
of pathological depressive rumination would only develop under
conditions in which the form of state repetitive thought contingent
on a stable context (e.g., depressed mood) was predominantly
passive, negative, and abstract. If, over time, an individual used a
range of different behaviors (e.g., distraction, abstract thought) in
response to unresolved goals and associated negative mood, there
may be no distinct contingency between context and response, and
no particular habit would develop. Conversely, the repeated use of
concrete action-oriented thinking in the face of goal discrepancy
might buffer against the development of unhelpful rumination-as-
a-habit (Watkins, 2008), instead forming a helpful habit of prob-
lem solving to sad mood. Thus, the co-occurrence of episodes of
goal-discrepancy rumination and negative mood is insufficient to
result in trait depressive rumination, explaining why many people
can experience occasional episodes of negative rumination without
necessarily developing depressive-rumination-as-a-habit.
This habit-goal framework suggests hypotheses about the de-
velopmental antecedents of individual differences in depressive
rumination. The formation of habitual depressive rumination is
hypothesized to involve two background conditions: (a) environ-
ments and situations that generate repeated and frequent episodes
of goal-discrepancy repetitive thought contiguous and contingent
with negative mood (and other contextual cues, e.g., behavior of
others; locations), such as the experience of having hard-to-
abandon personally important goals chronically thwarted (e.g.,
chronic stress, abuse, or trauma); and (b) a restricted repertoire of
coping behaviors and/or reduced flexibility in selecting behaviors
within this repertoire, constraining individuals toward passive and
abstract responses, which increases the likelihood of unhelpful
episodes of repetitive thought becoming contingent to negative
mood.
To date, these background conditions have tended to be studied
in isolation. A novel prediction from this framework is that the
effect of one background condition on the formation of
rumination-as-a-habit will be moderated by the presence or ab-
sence of the other condition. Being in an ongoing difficult situation
will not necessarily lead to habitual depressive rumination: Ongo-
ing difficulties may lead to episodes of state rumination as a
normal and often adaptive response to unresolved goals, but this
will not result in habitual depressive rumination unless unhelpful
episodes of repetitive thought are contingent on the associated context
(e.g., sad mood), which will be facilitated by a reduced range of
coping options. Conversely, having a tendency toward abstract
processing or passive responses will not necessarily lead to habit-
ual depressive rumination, unless there are ongoing unresolved
goals that afford the opportunity for repeated episodes of repetitive
thought in a co-occurring context, leading to repetitive thought
being contingent on negative mood. Of course, some circum-
stances may simultaneously produce both conditions (e.g., neglect,
abuse), and are thus significant risk factors for habitual rumination.
Consistent with exposure to chronically unresolved goals acting
as a setting condition, trait depressive rumination is associated
with history of sexual abuse, emotional abuse, and assault, which
reflect situations in which major personal goals of safety, affilia-
tion, and sense-making tend to be chronically unresolved (e.g.,
Conway, Mendelson, Giannopoulos, Csank, & Holm, 2004; Spa-
sojević & Alloy, 2002). Further, ongoing chronic stress and re-
duced sense of mastery in women predicted increases in trait
rumination over time (Nolen-Hoeksema, Larson, & Grayson,
1999).
 HABIT FRAMEWORK OF DEPRESSIVE RUMINATION
This account hypothesizes that developmental circumstances
and socialization processes that encourage limited and passive
responding increase the likelihood of developing habitual rumina-
tion. Such circumstances include overcontrolling parenting, re-
duced positive reinforcement, and socialization into the stereotypi-
cally feminine gender role identity.
Operant conditioning approaches hypothesize that when active
responses repeatedly fail to receive sufficient positive reinforce-
ment, the contingency learnt is that there is no value in being
active, leading to a reduced behavioral repertoire and increased
passive responding (Ferster, 1973). Thus, any environment char-
acterized by reduced positive reinforcement is hypothesized to
engender the development of passive responding. Mothers who did
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
not explicitly encourage their 5- to 7-year-old children to try
This document is copyrighted by the American Psychological Association or one of its allied publishers.
different approaches, and who told the child what to do, had
children who become more helpless, passive, and poorer at active
problem solving when faced with an upsetting and frustrating
situation (Nolen-Hoeksema, Mumme, Wolfson, & Guskin, 1995).
Consistent with such parenting being associated with habitual
rumination, in a retrospective design, Spasojević and Alloy (2002)
found that undergraduates who reported overcontrolling parenting
during their childhood reported increased trait rumination, after
controlling for levels of psychopathology. Further, in a multiwave
prospective study, Gaté et al. (2013) found that laboratory obser-
vation of reduced positive maternal behavior (indexed by fre-
quency of happy, pleasant, or caring affect, and approving, vali-
dating, affectionate, or humorous statements) when discussing and
planning a pleasant event with the child at age 12 prospectively
predicted trait depressive rumination in the child at age 15, but
only in girls. Thus, fewer validating and caring responses from
mothers to daughters when jointly planning a pleasant event pre-
dicted later rumination. This gender effect is consistent with the
commonly found gender difference in the expression of rumina-
tion, with rumination more frequent in girls and women than boys
and men (Nolen-Hoeksema et al., 1999), with this gender differ-
ence emerging in early adolescence (ages 12 to 13; Jose & Brown,
2008).
This gender difference in rumination may partially reflect de-
velopmental differences in the socialization into gender role iden-
tities. The stereotypically feminine gender role identity emphasizes
greater focus on emotions and passivity. Therefore, socialization
into this gender role is hypothesized to increase the likelihood of
developing rumination-as-a-habit. Consistent with this, in a pro-
spective longitudinal study, Broderick and Korteland (2004) found
that 9- to 12-year-old schoolchildren strongly identifying with the
feminine role identity reported significantly more trait rumination
at follow-up, regardless of gender. Cox, Mezulis, and Hyde (2010)
found that greater feminine gender role identity among children at
age 11, and maternal encouragement of emotion expression in
response to the child completing a difficult math task, each sepa-
rately mediated the relationship between child gender and the
development of trait depressive rumination at age 15, after con-
trolling for trait rumination at age 11.
An abstract style of responding to difficulties is also implicated
in impaired problem solving and reduced action orientation (Wat-
kins, 2008). In healthy individuals, there appears to be flexibility
and contextual variability in the use of abstract versus concrete
processing, with individuals adopting abstract processing as a
default mode, particularly when in positive mood, but switching
27
toward concrete processing in response to difficulties and sad
mood (Beukeboom & Semin, 2005; Watkins, 2011). This flexibil-
ity in regulating processing mode to different contexts is hypoth-
esized to be functional, with concrete processing aiding problem
solving, and thus adaptive for difficult situations, whereas abstract
processing aids longer-term planning, generalization, and consis-
tency and stability of behavior toward long-term goals across time
and situations, and is thus adaptive when things are going well.
There are individual differences in the tendency and/or ability to
regulate processing mode to context, with depressed patients im-
paired at becoming more concrete in response to sad mood relative
to healthy controls (Watkins, 2011; Watkins, Moberly, & Moulds,
2011). We hypothesize that these individual differences influence
the formation of habitual depressive rumination: Impaired regula-
tion of processing mode reduces flexibility in responding and
increases the likelihood of episodes of repetitive thought being
unhelpful, providing circumstances for such thinking to become
contingent on low mood.
Such dysregulation of processing mode could reflect deficits in
cognitive control: Individuals with deficits in executive/inhibitory
control, either because of greater cognitive load or reduced cog-
nitive resources, may be impaired at effectively regulating pro-
cessing in response to situational demands (Watkins, 2011). Dys-
regulation of processing mode could also be the consequence of
past developmental learning and parenting style. Praising or crit-
icizing a child generically (i.e., as a person, “You are a good boy”),
rather than giving feedback on the specific process and concrete
behavior (e.g., “You came up with a good solution”), can lead
children to think in abstract trait terms (for both positive and
negative events), leading to helplessness in response to later mis-
takes (Cimpian, Arce, Markman, & Dweck, 2007; Kamins &
Dweck, 1999).
The hypothesized relationship between habitual depressive ru-
mination and a restricted coping repertoire is probably bidirec-
tional. As well as a reduced range of coping strategies increasing
the chance of episodes of unhelpful repetitive thought becoming
contingent on negative mood, habitual rumination may also reduce
flexibility in selecting coping strategies. Rumination mediates the
relationship between induced negative affect and working memory
capacity (Curci, Lanciano, Soleti, & Rimé, 2013) and causally
reduces central executive functioning (Watkins & Brown, 2002),
which are implicated in flexible responding.
Second Principle: Habit Context–Response
Associations Are Not Mediated by Goals
Wood and Neal (2007) suggest that the habitual response is then
activated by the perception of the cues in the context in which the
behavior has been repeatedly performed, such that the habit is
performed without mediation by a goal. Supporting this account,
contexts, but not goals, automatically trigger overt habit perfor-
mance (Ji & Wood, 2007; Neal, Wood, Labrecque, & Lally, 2012).
For strong habits, attitudes and intentions have limited predictive
power on behavior, whereas one index of habit strength (opera-
tionalized by combining frequency and stability of behavior to the
same context) predicts how much people repeat behavior, even in
the absence of an explicit supporting goal (Ji & Wood, 2007; Neal
et al., 2012; Ouellette & Wood, 1998).
 28 WATKINS AND NOLEN-HOEKSEMA
Implications for depressive rumination. The goal-habit
framework suggests that depressive rumination would be triggered
by the relevant cueing context (e.g., negative mood) in the absence
of a personally important unresolved goal or of an intention to
think about the related concern. The habit being directly cued by
context means that depressive rumination can occur without direct
intention and without effort. Once the habit is formed, rumination
can still occur even if the unresolved goals that originally triggered
episodes of state repetitive thought have been attained or aban-
doned. Moreover, negative mood states induced by processes other
than a perceived goal discrepancy, such as direct physical or
biological effects (e.g., effect of weather, pharmacological agent),
or direct responses to stimuli (e.g., empathic response to another’s
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
distress, watching a sad movie), could then trigger rumination,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
extending the range of situations cueing this response.
This principle could be tested by adapting the methodologies
used by Ouellette and Wood (1998) and Neal et al. (2012) in which
habit strength, context, explicit goals, self-efficacy, attitudes, and
beliefs are examined as prospective predictors of the repeated
behavior (episodes of depressive rumination). Habit strength is
hypothesized to predict subsequent ruminative behavior, as as-
sessed by diary or experience sampling methods, over and above
motivational and belief variables.
Motivated cueing and instrumental learning. In addition to
associative learning, Wood and Neal (2007, p. 846) propose that
habit associations may also arise through a process in which the
reward value of response outcomes (e.g., positive affect from eating
popcorn) is conditioned onto context cues (e.g., movie theater) that
have historically accompanied the receipt of those rewards. This
analysis builds on instrumental learning theories in which habits
develop as organisms learn context–response associations in order to
obtain rewarding events.
In this process of motivated cueing, given sufficient repetitions,
the reward associated with a response is conditioned on the con-
textual cue, such that the cue then motivates the response by
activating a desire or urge toward the response as an opportunity to
acquire the associated reward.
Because depressive rumination has such clear negative conse-
quences, it is not obvious what potential reward value or instru-
mental function it may have. Nonetheless, motivated cueing could
potentially cement the strength of the habit if robust reinforcing
functions for rumination exist.
Rumination has been hypothesized to act as avoidance that is
negatively reinforced by reducing distress. Martell, Addis, and
Jacobson (2001, p. 121) proposed that “although rumination may
be experienced as aversive to the individual, it is possible that it is
maintained by the avoidance of even more aversive conditions.”
Rumination may put off overt action and avoid the risk of actual
failure and humiliation, or serve to avoid unwanted personal char-
acteristics (e.g., becoming selfish) through constant vigilance and
criticism of one’s performance. Nolen-Hoeksema et al. (2008)
hypothesized that rumination is reinforced by the reductions in
distress that come from withdrawing from aversive situations and
from being relieved of responsibility for outcomes. Although trait
rumination is correlated with avoidance (e.g., Giorgio et al., 2010),
this is not direct evidence that rumination has an avoidant function.
Rumination may also have perceived positive reward value as a
means to increase understanding and insight into self, feelings, and
problems, and to make sense of difficulties (Lyubomirsky &
Nolen-Hoeksema, 1993; Watkins & Baracaia, 2001). Believing
that rumination increases understanding predicts the maintenance
of trait rumination over 8 weeks prospectively (Kingston, Watkins,
& O’Mahen, 2013), while experimentally manipulating appraisals
that rumination is helpful versus unhelpful for understanding prob-
lems causally influenced state rumination occurring to a subse-
quent failure task (Kingston & Watkins, 2013).
However, there is no direct experimental evidence of whether
depressive rumination has any reinforcing qualities and whether
they consolidate it as a habit. Testing this functional hypothesis is
challenging: Each individual may have his or her own idiosyn-
cratic function(s) for rumination, making it hard to detect the
effects of manipulating rumination on functional outcomes within
standardized experimental conditions. Experimental studies also
need to ensure that the function under test is of value to the
participant. For example, there is no point in testing for an
avoidant function unless the study design exposes the participant
to a sufficiently negative experience that is meaningful and moti-
vating for the participant to avoid. Kingston, Watkins, and Nolen-
Hoeksema (2013) exposed undergraduates to the challenging sit-
uation of believing they had to be assertive to a friend or family
member about a difficult issue, before randomizing them to rumi-
nating about this situation versus thinking over the situation in a
nonruminative way, and assessed various indices of avoidance as
dependent variables. Consistent with Nolen-Hoeksema et al.’s
(2008) hypothesis, rumination increased justification for avoidance
by reducing sense of responsibility and increasing certainty that
efforts would be fruitless, relative to the control condition.
This design only tested whether manipulating rumination influ-
ences outcomes that enable inference about function. A necessary
further stage to test the functional hypothesis is to examine
whether such inferred functions increase the use of rumination.
One potential next step is to test whether individual differences in
the effect of manipulated rumination on justification for avoidance
prospectively predicts trait rumination.
Third Principle: Because Habits Are Acquired Slowly
With Experience, They Are Conservative and Do Not
Alter in Response to Current Goals or Infrequent
Counterhabitual Responses
Wood and Neal (2007, p. 844) argue that “habits arise from
context–response learning that is acquired slowly with experience.
As a result, habit dispositions do not alter in response to people’s
current goals or occasional counterhabitual responses. Thus, habits
possess conservative features that constrain their relation with
goals.” Hertel (2004, p. 208) noted that “habits of thought are
difficult to oppose.” Habits are insensitive and resistant to changes
in goals, outcomes, and intentions, including the devaluing of
outcomes (Dickinson, 1985).
Because control of a habitual behavior is outsourced directly to
those contextual cues that have been reliably paired with past
enactment of the behavior, habits are enacted and performed
without reference to goals or outcomes. Because a habitual behav-
ior is not especially sensitive to either outcomes or goals, it
continues to be performed even if it has unhelpful outcomes.
Further, because automatic cueing of behavior in response to
contextual cues is likely to be more immediately available and
 HABIT FRAMEWORK OF DEPRESSIVE RUMINATION
require less executive capacity than more deliberate and thoughtful
actions, it will tend to take precedence over intentional behaviors.
Thus, when in conflict with habits, goals have little impact on
actual behavior. Similarly, when goals are in concert with habits
and both dictate the same response, goals effectively become
epiphenomena.
Implications for depressive rumination. This analysis sug-
gests that once rumination becomes a habit, it will be hard to stop
even if new behavioral goals are adopted, even if it has negative
consequences, or even if the rumination is at odds with an indi-
vidual’s attitudes and intentions. This is consistent with the phe-
nomenological quality reported by patients of “not being able to
help” ruminating (Watkins & Baracaia, 2001). Intentions to stop
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ruminating alone are unlikely to be successful and habitual rumi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
nation will be resistant to change. The goal-habit framework thus
provides a partial explanation of why patients find it so difficult to
break free of rumination once it has begun, even when they
recognize it has negative consequences.
Implications for the Treatment of
Depressive Rumination
This framework offers a fresh approach to considering interven-
tions for depressive rumination when allied with the literature on
successful habit change. It suggests, for the reasons noted thus far,
that interventions focused on changing individual’s beliefs, atti-
tudes, and intentions, and on providing new information, are
unlikely to be effective at changing habitual behaviors, as empir-
ically confirmed by Verplanken and Wood (2006) and Webb and
Sheeran (2006). Changing goals, persuasion, cognitive restructur-
ing, or psychoeducation is thus hypothesized to be less effective at
changing depressive rumination. Although trait rumination has not
been frequently examined in randomized controlled trials of psy-
chological interventions, the few extant studies suggest that rumi-
nation does not change significantly in response to cognitive–
behavioral therapy (CBT; Schmaling, Dimidjian, Katon, &
Sullivan, 2002) or that elevated rumination predicts worse out-
comes to CBT (Jones, Siegle, & Thase, 2008). We hypothesize
that the more rumination is a strong (vs. weak) habit, the harder it
will be for patients to stop, and the less successful information-
based interventions will be (Webb & Sheeran, 2006).
Instead, Wood and Neal (2007, p. 860) propose that “interven-
tions to maximize habit change provide people with concrete tools
for controlling habit cueing.” The current framework suggests that
habits can be broken by altering or avoiding exposure to the cues
that trigger habit performance (Verplanken & Wood, 2006). Re-
moving contextual cues could occur as a consequence of deliberate
attempts to control cue exposure, which may require ongoing and
sustained effortful control to identify and avoid relevant cues, or
from unplanned changes that occur naturally as a consequence of
life events, such as moving to a new house or changing jobs. Wood
et al. (2005) found that when the location or environment associ-
ated with a habitual behavior (e.g., TV watching) changed when
students transferred to a new university, the behavior became less
habitual.
Hence, this framework suggests that one way to reduce depres-
sive rumination is to alter or reduce exposure to its triggering
context. When the cueing context for rumination involves a par-
ticular location (e.g., bedroom), person, preceding behavior (e.g.,
29
sitting down for a coffee after work), or environmental feature
(e.g., sad music), environmental modification to remove or avoid
the triggering context is hypothesized to interrupt depressive ru-
mination.
One context identified by Hertel (1998, 2004) is situations that
lack external control over attention and that therefore afford mind
wandering. Habitual rumination is more likely to occur when the
current activity or task does not externally constrain attentional
focus. Habitual rumination will be triggered during tasks that are
open-ended, unstructured, and rely on self-control, and when an
individual has an “empty” unplanned period of time, such as lying
in bed doing nothing. Introducing more structure and constraints
into activities may change the context to reduce the triggering of
habitual rumination.
A long-term reduction in rumination is possible if contextual
cues can be permanently removed or consistently avoided. How-
ever, this strategy is dependent on individuals being able to accu-
rately identify cues for habits, requiring a detailed functional and
contextual analysis to determine the triggering cues for rumination,
although this may not always be successful. Consistent with this
hypothesis, an intervention incorporating functional analysis and
stimulus control (rumination-focused CBT, Watkins et al., 2007;
utilizing behavioral activation principles, Martell et al., 2001)
successfully reduced depression and rumination (Watkins et al.,
2011) in a randomized controlled trial. However, this treatment
had multiple components, and the effectiveness of stimulus control
has not been tested as a separate intervention. This framework
predicts that stimulus control alone will reduce trait depressive
rumination when there are identifiable environmental cues tracta-
ble to modification.
However, when the cueing context is an internal state, such as a
negative mood, as typical for depressive rumination, it is less
straightforward to permanently alter or avoid the context. This
framework predicts that reducing the negative mood would, at
least temporarily, reduce the enactment of the depressive rumina-
tion habit. This framework predicts that any intervention that
reduces negative mood, whether antidepressant medication, psy-
chotherapy, distraction, or change in circumstances, would reduce
the expression of habitual rumination by removing one cue that
triggers the ruminative habit. Consistent with this, self-reported
trait rumination reduces as acute symptoms of depression diminish
(Roberts, Gilboa, & Gotlib, 1998).
However, a key implication of this framework is that removing
the context does not change the context–response association
underlying the habit but only limits its expression. On any subse-
quent return of the triggering context, such as another period of sad
or depressed mood, the ruminative habit would reoccur. This
analysis therefore predicts that any interventions that improve
mood state will temporarily disrupt depressive rumination, but
only interventions that directly modify the context–response asso-
ciation will lead to long-standing reductions in depressive rumi-
nation and associated depressive vulnerability. In sum, improve-
ments in mood can temporarily reduce rumination by removing a
potential cue for the habit. However, because of their conservative
nature, habits are easily reactivated and do not change unless the
S-R association itself is counterconditioned.
This account can partially explain the high rates of relapse and
recurrence (50% to 80%) and the limited sustained recovery (less
than one third at 1 year) obtained for even our best interventions
 30 WATKINS AND NOLEN-HOEKSEMA
for depression (Hollon et al., 2002). We hypothesize that many
interventions for depression improve symptoms in the short term
through positive expectancy, remoralization, increased activation,
and therapist support, but because they do not directly target
depressogenic habits like rumination, they leave patients vulnera-
ble to relapse. A novel testable prediction is that posttreatment
measures indexing habit strength for rumination (and alternative
coping strategies) will better prospectively predict sustained re-
covery for depression than posttreatment beliefs, expectancies, and
symptoms. One such measure is the Self-Report Habit Index,
which measures the frequency and automaticity of behavior, has
good reliability and validity predicting actual habitual behavior
(Orbell & Verplanken, 2010; Verplanken & Orbell, 2003; Ver-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
planken et al., 2007), and is adaptable for different behaviors.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This key implication suggests that directly targeting the auto-
matic context–response association will improve the efficacy and
durability of interventions for rumination and depression. Marteau,
Hollands, and Fletcher (2012) noted that most interventions aimed
at changing health-related behaviors target reflective processes
rather than automatic habits, and that this may explain the relative
inefficacy of these interventions. We propose that this analysis is
equally pertinent to improving mental health, because automatic
mental habits (e.g., rumination) are a central process within psy-
chopathology.
To be effective at modifying the context–response association,
and thereby reducing habitual rumination, the unhelpful rumina-
tive response to the cueing context needs to be replaced with a
more helpful response, with the patient in effect learning a new
more adaptive habit. This echoes Hertel’s (2004, p. 209) recom-
mendation toward “the training of new habits through practice in
control.” Changing these habits requires “counterconditioning or
training to associate the triggering cue with a response that is
incompatible and thereby conflicts with the unwanted habit”
(Wood & Neal, 2007, p. 859). Such an intervention involves
repeated practice at utilizing an alternative incompatible coping
strategy (e.g., concrete thinking, relaxation) in response to the
identified triggering contextual cue (e.g., sad mood) to develop the
new context–response association. Critically, this habit analysis
predicts that systematically counterconditioning alternative func-
tional responses to the cueing context for rumination will improve
prevention of relapse and recurrence, whether as a distinct acute
treatment or following successful acute treatment of depression.
This analysis suggests that traditional CBT approaches such as
thought challenging will not necessarily be effective at changing
habitual rumination. Successfully challenging a thought or belief
would not change the underlying S-R habit. However, thought
challenging could potentially be an effective intervention for ru-
mination if repetitively trained as an alternative coping skill in
response to cues for rumination.
By the same logic, cognitive bias modification (CBM) ap-
proaches used to change negative biases (Hertel & Mathews, 2011)
are likely to have value in reducing habitual rumination. CBM
typically involves the repeated training of desired responses; in-
corporates programmed contingencies between stimuli, responses,
and desirable outcomes (e.g., faster task performance); and is
assumed to involve the same associative and instrumental learning
processes as habit formation, and to work on processes that are
often outside conscious awareness. Thus, Hertel and Mathews
(2011, p. 528) argued that “in interpretation retraining, contingen-
cies between ambiguous situations and their preexperimental res-
olutions are, in effect, counterconditioned.” There is now initial
evidence supporting the assumption that CBM influences mental
habits: Hertel, Holmes, and Benbow (in press) used a variant of
Jacoby’s (1996) process dissociation approach to confirm that
interpretation bias training changed automatic, habitual processes
rather than controlled recollection. Further, there is growing evi-
dence that rumination is associated with biases in disengaging
attention from negative information (Koster, De Lissnyder, Derak-
shan, & De Raedt, 2011), abstract construal (Watkins, 2008), and
negative interpretative bias (Mor, Hertel, Ngo, Schachar, & Redak,
in press). CBM approaches therefore seem particularly pertinent to
changing mental habits. The effects of CBM may perhaps be
enhanced by training adaptive biases in response to contextual
cues for unwanted habits, with Hertel and Mathews (2011, p. 529)
arguing that “training should incorporate situational cues that
typically provoke the deployment of negative biases in real life.”
Practice may need to be more extensive than typically employed
in current psychological interventions. Lally, Van Jaarsveld, Potts,
and Wardle (2010) examined the development of habits for vol-
unteers practicing an eating, drinking, or exercise behavior daily in
the same context. Self-reported automaticity of the trained re-
sponse increased with repetition in the consistent context, but took
a median of 66 repetitions to reach a training plateau, with con-
siderable individual variation. Moreover, to ensure the required
cueing context (e.g., mood state) is active prior to practicing the
alternative response, treatment may require active attempts to
engineer the cueing context in therapy sessions, such as mood
challenges or in imago exposure to cues. Such challenges provide
an opportunity to test whether the habitual response to the cue has
changed. This framework suggests that assessing the habitual
nature of rumination and alternative coping strategies during ther-
apy has clinical value to determine whether a new habit has
successfully been trained. Practice and training would optimally
continue until the new adaptive habit reaches a self-reported as-
ymptote.
To our knowledge, there has only been one explicit attempt to
test repeated training of an alternative response to the contextual
cues for rumination as an intervention. Building on the evidence
that concrete processing is an adaptive alternative to depressive
rumination (Watkins, 2008), a proof-of-principle randomized con-
trolled trial found that training dysphoric individuals to be more
concrete when faced with difficulties reduced depression, anxiety,
and rumination relative to a no-treatment control (Watkins, Baey-
ens, & Read, 2009). Concreteness training is a form of CBM that
involves repeated practice at focusing on the specific details,
context, and sequence of difficult events, and asking “How the
event happened?” During the training, concrete thinking is explic-
itly used as a response to identified warning signs for rumination
and is practiced repeatedly with audio-recorded exercises.
In a Phase II randomized controlled trial, concreteness training
was found to be superior to treatment-as-usual in reducing rumi-
nation, worry, and depression in patients with major depression
recruited in primary care (Watkins et al., 2012). A control relax-
ation training condition, in which patients practiced progressive
relaxation as an alternative response to cues for rumination, also
significantly reduced depression relative to treatment-as-usual and
did not significantly differ from concreteness training. These re-
sults suggest that repeated training in any functional alternative
 HABIT FRAMEWORK OF DEPRESSIVE RUMINATION
response to rumination may be an effective intervention, consistent
with the current habit analysis.
Relative to relaxation training, concreteness training signifi-
cantly reduced trait rumination, suggesting that interposing a cog-
nitive alternative may be more effective at reducing habitual
rumination, perhaps because it facilitates adaptive repetitive
thought in place of maladaptive repetitive thought as an example
of transfer-appropriate processing (see Hertel & Mathews, 2011,
for full discussion). Moreover, consistent with the hypothesis that
concreteness training works through establishing an alternative
habitual response to replace rumination, the benefits of concrete-
ness training were significantly stronger in those patients who
reported that the practiced self-help response had become habitual
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
(Watkins et al., 2012).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This habit analysis has implications for the generalizability and
sustainability of clinical improvement. Several theorists (Bouton,
2000; Hertel & Mathews, 2011; Wood & Neal, 2007) have noted
that new patterns of associative learning (i.e., new functional S-R
associations) intended to replace older patterns of associations are
inherently unstable, enabling the original patterns of learning to
recur under certain circumstances. These recurrences occur be-
cause the original learning that underpins the unwanted habit
remains intact following counterconditioning or extinction, with
new learning overlaid on the original learning. In these circum-
stances, a conditioned cue (e.g., sad mood) is in the position of
potentially signaling two available cue–response associations (e.g.,
rumination vs. concrete thinking). Which association is predomi-
nant and which behavior it evokes will be determined by the
current occasion-setting context. If there are differences between
the setting contexts under which the original learning and subse-
quent counterconditioning occur, shifts back to the original setting
context can renew the original relationship and original contingen-
cies. A key implication of considering rumination-as-habit is that
despite conditioning of new responses, the old habitual response
can recur under particular circumstances, potentially leading to
relapse. Thus, even after successful counterconditioning of a new
habitual response to sad mood, there may be a lapse back into
rumination when the conditioned cue is encountered in a different
setting context.
Moreover, because the original unwanted habit typically results
from a large number of learning trials in which the triggering cue
and the habitual response were paired over many different settings
and situations, more setting contexts will be associated with the
original unwanted habit than the replacement habit, which may be
constrained to fewer contexts (e.g., the therapeutic setting). One
way to counter these renewal effects is to promote the retrieval of
the new learning (i.e., the new habit) through explicit retrieval cues
and reminders throughout and after the training (e.g., flashcards,
texts from therapist; Hertel & Mathews, 2011). Another approach
is to increase the generalization of the counterconditioning by
conducting the training across different and varied contexts (e.g.,
using multiple variants of the cueing context; pairing sad mood
with new response in different settings, such at home, at the office,
when alone, and with other people; see Hertel & Mathews, 2011,
for further detail).
In addition, the performance of unwanted habits becomes more
likely when individuals are under cognitive load, distracted, pre-
occupied, or stressed (Botvinick & Bylsma, 2005; Schwabe &
Wolf, 2011). We hypothesize that the habitual ruminative response
31
is more likely to be triggered under circumstances of stress, pres-
sure, or fatigue. The ruminative response is thus likely to be
hardest to stop at those times when individuals most need to stop
it.
The habit literature suggests that a further way to control habit-
ual rumination is by “inhibiting the performance of the habitually
cued response once it has been activated” (Wood & Neal, 2007, p.
854). Quinn, Pascoe, Wood, and Neal (2010) provided evidence
from episode-sampling studies that vigilant monitoring (thinking
“don’t do it”; watching for slipups) was the best strategy for
self-control of strong habits in daily life, relative to do nothing,
stimulus control, and distraction. However, inhibition of habitual
behaviors requires the availability of sufficient levels of effortful
self-control: On days when students had a sustained load on
self-control, they were more likely to fail at overriding habitual
behaviors but not nonhabitual behaviors (Neal & Wood, 2007).
Moreover, self-control is a finite resource that is depleted follow-
ing effortful attempts to inhibit or control thoughts, feelings, and
actions (Muraven & Baumeister, 2000). It is likely to be reduced
in patients with depression (Joormann, 2010). This raises questions
over the long-term sustainability of effortful inhibition of strong
habits in daily life. Inhibition of habitual responses may most
usefully contribute to long-term change by interrupting a habitual
response to enable a new pattern of responding to be countercon-
ditioned.
Focusing on self-control and inhibition in habits is particularly
relevant to habitual rumination because difficulties in turning
attention away from negative stimuli are characteristic of rumina-
tion. It remains unresolved to what extent these difficulties reflect
difficulties in disengaging attention from negative material (Koster
et al., 2011) and/or difficulties in inhibiting no-longer-relevant
material (Joormann, 2006, 2010). Consistent with the former,
habitual rumination is correlated with selective attentional bias
toward sad faces (Joormann, 2004; Joormann, Dkane, & Gotlib,
2006) and negative words (Donaldson, Lam, & Mathews, 2007),
although the causal direction of this relationship is not yet deter-
mined. Consistent with the latter, habitual rumination is correlated
with poor performance on inhibition-related tasks, including neg-
ative affective priming (e.g., Joormann, 2006) and suppression-
induced forgetting (Hertel & Gerstle, 2003), and with difficulties
in removing now-irrelevant information from working memory
(Joormann & Gotlib, 2008; Joormann, Levens, & Gotlib, 2011).
Interventions that strengthen inhibitory control or effortful self-
control may thus have potential to reduce the expression of habit-
ual rumination. There is preliminary evidence that it may be
possible to improve working memory capacity through repeated
training on cognitive tasks (e.g., Jaeggi, Buschkuehl, Jonides, &
Perrig, 2008), and that repeated practice can strengthen self-
regulatory control (Muraven & Baumeister, 2000). If robust and
replicable, such cognitive training has potential to improve self-
control of habitual rumination as part of a treatment intervention,
particularly if combined with counterconditioning of alternative
responses.
We note that this habit analysis is not necessarily unique to
rumination. Because the theoretical framework is based on generic
concepts of conditioning and habit formation, the principles artic-
ulated could be applied to other behaviors and emotional regula-
tion strategies. Nevertheless, our focus in this article is to consider
the value of applying this approach to habitual rumination, and to
 32 WATKINS AND NOLEN-HOEKSEMA
reconcile state episodes of repetitive thought engendered by goal
discrepancy with trait habitual tendency toward rumination. We
further note that rumination may be harder to break than other
habits, such as those involving overt behavioral routines. First,
rumination tends to be triggered by internal states, which makes it
harder to successfully use stimulus and environmental control.
Second, changing habits requires a concerted and deliberate at-
tempt at self-control in the form of generating an alternative
nonhabitual response. However, habitual rumination reduces an
individuals’ ability to exert such self-control because of its impact
on executive control and working memory (e.g., Curci et al., 2013;
Watkins & Brown, 2002). Thus, unlike many habits, by dint of
occupying mental resources, rumination may be harder to change,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
indicating the importance of training ruminators into new habits,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
for example, by CBM.
Conclusion
We have articulated a habit-goal framework of depressive ru-
mination that elaborates on conceptualizations of depressive rumi-
nation as a mental habit (Hertel, 2004) and recent work consider-
ing the relationship between goals and habits (Wood & Neal,
2007). This approach links RST and control theory by explicating
the mechanisms and conditions by which state episodes of goal-
discrepancy repetitive thought could develop into habitual depres-
sive rumination. Further, this habit account generates new insights
and testable predictions concerning the developmental antecedents
of individual differences in rumination. The habitual nature of
depressive rumination provides a partial account of why it can be
so difficult to break free from rumination. Understanding the
possible conditions and mechanisms that influence habit formation
and habit change provides a new perspective on how to optimize
psychological interventions for depressive rumination.
This manuscript reflects an ongoing collaboration between the
authors, with an original draft complete when Susan Nolen-
Hoeksema tragically passed away. It is hoped that, in some small
part, this final manuscript pays tribute to Susan, her wisdom and
thoroughness, her friendship, and her work by elaborating on her
seminal conceptualization of rumination and showing new insights
and value. Susan was first in research, first in teaching, and first in
the hearts of her peers and students; as a collaborator, she made
ideas flow in a way that was exciting and exhilarating: She will be
greatly missed.
References
Bargh, J. A. (1994). The four horsemen of automaticity: Awareness,
intention, efficiency, and control in social cognition. In R. S. Wyer &
T. K. Srull (Eds.), Handbook of social cognition (Vol. 1, pp. 1– 40).
Hillsdale, NJ: Erlbaum.
Beukeboom, C. J., & Semin, G. R. (2005). Mood and representations of
behaviour: The how and why. Cognition and Emotion, 19, 1242–1251.
doi:10.1080/02699930500203369
Botvinick, M. M., & Bylsma, L. M. (2005). Distraction and action slips in
an everyday task: Evidence for a dynamic representation of task context.
Psychonomic Bulletin & Review, 12, 1011–1017. doi:10.3758/
BF03206436
Bouton, M. E. (2000). A learning theory perspective on lapse, relapse, and
the maintenance of behavior change. Health Psychology, 19, 57– 63.
doi:10.1037/0278-6133.19.Suppl1.57
Broderick, P. C., & Korteland, C. (2004). A prospective study of rumina-
tion and depression in early adolescence. Clinical Child Psychology and
Psychiatry, 9, 383–394. doi:10.1177/1359104504043920
Brunstein, J. C., & Gollwitzer, P. M. (1996). Effects of failure on subse-
quent performance: The importance of self-defining goals. Journal of
Personality and Social Psychology, 70, 395– 407. doi:10.1037/0022-
3514.70.2.395
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual
framework for personality-social, clinical, and health psychology. Psy-
chological Bulletin, 92, 111–135. doi:10.1037/0033-2909.92.1.111
Cimpian, A., Arce, H.-M. C., Markman, E. M., & Dweck, C. S. (2007).
Subtle linguistic cues affect children’s motivation. Psychological Sci-
ence, 18, 314 –316. doi:10.1111/j.1467-9280.2007.01896.x
Conway, M., Mendelson, M., Giannopoulos, C., Csank, P. A. R., & Holm,
S. L. (2004). Childhood and adult sexual abuse, rumination on sadness,
and dysphoria. Child Abuse & Neglect, 28, 393– 410. doi:10.1016/j
.chiabu.2003.05.004
Cox, S. J., Mezulis, A. H., & Hyde, J. S. (2010). The influence of child
gender role and maternal feedback to child stress on the emergence of
the gender difference in pathological rumination in adolescence. Devel-
opmental Psychology, 46, 842– 852. doi:10.1037/a0019813
Curci, A., Lanciano, T., Soleti, E., & Rimé, B. (2013). Negative emotional
experiences arouse rumination and affect working memory capacity.
Emotion, 13, 867– 880. doi:10.1037/a0032492
Dickinson, A. (1985). Actions and habits: The development of behavioural
autonomy. Philosophical Transactions of the Royal Society of London:
Series B: Biological Sciences, 308, 67–78. doi:10.1098/rstb.1985.0010
Donaldson, C., Lam, D., & Mathews, A. (2007). Rumination and attention
in major depression. Behaviour Research and Therapy, 45, 2664 –2678.
doi:10.1016/j.brat.2007.07.002
Ehring, T., & Watkins, E. R. (2008). Repetitive negative thinking as a
transdiagnostic process. International Journal of Cognitive Therapy, 1,
192–205. doi:10.1521/ijct.2008.1.3.192
Ferster, C. B. (1973). Functional analysis of depression. American Psy-
chologist, 28, 857– 870. doi:10.1037/h0035605
Gaté, M. A., Watkins, E. R., Simmons, J. G., Byrne, M. L., Schwartz,
O. S., Whittle, S., . . . Allen, N. B. (2013). Maternal parenting behaviors
and adolescent depression: The mediating role of rumination. Journal of
Clinical Child and Adolescent Psychology, 42, 348 –357. doi:10.1080/
15374416.2012.755927
Giorgio, J. M., Sanflippo, J., Kleiman, E., Reilly, D., Bender, R. E.,
Wagner, C. A., . . . Alloy, L. B. (2010). An experiential avoidance
conceptualization of pathological rumination: Three tests of the model.
Behaviour Research and Therapy, 48, 1021–1031. doi:10.1016/j.brat
.2010.07.004
Goschke, T., & Kuhl, J. (1993). Representation of intentions: Persisting
activation in memory. Journal of Experimental Psychology: Learning,
Memory, & Cognition, 19, 1211–1226. doi:10.1037/0278-7393.19.5
.1211
Gupta, P., & Cohen, N. J. (2002). Theoretical and computational analysis
of skill learning, repetition priming, and procedural memory. Psycho-
logical Review, 109, 401– 448. doi:10.1037/0033-295X.109.2.401
Harvey, A. G., Watkins, E., Mansell, W., & Shafran, R. (2004). Cognitive
behavioural processes across psychological disorders. Oxford, UK:
Oxford University Press.
Hertel, P. T. (1998). Relation between rumination and impaired memory in
dysphoric moods. Journal of Abnormal Psychology, 107, 166 –172.
doi:10.1037/0021-843X.107.1.166
Hertel, P. T. (2004). Memory for emotional and non-emotional events in
depression: A question of habit. In D. Reisberg & P. Hertel (Eds.),
Memory and emotion (pp. 186 –216). New York, NY: Oxford University
Press. doi:10.1093/acprof:oso/9780195158564.003.0006
 HABIT FRAMEWORK OF DEPRESSIVE RUMINATION
Hertel, P. T., & Gerstle, M. (2003). Depressive deficits in forgetting.
Psychological Science, 14, 573–578. doi:10.1046/j.0956-7976.2003
.psci_1467.x
Hertel, P. T., Holmes, M., & Benbow, A. (in press). Interpretative habit is
strengthened by cognitive bias modification. Memory.
Hertel, P. T., & Mathews, A. (2011). Cognitive bias modification: Past
perspectives, current findings, and future applications. Perspectives on
Psychological Science, 6, 521–536. doi:10.1177/1745691611421205
Hertel, P. T., & Rude, S. S. (1991). Depressive deficits in memory:
Focusing attention improves subsequent recall. Journal of Experimental
Psychology: General, 120, 301–309. doi:10.1037/0096-3445.120.3.301
Hollon, S. D., Munoz, R. F., Barlow, D. H., Beardslee, W. R., Bell, C. C.,
Bernal, G., . . . Sommers, D. (2002). Psychosocial intervention devel-
opment for the prevention and treatment of depression: Promoting
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
innovation and increasing access. Biological Psychiatry, 52, 610 – 630.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
doi:10.1016/S0006-3223(02)01384-7
Hull, C. L. (1943). Principles of behavior. New York, NY: Appleton-
Century-Crofts.
Jacoby, L. L. (1996). Dissociating automatic and consciously controlled
effects of study/test compatibility. Journal of Memory and Language,
35, 32–52. doi:10.1006/jmla.1996.0002
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Im-
proving fluid intelligence with training on working memory. PNAS
Proceedings of the National Academy of Sciences of the United States of
America, 105, 6829 – 6833. doi:10.1073/pnas.0801268105
Ji, M. F., & Wood, W. (2007). Purchase and consumption habits: Not
necessarily what you intend. Journal of Consumer Psychology, 17,
261–276. doi:10.1016/S1057-7408(07)70037-2
Jones, N. P., Siegle, G. J., & Thase, M. E. (2008). Effects of rumination
and initial severity on remission to cognitive therapy for depression.
Cognitive Therapy and Research, 32, 591– 604. doi:10.1007/s10608-
008-9191-0
Joormann, J. (2004). Attentional bias in dysphoria: The role of inhibitory
processes. Cognition and Emotion, 18, 125–147. doi:10.1080/
02699930244000480
Joormann, J. (2006). The relation of rumination and inhibition: Evidence
from a negative priming task. Cognitive Therapy and Research, 30,
149 –160. doi:10.1007/s10608-006-9035-8
Joormann, J. (2010). Cognitive inhibition and emotion regulation in de-
pression. Current Directions in Psychological Science, 19, 161–166.
doi:10.1177/0963721410370293
Joormann, J., Dkane, M., & Gotlib, I. H. (2006). Adaptive and maladaptive
components of rumination? Diagnostic specificity and relation to de-
pressive biases. Behavior Therapy, 37, 269 –280. doi:10.1016/j.beth
.2006.01.002
Joormann, J., & Gotlib, I. H. (2008). Updating the contents of working
memory in depression: Interference from irrelevant negative material.
Journal of Abnormal Psychology, 117, 182–192. doi:10.1037/0021-
843X.117.1.182
Joormann, J., Levens, S. M., & Gotlib, I. H. (2011). Sticky thoughts:
Depression and rumination are associated with difficulties manipulating
emotional material in working memory. Psychological Science, 22,
979 –983. doi:10.1177/0956797611415539
Jose, P., & Brown, I. (2008). When does the association between gender
and rumination begin? Gender and age differences in the use of rumi-
nation by adolescents. Journal of Youth and Adolescence, 37, 180 –192.
doi:10.1007/s10964-006-9166-y
Kamins, M. L., & Dweck, C. S. (1999). Person versus process praise and
criticism: Implications for contingent self-worth and coping. Develop-
mental Psychology, 35, 835– 847. doi:10.1037/0012-1649.35.3.835
Kingston, R. E. F., & Watkins, E. R. (2013). Believing that rumination is
helpful increases rumination in response to stress. Manuscript submitted
for publication.
33
Kingston, R. E. F., Watkins, E. R., & Nolen-Hoeksema, S. (2013). Inves-
tigating the reinforcing functions of depressive rumination. Manuscript
submitted for publication.
Kingston, R. E. F., Watkins, E. R., & O’Mahen, H. A. (2013). A prospec-
tive examination of risk factors for repetitive negative thought. Manu-
script submitted for publication.
Koster, E. H. W., De Lissnyder, E., Derakshan, N., & De Raedt, R. (2011).
Understanding pathological rumination from a cognitive science per-
spective: The impaired disengagement hypothesis. Clinical Psychology
Review, 31, 138 –145. doi:10.1016/j.cpr.2010.08.005
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. doi:10.1002/
ejsp.674
Lyubomirsky, S., & Nolen-Hoeksema, S. (1993). Self-perpetuating prop-
erties of dysphoric rumination. Journal of Personality and Social Psy-
chology, 65, 339 –349.
Mackintosh, N. J. (1983). Conditioning and associative learning. Oxford,
UK: Oxford University Press.
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. doi:10.1126/science.1226918
Martell, C. R., Addis, M. E., & Jacobson, N. S. (2001). Depression in
context: Strategies for guided action. New York, NY: Norton.
Martin, L. L., & Tesser, A. (1989). Toward a motivational and structural
theory of ruminative thought. In J. S. Uleman & J. A. Bargh (Eds.),
Unintended thought (pp. 306 –326). New York, NY: Guilford Press.
Martin, L. L., & Tesser, A. (1996). Some ruminative thoughts. In R. S.
Wyer (Ed.), Ruminative thoughts. Advances in social cognition (Vol. 9,
pp. 1– 47). Hillsdale, NJ: Erlbaum.
Mor, N., Hertel, P. T., Ngo, T. A., Schachar, T., & Redak, S. (in press).
Interpretation bias characterizes trait rumination. Journal of Behavior
Therapy and Experimental Psychiatry.
Muraven, M., & Baumeister, R. F. (2000). Self-regulation and depletion of
limited resources: Does self-control resemble a muscle? Psychological
Bulletin, 126, 247–259. doi:10.1037/0033-2909.126.2.247
Neal, D. T., & Wood, W. (2007). Automaticity in situ: Direct context cuing
of habits in daily life. In J. A. Bargh, P. Gollwitzer, & E. Morsella (Eds.),
Psychology of action. Volume 2: Mechanisms of human action (pp.
442– 457). London, UK: Oxford University Press.
Neal, D. T., Wood, W., Labrecque, J. S., & 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. doi:10.1016/
j.jesp.2011.10.011
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on
the duration of depressive episodes. Journal of Abnormal Psychology,
100, 569 –582. doi:10.1037/0021-843X.100.4.569
Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders
and mixed anxiety/depressive symptoms. Journal of Abnormal Psychol-
ogy, 109, 504 –511. doi:10.1037/0021-843X.109.3.504
Nolen-Hoeksema, S., Larson, J., & Grayson, C. (1999). Explaining the
gender difference in depressive symptoms. Journal of Personality and
Social Psychology, 77, 1061–1072. doi:10.1037/0022-3514.77.5.1061
Nolen-Hoeksema, S., & Morrow, J. (1991). A prospective study of depres-
sion and posttraumatic stress symptoms after a natural disaster: The
1989 Loma-Prieta earthquake. Journal of Personality and Social Psy-
chology, 61, 115–121. doi:10.1037/0022-3514.61.1.115
Nolen-Hoeksema, S., Mumme, D., Wolfson, A., & Guskin, K. (1995).
Helplessness in children of depressed and nondepressed mothers. De-
velopmental Psychology, 31, 377–387. doi:10.1037/0012-1649.31.3.377
Nolen-Hoeksema, S., & Watkins, E. R. (2011). A heuristic for transdiag-
nostic models of psychopathology: Explaining multifinality and diver-
gent trajectories. Perspectives on Psychological Science, 6, 589 – 609.
doi:10.1177/1745691611419672
 34 WATKINS AND NOLEN-HOEKSEMA
Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Rethinking
rumination. Perspectives on Psychological Science, 3, 400 – 424. doi:
10.1111/j.1745-6924.2008.00088.x
Oatley, K., & Johnson-Laird, P. N. (1987). Towards a cognitive theory of
emotions. Cognition and Emotion, 1, 29 –50. doi:10.1080/
02699938708408362
Orbell, S., & Verplanken, B. (2010). The automatic component of habit in
health behavior: Habit as cue-contingent automaticity. Health Psychol-
ogy, 29, 374 –383. doi:10.1037/a0019596
Ouellette, J. A., & Wood, W. (1998). Habit and intention in everyday life:
The multiple processes by which past behavior predicts future behavior.
Psychological Bulletin, 124, 54 –74. doi:10.1037/0033-2909.124.1.54
Quinn, J. M., Pascoe, A., Wood, W., & Neal, D. T. (2010). Can’t control
yourself? Monitor those bad habits. Personality and Social Psychology
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Bulletin, 36, 499 –511. doi:10.1177/0146167209360665
Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian condi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
tioning: Variations in the effectiveness of reinforcement and nonrein-
forcement. In A. H. Black and W. F. Prokasy (Eds.), Classical condi-
tioning II: Current research and theory (pp. 64 –99). New York, NY:
Appleton-Century-Crofts.
Roberts, J. E., Gilboa, E., & Gotlib, I. H. (1998). Ruminative response style
and vulnerability to episodes of dysphoria: Gender, neuroticism, and
episode duration. Cognitive Therapy and Research, 22, 401– 423. doi:
10.1023/A:1018713313894
Schmaling, J., Dimidjian, S., Katon, W., & Sullivan, M. (2002). Response
styles among patients with minor depression and dysthymia in primary
care. Journal of Abnormal Psychology, 111, 350 –356. doi:10.1037/
0021-843X.111.2.350
Schwabe, L., & Wolf, O. T. (2011). Stress-induced modulation of instru-
mental behaviour: From goal-directed to habitual control of action.
Behavioural Brain Research, 219, 321–328. doi:10.1016/j.bbr.2010.12
.038
Spasojević, J., & Alloy, L. B. (2002). Who becomes a depressive rumina-
tor? Developmental antecedents of ruminative response style. Journal of
Cognitive Psychotherapy: An International Quarterly, 16, 405– 419.
doi:10.1891/088983902780935713
Verplanken, B. (2006). Beyond frequency: Habit as mental construct.
British Journal of Social Psychology, 45, 639 – 656. doi:10.1348/
014466605X49122
Verplanken, B., Friborg, O., Wang, C. E., Trafimow, D., & Woolf, K.
(2007). Mental habits: Metacognitive reflection on negative self-
thinking. Journal of Personality and Social Psychology, 92, 526 –541.
doi:10.1037/0022-3514.92.3.526
Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: A
self-report index of habit strength. Journal of Applied Social Psychol-
ogy, 33, 1313–1330. doi:10.1111/j.1559-1816.2003.tb01951.x
Verplanken, B., & Wood, W. (2006). Interventions to break and create
consumer habits. Journal of Public Policy & Marketing, 25, 90 –103.
doi:10.1509/jppm.25.1.90
Watkins, E. R. (2008). Constructive and unconstructive repetitive thought.
Psychological Bulletin, 134, 163–206. doi:10.1037/0033-2909.134.2
.163
Watkins, E. R. (2011). Dysregulation in level of goal and action identifi-
cation across psychological disorders. Clinical Psychology Review, 31,
260 –278. doi:10.1016/j.cpr.2010.05.004
View publication stats
Watkins, E. R., Baeyens, C. B., & Read, R. (2009). Concreteness training
reduces dysphoria: Proof-of-principle for repeated cognitive bias mod-
ification in depression. Journal of Abnormal Psychology, 118, 55– 64.
doi:10.1037/a0013642
Watkins, E., & Baracaia, S. (2001). Why do people ruminate in dysphoric
moods? Personality and Individual Differences, 30, 723–734. doi:
10.1016/S0191-8869(00)00053-2
Watkins, E., & Brown, R. G. (2002). Rumination and executive function in
depression: An experimental study. Journal of Neurology, Neurosurgery
& Psychiatry, 72, 400 – 402. doi:10.1136/jnnp.72.3.400
Watkins, E. R., Moberly, N. J., & Moulds, M. (2008). Processing mode
causally influences emotional reactivity: Distinct effects of abstract
versus concrete construal on emotional response. Emotion, 8, 364 –378.
doi:10.1037/1528-3542.8.3.364
Watkins, E. R., Moberly, N. J., & Moulds, M. L. (2011). When the ends
outweigh the means: Mood and level-of-identification in depression.
Cognition and Emotion, 25, 1214 –1227. doi:10.1080/02699931.2010
.532389
Watkins, E. R., & Moulds, M. (2005). Distinct modes of ruminative
self-focus: Impact of abstract versus concrete rumination on problem
solving in depression. Emotion, 5, 319 –328. doi:10.1037/1528-3542.5
.3.319
Watkins, E. R., Mullan, E. G., Wingrove, J., Rimes, K., Steiner, H.,
Bathurst, N., . . . Scott, J. (2011). Rumination-focused cognitive behav-
iour therapy for residual depression: Phase II randomized controlled
trial. The British Journal of Psychiatry, 199, 317–322. doi:10.1192/bjp
.bp.110.090282
Watkins, E., Scott, J., Wingrove, J., Rimes, K., Bathurst, N., Steiner, H.,
. . . Malliaris, Y. (2007). Rumination-focused cognitive-behaviour ther-
apy for residual depression: A case series. Behaviour Research and
Therapy, 45, 2144 –2154. doi:10.1016/j.brat.2006.09.018
Watkins, E. R., Taylor, R. S., Byng, R., Baeyens, C. B., Read, R., Pearson,
K., & Watson, L. (2012). Guided self-help concreteness training as an
intervention for major depression in primary care: A Phase II random-
ized controlled trial. Psychological Medicine, 42, 1359 –1371. doi:
10.1017/S0033291711002480
Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions
engender behaviour change? A meta-analysis of the experimental evi-
dence. Psychological Bulletin, 132, 249 –268. doi:10.1037/0033-2909
.132.2.249
Wood, W., & Neal, D. T. (2007). A new look at habits and the habit-goal
interface. Psychological Review, 114, 843– 863. doi:10.1037/0033-295X
.114.4.843
Wood, W., Tam, L., & Witt, M. G. (2005). Changing circumstances,
disrupting habits. Journal of Personality and Social Psychology, 88,
918 –933. doi:10.1037/0022-3514.88.6.918
Zeigarnik, B. (1938). On finished and unfinished tasks. In W. D. Ellis
(Ed.), A source book of gestalt psychology (pp. 300 –314). New York,
NY: Harcourt, Brace, & World. doi:10.1037/11496-025
Received June 29, 2013
Revision received November 21, 2013
Accepted December 2, 2013 䡲