The scientific endeavour to
understand the determinants of human action and behaviour is one that underlies
a variety of disciplines: psychology, epidemiology, law, machine learning and
economics among them. The assumption of rationality in which human beings do
what they believe is best for them is the fundamental behavioural foundation
for many economic theories and artificial intelligence algorithms, but can be
traced back to Ancient Greece. Socrates construed human beings as rational
agents, capable of understanding the full utility of their decisions, thus acting
upon the instinct of their own benefit. It is clear, however, that humans do
not always act on what appears the most beneficial course of action, a
phenomenon contradictory to Socrates’s claim that the Ancient Greeks called akrasia (Henry, 2002). According to
Socrates, the way to reconcile his perspective with the dilemma of akrasia was to presume that those who knowingly
acted in a way that would disadvantage them must be subject to a lack of full
knowledge on the matter (Hamilton, Cairns & Cooper, 1961). Despite his
philosophical zeal, Socrates’s viewpoint neglects to consider any additional
principles that may govern human behaviour.

The conviction that human behaviour is guided by more
than one underlying process has resulted in the emergence of a wide range of dual-process theories,
the essence of which remains two distinct processing modes: the reflective and
the impulsive (Strack & Deutsch, 2004; Gardener, 2014). Despite growing
literature supporting the significance of habit (represented by the impulsive
pathway) harkening back to the late nineteenth century (James, 1891), initial behavioural
theories and interventions have primarily focused on reasoned-based processes. The
assumption behind interventions based solely on the intentional, reflective
pathway is that one need only manipulate someone’s conscious cognitions to
achieve a certain behavioural outcome (Strack & Deutsch, 2004). For example,
in health psychology, models intended to alter eating behaviour emphasise
educational strategies, thus addressing deliberative processes almost
exclusively, though such interventions have been found to be largely
ineffective (Chadwick, 2017). These findings correspond with Socrates’s
assertion that addressing one’s lack of appropriate knowledge will give rise to
its complementary behaviour change. The lack of success in longitudinal designs
of such behavioural interventions and predictions suggests that while intention
may play a highly influential role in the initiation of new behaviours,
automatic processes like habit may provide further insight into the support of
long-term behaviour change (Sniehotta, Presseau & Araújo-Soares, 2014;
Rothman, Sheeran & Wood, 2009). Thereafter, through the examination of reviews
and meta-analyses, the aim of this paper is to evaluate the study of habit and
its contributions to our understanding of the maintenance and upkeep of
behavioural patterns over time.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

Prior to the contributions of habit,
it is first necessary to outline the meaning of ‘habit’ beyond its value in
common parlance.  Within
health psychology, habitual behaviour has been investigated in contexts ranging
from substance abuse and addiction (West & Brown, 2013) to dietary
consumption (Rothman et al., 2009) and beyond. A literature review of eight
definitions conducted by Gardner (2015) conceptualised the essential elements
of the habitual process: “cue dependence, automacity, and conditioned
stimulus-response associations” (pp. 4), finding that habit was commonly viewed
as a process in which specific environmental cues automatically activate
behavioural impulses through context-action associations, without deliberative,
purposeful thinking (Verplanken & Aarts, 1999). In the context of behaviour, automacity
further includes a lack of control or absence of awareness, and a sense of
mental efficiency (Gardner, 2015). Gardner (2012) further argued that while it
may be true that repetition reinforces context-action associations, frequency
is not an essential component of habit, as this cue-response mechanism in
habitual behaviour may persist even if the associated context is seldom
encountered. Habit can play a role in the initiation or performance of a
behaviour, a significant distinction that may yield higher precision in
understanding complex behaviours with composite sub-actions; habitually
initiated behaviour may necessitate further deliberative cognition in
sequential actions and conversely, habitually performed behaviour may be
consciously activated (Gardner, 2012).

Prior to discussing the findings and
implications of the study of habit, the empirical evidence warranting its study
and application should be explored.  A
systematic literature review on the influence of habit on dietary and physical
activity behaviours suggests that previous behavioural models have
overemphasised deliberative reason-based processes (Gardner, Bruijn &
Lally, 2011). To some extent, the high concentration in the study and use of intentionality
is not surprising, considering the magnitude of its revealed effect size,
explaining an averaged 28% of variance in behaviour (Sheeran, 2002, cited in
Chadwick, 2017). The dominant, widespread theory
of planned behaviour (TPB; Ajzen, 1985) pushes intentional, volitional
processes to the forefront, holding that intention is the composed aggregate of
an individual’s beliefs towards particular behaviours, perceived social
pressure and perceived capability in impacting behavioural performance. Although
studies assessing the TPB’s assumptions have been few and far between, the ones
that have done so do not appear promising (Sniehotta et al., 2014). In addition
to diminished predictive utility in longitudinal designs (Sniehotta et al., 2014),
its crown jewel of intentionality appears considerably less predictive of
repeated actions, which compose 43% of behaviour, occurring daily and largely
in the same contexts (as cited in Chadwick, 2017). At this point it should be
noted that despite this paper’s support of the study of habit, it is by no
means advocating for exclusive study of it, instead aiming to outline the
limitations of entirely focusing on intention and to discuss both the
relationship between the two pathways and the implications of it.

            In the endeavour to understand the
governance of human behaviour, Triandis (1977) postulated a relationship
between three determinants: intentions, habit strength, and the presence of
conditions facilitating behavioural performance or lack thereof. Habit strength
was hypothesised to be a moderating factor in the intention-behaviour
relationship. That is to say, intention predicts behaviour when habit strength
is weak, and to a much lesser extent when habit is strong. In a study assessing
behaviour regarding travel mode, habit was measured through self-report on past
travel mode frequency and through a more covert index of habit strength, developed
through analysing the amount of participant nominations for particular methods
of transport across a specified set of destinations (Verplanken, Aarts,
Knippenberg & Moonen, 1998). Intention was found to have significant predictive
utility only when habit was weak, whereas behaviour was found to be independent
of intention in the presence of strong habit, largely supporting Triandis’s
(1977) hypothesis. In a corresponding study, Aarts, Verplanken & Knippenberg
(1997) found that strong habit reduced the complexity of information utilised
in judgement of travel modes, supporting Gardner’s (2015) notion of habit’s
mental efficiency in its automacity. Though it seems habits, once activated by
their associated cues, often supersede intentions when they are in conflict
(Gardner et al., 2011), the distinction of habit as a process stimulating
cue-dependent impulses rather than behaviour suggests such tendencies may be
inhibited prior to action, given adequate use of intention and self-regulation
(Gardner, 2015). These findings hold promising implications for the
study and application of habit on behaviour change interventions.  

The primary methods of utilising
habit theory in behaviour change thus far largely consist of establishing habits
for target behaviours, manipulating the contexts and cues that activate
behavioural impulses, and reinforcing deliberative, reflective processes (Chadwick,
2017). The fundamental mechanisms in developing new habits comprise of behavioural
repetition in the consistent presence of specific cues and contexts to
condition and encourage automacity (Lally, Jaarsveld, Potts & Wardle, 2010).

The formation of a habit may also be dependent on an individual’s perceptions
of the desirability of an outcome (e.g. taking the car versus the train when
there are often scheduled delays) and may be reinforced by convenience and efficiency
(Verplanken & Wood, 2006). Through repetition in consistent contexts, the
development of cue-associations may result in habit formation and thus,
achievement of the behavioural target (Lally, Wardle & Gardner, 2011). The
second use of habit theory in behaviour change is through the manipulation of
those very environment-response associations. Intervention efficiency may be
enhanced when implemented alongside disruption in context or stimuli upon which
behavioural impulses may be dependent (Verplanken, Walker, Davis & Jurasek,
2008). Additional cue manipulation may come in the form of the cautious
monitoring and inhibition of any habitual impulses activated by their
respective cues (Quinn, Pascoe, Wood & Neal, 2009). Though there is evidence
to support both context-manipulation approaches, it is possible that despite a cessation
in contact with particular contexts, cue-response associations, albeit dormant,
may persist and even re-emerge upon encountering cues once more (Gardner, 2012).

Another method of behaviour change consists of the reinforcement of reflective
processes through the use of ‘implementation intentions’, in which a goal
intention is formed alongside specific, concrete plans determining the appropriate,
motivated responses to specific cues (Verplanken & Faes, 1999; Sheeran,
2002). These proposed methods within the study of habit may aid in the
enhancement of efficiency in interventions, as they prioritize the inclusion of
automatic processes, acknowledging that the reflective system is not the sole
antecedent of behaviour.


            In light of these findings, habit
has re-emerged in the spotlight of psychological research on behaviour, however
theoretical and methodological limitations remain. The beginnings of
behaviourist endeavour focused on behavioural minutiae like lever-pushing and
string-pulling, an emphasis that persists in habit’s study today, but may fail
in its application to more complex behaviours.  Indeed, Gardner (2015) questioned the translation
of habit’s study of simple actions to the elaborate, complex nature of human
behaviour, writing:

Behaviourists have traditionally
viewed complex habitual actions as concatenated sequences of simpler habitual
actions… however the notion that everyday health behaviours may be wholly
rigidly automated and performed with little control, awareness or intention
does not match the subjective experience of most health behaviours (pp. 5).

This corresponds to another criticism of the study of habit in the failure to
distinguish between habitually initiation and habitual performance of a
behaviour in dominant measures of habit through the Self-Report Behavioural
Automacity Index (SRBAI) and the Self-Report Habit Index (SRHI). Not only is
the accuracy of self-report questionable, there is also a failure to
distinguish habit from other forms of automacity (Gardner, 2015). Despite these
slight criticisms, it is worth noting that the study of habit is well in its
infancy, as only 21 studies on behavioural intervention have incorporated habit,
thirteen merely consider habit a potential covariate and only eight
interventions explicitly address habit change (Gardner, 2015; Chadwick, 2017).