Habit predicts in-the-moment alcohol consumption

Habit Predicts In-The-Moment Alcohol Consumption

1


Short Communication

Word count: 1,862 words (excluding References and Tables)

Ian P. Albery

Isabelle Collins

Antony C. Moss

Daniel Frings

and

Marcantonio M. Spada

London South Bank University, UK

Revision 1

Author Notes


2

Ian P. Albery, Isabelle Collins, Antony C. Moss, Daniel

Frings and Marcantonio M. Spada

Department of Psychology, London South Bank University, UK.

Correspondence concerning this article should be addressed to

Marcantonio M. Spada, Department of Psychology, London South

Bank University, United Kingdom. Contact: +44 (0)20 7815 5760

or e-mail [email protected].

Abstract

The objective of this study was to examine whether habit

predicts in-the-moment behavioural intention (amount of

alcohol poured) and behavioural enactment (amount and

proportion of alcohol consumed) controlling for craving and

positive alcohol expectancies. Forty-six college students,

who defined themselves as social drinkers, were tested

individually in a laboratory setting. After completing a

measure of craving they were given a bottle of non-alcoholic

beer and a cup, asked to pour a drink, and then drink as much

as they liked. They were not informed that the beer was non-

mailto:[email protected]


3

alcoholic. They were subsequently asked to complete measures

of alcohol use and misuse, positive alcohol expectancies and

habit. Results showed that positive alcohol expectancies were

positively and significantly associated with amount of

alcohol poured and amount and proportion of alcohol consumed.

Habit was positively and significantly associated with amount

and proportion of alcohol consumed but not with the amount of

alcohol poured. Hierarchical regression analyses revealed

that only habit was a significant predictor of both amount

and proportion of alcohol consumed. Even though measures of

intention (amount of alcohol poured) and behaviour (amount

and proportion of alcohol consumed) were positively

correlated, habit was shown to effectively discriminate

between these measures. These findings suggest that habit

predicts in-the-moment behavioural enactment in terms of

amount and proportion of alcohol consumed.


4

Key words: dual process model; habit; in-the-moment alcohol

consumption; metacognitive monitoring; positive alcohol

expectancies; proportion of alcohol consumed.

1. Introduction

Changing unhelpful drinking patterns relies on

identifying and modifying determinants of action.

Traditionally, approaches to behaviour change have been based

on reasoned action models, which portray behaviour as the

outcome of conscious intention in the form of, for example,

attitudes, beliefs and self-efficacy (Armitage & Conner,

2001). Explaining repeated actions, however, requires the

consideration of determinants of behaviour beyond consciously

experienced intentions especially in view of the fact that

findings appear to indicate that intention has a small-sized

effect in predicting ongoing behaviour (Webb & Sheeran,

2006).

Dual-process models (e.g. Moss & Albery, 2009; Strack &

Deutsch, 2004; Wiers et al, 2010) may offer a valuable

framework for tackling the difficulties of the ‘intention-


5

behaviour’ gap (see Sheeran, Gollwitzer & Bargh, 2013). These

models propose that behaviour results from both a

‘reflective’ pathway, which involves effortful forethought,

and an‘impulsive’ pathway, characterized by immediate

stimulus-response relationships mediated by the activation of

associative knowledge. From this standpoint, repetition can

initially lead reasoned actions to become impulsive through

the formation of ‘habits’, which are automatic behavioural

responses to contextual cues, acquired through context-

dependent repetition (see Ouellette and Wood, 1998;

Verplanken & Aarts, 1999). Repeated performance, in stable

settings, strengthens habits (Lally, van Jaarsveld, Potts, &

Wardle, 2010; Lally, Wardle & Gardner, 2011; Neal, Quinn

&Wood, 2006).

It has been argued that habit reflects an automated mode

of response which is reflected in core features such as lack

of awareness, control and conscious intent, and mental

efficiency (e.g., Wood & Neal, 2007). The Self-Report Habit

Index (SRHI) was developed as a unidimensional measure of


6

habit strength along these lines, by measuring the degree to

which a target behavior occurs frequently, requires conscious

awareness, thought and effort, is difficult to control, and

is relevant in terms of personal identity (Verplanken and

Orbell (2003). Among others, choices made around food (e.g.,

De Bruijn, 2010; De Bruijn et al, 2007), physical exercise

(Rhodes, De Bruijn, & Matheson, 2010), and self-reported

alcohol consumption (Gardner, De Bruijn & Lally, 2012) have

been shown to have a habitual component. A recent meta-

analysis (in nutritional/physical exercise) also showed a

medium to strong relationship between habit (i.e. SRHI) and

behaviour. The SRHI has also been found to moderate the

effect of intended behaviour on behavioural enactment (see

Gardner, De Bruijn & Lally, 2011).

A weakness of the evidence presented is that it is

predominantly limited to retrospective reports using

correlational designs. For example, Gardner and colleagues

(2012) used self-reported drinking in the past week whilst

prospectively predicting binge drinking from initial habit.


7

The current study attempts to address this limitation by

measuring in-the-moment alcohol consumption, planning and

enactment and investigating whether habit predicts the

proportion of alcohol consumed (an objective behavioural

outcome measure) controlling for craving and positive alcohol

expectancies (Christiansen et al. 1989). One way of measuring

in-the-moment drinking is to use a Taste Preference Task

(TPT) (Morrison, Noel & Ogle, 2012). In a TPT participants

are given a number of alcohol placebos and soft drinks to

taste and rate on a number of dimensions (e.g., quality,

taste, colour, etc). Consumption is calculated as the amount

drunk from a predefined known quantity. In this study we

asked participants to pour their own measure (from a known

quantity) to consume. This represents a measure of both

behavioural intention (amount of alcohol poured) and

behavioural enactment (amount and proportion of alcohol

consumed).

2. Method


8

2.1. Participants

The sample consisted of 46 college students who reported

being social drinkers (38 female and 8 male, mean age = 24.7

years, sd = 7.9, range = 18-53; mean AUDIT-C = 5.5, sd = 2.3,

range = 1-12).

2.2. Materials and Procedure

Ethics approval for the study was obtained from a UK

University. Once briefed and consent achieved participants

completed the Penn Alcohol Craving Scale (PACS; Flannery,

Volpicelli & Pettinati, 1999) which measures duration,

frequency and intensity of craving. They were then given a

bottle of non-alcoholic beer and a cup, and asked to pour

themselves a drink (behavioural intention) and drink as much

as they liked (behavioural enactment). Participants were not

informed that the beer was non-alcoholic. They were then

asked to complete the Alcohol Use Disorder Identification

Test Consumption (AUDIT-C; Bush, Kivlahan, McDonell, Fihn, &

Bradley, 1998) which comprises three items assessing quantity

and frequency of alcohol use, the Alcohol Outcome


9

Expectancies Scale (AOES; Leigh & Stacy, 1993) measuring

social facilitation, fun, sex, and tension reduction and the

SRHI (Verplanken & Orbell, 2003) comprising 12 items

assessing perceived habitual or automatic dimensions of

behaviour. Following completion of the study all participants

were debriefed.

3. Results

An inspection of histograms, skewness and kurtosis

showed that several variables were not normally distributed.

One-tailed Spearman rho correlation analyses showed that

positive alcohol expectancies were positively and

significantly associated with amount of alcohol poured,

amount of alcohol consumed and proportion of alcohol

consumed. Habit was found to be positively and significantly

associated with amount and proportion of alcohol consumed and

not amount of alcohol poured. No significant associations

between age, self-reported alcohol use, and craving, on the

one hand, and amount of alcohol poured, amount of alcohol

consumed and proportion of alcohol consumed were observed.


10

Amount of alcohol poured, amount of alcohol consumed and

proportion of alcohol consumed were all positively and

significantly correlated (see Table 1).

To evaluate the contribution of habit beyond that

accounted for by positive alcohol expectancies, hierarchical

regression analyses were run with amount of alcohol poured,

amount of alcohol consumed and proportion of alcohol consumed

as criterion variables. Positive alcohol expectancies were

entered in step 1 and habit was entered in step 2 (see Table

2). Together habit and positive alcohol expectancies were

shown to significantly predict amount of alcohol consumed

[F(2, 43) = 5.38, p < .001 (R2 = .20, 2= .16)] and proportion

of alcohol consumed [F(2, 43) = 5.92, p < .01 (R2 = .22,

2= .18)] but not amount of alcohol poured [F (2, 43) = 2.23, p

= .12 (R2 = .09, 2 = .05)]. An inspection of the final

equations revealed that habit significantly increased the


11

variance explained in amount of alcohol consumed [F∆ (1, 43)

= 4.58, p < 0.05 (R2 = .16, R2∆= .09, B = .31,)] and

proportion of alcohol consumed [F∆ (1, 43) = 7.39, p < 0.01

(R2 = .22, R2∆= .14, B = .39)] over and above the variance

accounted for by positive alcohol expectancies (see Table 2).

4. Discussion

This study sought to examine whether habit predicts in-

the-moment amount of alcohol poured, and amount and

proportion of alcohol consumed controlling for craving and

positive alcohol expectancies. A focus on how much an

individual pours to drink, and the actual amount they drink,

allows us to distinguish between one’s intention and the

enactment of this behavioural intention. Importantly,

however, these measures are both behavioural in nature and as

such provide evidence as indirect measures. Correlational

analyses showed dissociation between positive alcohol

expectancies and habit, with amount of alcohol poured

(behavioural intention) and amount and proportion of alcohol

consumed (behavioural enactment). Positive alcohol


12

expectancies were positively and significantly associated

with amount of alcohol poured, and amount and proportion of

alcohol consumed. Habit was positively and significantly

associated with amount of alcohol consumed and proportion of

alcohol consumed. On this basis positive alcohol expectancies

appear to be associated with the planning of behaviour. In

contrast, habit appears to be associated with behavioural

enactment. These indications were confirmed by hierarchical

regression analyses which revealed that: (i) positive alcohol

expectancies and habit did not predict amount of alcohol

poured; (ii) habit added significant variance to the

prediction of amount and proportion of alcohol consumed

beyond expectancies; and (iii) only habit was a significant

predictor of both amount and proportion of alcohol consumed.

Although measures of intention (alcohol poured) and

behavioural enactment (amount and proportion of alcohol

consumed) were positively and significantly correlated, habit

was shown to effectively discriminate between these measures.


13

This highlights the role of habit in guiding behavioural

enactment and not behavioural intention.

These results have potential implications for behaviour

change practice, highlighting the importance of disrupting

the cue–response association underpinning habit (Verplanken &

Wood, 2006). The purposive discontinuation of exposure to the

everyday cues that support habit may be an unrealistic

intervention strategy, however, using volitional strategies

such as vigilant monitoring (Quinn, Pascoe, Wood & Neal,

2006) or the enhancement of metacognitive monitoring (Spada &

Wells, 2006; Spada, Caselli & Wells, 2013), respectively

aimed at heightening attention to behaviour so as to detect

habit initiation or stop signals for engaging in a behaviour,

may be helpful in inhibiting the performance of unhelpful

habits. An example of a metacognitive monitoring enhancement

strategy would be the use of Situational Attentional

Refocusing (SAR; Wells 2000), which aims to increase the flow

of adaptive information in awareness so the individual is

better able to regulate cognition and behavior. This


14

technique would require the purposeful direction of attention

onto cues related to the use of alcohol, such as quantity of

alcohol consumed and proximity to desired goals, with the

objective of enhancing self-awareness during enactment and

help identify a stop signal for use.

This study has several limitations. Firstly, it partially

relies on self-report instruments, which are subject to errors

in measurement. Secondly, a cross-sectional design was adopted

which precludes causal inferences. Thirdly, the sample

comprised predominantly female college students so

generalisations based on the current findings should be

considered with caution.

Directions for future research include ascertaining

further the role of habit in predicting drinking behaviour,

particularly through in-the-moment dynamic longitudinal

studies and through the examination of more representative

samples (including clinical samples). It would also be

interesting to ascertain the efficacy of treatments, aimed at


15

problem drinking, that focus on heightening attention to

behaviour and stop signals for engaging in behavior.

References

Armitage, C. J. & Conner, M. (2001). Efficacy of the Theory

of Planned Behaviour: A

meta-analytic review. British Journal of Social Psychology, 40,

471-99.

Bush, K., Kivlahan, D. R., McDonell, M. B., Fihn, S. D. &

Bradley, K. A. (1998). For the

Ambulatory Care Quality Improvement Project (ACQUIP).

The AUDIT alcohol consumption questions (AUDIT-C): An

effective brief screening test for problem drinking.

Archives of Internal Medicine, 158, 1789-1795.

Christiansen, B. A., Smith, G. T., Roehling, P. V. & Goldman,

M. S. (1989). Using alcohol


16

expectancies to predict adolescent drinking behaviour

after one year. Journal of Consulting and Clinical Psychology, 57,

93-99.

De Bruijn, G.J. (2010). Understanding college students’ fruit

consumption: Integrating

habit strength in the theory of planned behaviour.

Appetite, 54, 16–22.

De Bruijn, G. J., Kremers, S. P., De Vet, E., De Nooijer, J.,

Van Mechelen, W., & Brug, J. (2007). Does habit strength

moderate the intention–behaviour relationship in the

Theory of Planned Behaviour? The case of fruit

consumption. Psychology and Health, 22, 899-916.

Flannery, B. A., Volpicelli, J. R. & Pettinati, H. M. (1999).

Psychometric properties of the Penn Alcohol Craving

Scale. Alcoholism: Clinical and Experimental Research, 23, 1289-1295.

Gardner, B., de Bruijn, G. J., & Lally, P. (2011). A

systematic review and meta-analysis of applications of

the self-report habit index to nutrition and physical


17

activity behaviours. Annals of Behavioral Medicine, 42, 174-

187.

Gardner, B., De Bruijn, G. J., & Lally, P. (2012). Habit,

identity, and repetitive action: A prospective study of

binge‐drinking in UK students. British Journal of Health

Psychology, 17, 565-581.

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.

Lally, P., Wardle, J. & Gardner, B. (2011). Experiences of

habit formation: A qualitative

study. Psychology, Health & Medicine, 16, 484-489.

Leigh, B. C. & Stacy, A.W. (1993). Alcohol outcome

expectancies: Scale construction and

predictive utility in higher order confirmatory models.

Psychological Assessment, 5, 216-229.

Morrison, P.M., Noel, N.E. & Ogle, R.E. (2012) Do angry women

choose alcohol? Addictive Behaviors, 37, 908-13.


18

Moss, A. C. & Albery, I. P. (2009). A dual-process model of

the alcohol-behavior link for

social drinking. Psychological Bulletin, 135, 516-530.

Neal, D. T., Wood, W. & Quinn, J. M. (2006). Habits: A repeat

performance. Current

Directions in Psychological Science, 15, 198–202.

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.

Quinn, J. M., Pascoe, A., Wood, W. & Neal, D. T. (2010).

Can't control yourself? Monitor

those bad habits. Personality and Social Psychology Bulletin, 36,

499-511.

Rhodes, R., De Bruijn, G.-J. & Matheson, D. H. (2010). Habit

in the physical activity


19

domain: Integration with intention temporal stability

and action control. Journal of Sport and Exercise Psychology, 32,

84-98.

Sheeran, P., Gollwitzer, P. M., & Bargh, J. A. (2013).

Nonconscious processes and health. Health Psychology, 32,

460.

Spada, M. M., Caselli, G. & Wells, A. (2013). A triphasic

metacognitive formulation of

problem drinking. Clinical Psychology & Psychotherapy, 20, 494-

500.

Spada, M. M. & Wells, A. (2006). Metacognitions about alcohol

use in problem drinkers.

Clinical Psychology & Psychotherapy, 13, 138-143.

Strack, F. & Deutsch, R. (2004). Reflective and impulsive

determinants of social behaviour.

Personality and Social Psychology Review, 8, 220-247.

Verplanken, B. & Aarts, H. (1999). Habit, attitude, and

planned behavior: Is habit an empty


20

construct or an interesting case of goal-directed

automaticity? European Review of Social Psychology, 10, 101–134.

Verplanken, B. & Orbell, S. (2003). Reflections on past

behavior: A self-report index of

habit strength. Journal of Applied Social Psychology, 33, 1313-

1330.

Verplanken, B. & Wood, W. (2006). Interventions to break and

create consumer habits.

Journal of Public Policy and Marketing, 25, 90-103.

Webb, T. L & Sheeran, P. (2006). Does changing behavioral

intentions engender behavior

change? An analysis of the experimental evidence.

Psychological Bullettin, 132, 249-68.

Wells, A. (2000). Emotional Disorders & Metacognition: Innovative

Cognitive Therapy.

Chichester, UK: Wiley.

Wiers, R. W., Houben, K., Roefs, A., de Jong, P. & Stacy, A.

W. (2010). Implicit cognition in health psychology: Why


21

common sense goes out the window. In B. Gawronski & B.

Keith Payne (Eds.), Handbook of Implicit Social Cognition:

Measurement, Theory, and Applications, (pp. 463-488). New York,

London: The Guilford Press.

Wood, W. & Neal, D. T. (2007). A new look at habits and the

habit-goal interface. Psychological Review, 114, 843.

Table 1: Descriptive statistics and Spearman’s rho correlations coefficients (one

tailed).

Mean SD Range AUDIT-C PACS AOEQ

Positiv

e

SRHI Amount

of

alcohol

poured

(ml)

Amount

of

alcohol

consume

d (ml)

Proportion

of alcohol

consumed

(%)

Age 24.7 7.9 18-53 -.04 -.11 -.01 .17 .01 .09 .07

AUDIT-C 5.5 2.3 1-12 - .36** .32** .41** .06 .05 .09

PACS 7.2 4.3 0-23 - .27* .35** .21 .02 -.10

AOEQ Positive 81.0 9.1 64-100 - .36** .25* .27* .26*

SHRI 30.4 7.5 13-49 - .19 .32* .45**

Amount of alcohol

poured (ml)

52.2 36.

8

7-175 - .73** .28*

Amount of alcohol 34.9 37. 2-146 - .74**

consumed (ml) 9

Proportion of

alcohol consumed

(%)

60.9 30.

4

7.4-

98.7

-

* p < .05, ** p < .001

Table 2: Hierarchical multiple linear regression statistics with amount of alcohol poured

and consumed and proportion of alcohol consumed as criterion variables and positive

alcohol expectancies and habit as predictor variables.

Amount of alcohol

poured

(ml)

Amount of alcohol

consumed (ml)

Proportion of alcohol

consumed (%)

Step 1 β t p β t p β t p

AOEQ Positive .25 1.67 .10 0.34 2.39 0.02 0.29 1.97 0.05

r2 = 0.06 r2 = 0.12 r2 = 0.08

Step 2

AOEQ Positive .18 1.19 .24 0.24 1.63 0.10 0.16 1.01 0.28

SRHI .19 1.26 .22 0.31 2.14 0.04 0.39 2.71 0.01

r2 = 0.09 r2 = 0.20 r2 = 0.22

Habit predicts in-the-moment alcohol consumption

Documents

Transcript of Habit predicts in-the-moment alcohol consumption