Habits

1
Habits

• And time for your questions

2
Evidence for intentions

• Intentions account for 28 of variance in
  behaviour on average (Sheeran, 2002)
• When past behaviour is controlled for, a
  meta-analysis showed that past behaviour accounts
  for about 26 of the variance in behaviour and
  intentions for an additional 7 over and above
  past behaviour (Sutton Sheeran 2003)

3
Background Habits

• Much of our behaviour is frequently exhibited in
  the same environment and has habitual character
• Habits enable us to act accurately without
  spending much attention in an automatic fashion.
• For example, driving, typing and brushing ones
  teeth would take more effort if is was not
  habitual.

4
Early study on the role of habitsBentler and
Speckart (1979)

• Study on students' consumption of alcohol and
• marijuana.
• Findings measure of habit (obtained by
  self-reported frequency of behaviour in the
  previous 2 weeks) predicted future behaviour
  directly without mediation of attitudes or
  intentions.
• Conclusion actions become habitual over time and
  can be instigated without "products of
  reasoning, i.e. without deliberation and thought

5
Ouellette, J., Wood, W. (1998). Habit and
intention in everyday life Themultiple
processes by which past behavior predicts future
behavior. Psychological Bulletin, 124, 54-74.

• Meta-analysis of studies on habits
• showed that the direct influence of past
  behaviour on future behaviour was strongest for
  behaviours executed frequently and consistently
  in a stable context.
• Behaviours carried out less often were more
  accurately predicted by consciously formed
  intentions.
• This confirms the assumption that when a
  behaviour has been performed many times in the
  past, future behaviour becomes increasingly under
  control of an automatised process, whereas a
  behaviour executed less frequently is (still)
  guided by evaluative interpretations and
  considerations.

6
Habits as goal-directedautomatic behaviour

• Habits are represented as links between a goal
  and actions that are instrumental in attaining
  this goal.
• The strength of such links is dependent on
  frequent co-activation of the goal and the
  relevant actions in the past.
• Goals do not need to be conscious.

Aarts Dijkstrahuis, 2000
7
Experiment 1 Hypotheses

• Tested the assumption that action is activated
  automatically on the instigation of a goal if and
  only if the behaviour is habitual.
• A secondary aim was to show that goals can exert
  their influence on habitual responses when people
  are not aware of the relation between an
  earlier-primed goal and the habitual response.

Aarts Dijkstrahuis, 2000
8
Experiment 1 Methods

• 54 students owning a bicycle living near the
  university
• 2 (habitual vs. non habitual bikers) 2 (goal
  priming vs. no goal priming) design
• Pilot study on locations for cycling and goals
  associated with going to these places
• Three tasks First task served as the
  manipulation phase for goal priming, and the
  second task was designed to study the effects of
  habit and goal priming on the speed of responding
  to the target bicycle trips. Habit strength was
  assessed in the third task.

9
Experiment 1 Methods

• Task 1 Experimental group read 5 sentences
  describing the five different travel goals (e.g.,
  going shopping at the city centre mall) that
  corresponded to the five travel locations
  obtained in the pilot study without providing the
  travel locations per se
• Task 2 Association task - 40 different location
  words appeared briefly on the screen followed by
  a mode of transport (300ms).
• Task 3 Habit strength assessment Participants'
  estimates of frequency of bicycle use in the
  recent past for different trips were used to
  obtain information on their bicycle habit
  strength.

10
The results of Experiment 1 supported the
predictions. Habitual bicycle users who were
primed with travel goals showed faster responses
than nonhabitual bicycle users. Furthermore, this
effect did not appear in the absence of goal
priming.
11
Experiment 2

• Similarities between habits and implementation
  intentions (IMPS)
• Hypothesis that IMPS can lead to habit-like
  effects but do not facilitate action in habitual
  actors
• To test these ideas, the authors examined the
  interaction between bicycle habit strength and
  the formation of implementation intentions (by
  means of planning travel goals) using the same
  association task as in Experiment 1.
• Experimental group planned travel goals, controls
  planned repairing a flat tire.

12
These latter results show that planning (or
formation of implementation intentions)
facilitated the speed of nonhabitual
participants' responses, whereas this was not the
case for habitual participants.
13
Questions about the exam
14
General points

• Questions will be similar to last years and to
  the test questions for each lecture
• The blocks will not be ordered according to the
  lectures (e.g. 1-6 and 7-12).
• Reading the papers I suggested for preparation of
  the lectures is a good idea.

15
Question

• Are stage models classified as social cognition
  models?
• Answer NO, both are distinct classes of models
  with different structure and aims.

16
Could you please go over the tests of stage
assumptions regarding theTTM/PAPM -- in other
words, the cross-sectional/longitudinal/experiment
al evidence? 
17
Tests of stage assumptions

• Cross-sectional
• Longitudinal
• Experimental

18
Sutton 2000
19
Linear Patterns of Self-Efficacy over TTM Stages
of Change Armitage Arden, 2002
20
Discontinuity Patterns in an extended Version of
the PAPMSniehotta et al., 2005
Outcome expectancies (cons)
Vulnerability
21
Tests of stage assumptions

• Cross-sectional
• Longitudinal
• Experimental

22
Predictors of non-smokingchange their pattern
across stages
Outcome Expectancy
Self-Efficacy
De Vries, H. Mudde, A. N. (1998). Predicting
stage transitions for smoking cessation applying
the attitude-social influence-efficacy model.
Psychology and Health, 13, 369-385.
23
Tests of stage assumptions

• Cross-sectional
• Longitudinal
• Experimental

24
General Design to Test the Matching of Treatments
to Stages experimentally
Treatment 1 Treatment 2
matched
Move from stage A to stage B
Move from stage B to stage C
matched
Weinstein, N. D., Lyon, J. E., Sandman, P. M.
Cuite, C. L. (1998). Experimental evidence for
stages of health behavior change The precaution
adoption process model applied to home radon
testing. Health Psychology, 17(5), 445-453.
25
Progress Toward Acting of one Stage or More (in )
Weinstein, N. D., Lyon, J. E., Sandman, P. M.
Cuite, C. L. (1998). Experimental evidence for
stages of health behavior change The precaution
adoption process model applied to home radon
testing. Health Psychology, 17(5), 445-453.
26
Test Orders (in )
Weinstein, N. D., Lyon, J. E., Sandman, P. M.
Cuite, C. L. (1998). Experimental evidence for
stages of health behavior change The precaution
adoption process model applied to home radon
testing. Health Psychology, 17(5), 445-453.
27
Lecture 4 Stage models

• There is a graph by Dijkstra, De Vries
  Roijackers (1998), I wonderedwhat it was
  showing and how it fits into the stage models
  part.  Is it anything to do with interventions?

28
Percent of ex-smokers 10 weeks post-treatment
Dijkstra, A., De Vries, H., Roijackers, J.
(1998). Computerized tailored feedback to change
cognitive determinants of smoking A Dutch field
experiment. In Health Education Research, 13(2),
197-206.
29
Question

• Could you clarify the difference between
  variables which mediate andthose which moderate?

30
Mediator-Variables Definition

• MEDIATOR
• A variable is a mediator of the relationship
  between two other variables, if it is related to
  both other variables and accounts for a share or
  all of the shared variance between the other two
  variables
• Mediators explain why effects occur.
• Relationships between variables can be broken
  down into direct and indirect effects

31
Mediator
Mediator (e.g., Self-efficacy)
outcome (e.g., behaviour)
predictor (e.g., Intervention)
32
Moderator or Mediator?

• Baron, R.M. Kenny, D.A. (1986). The
  mediator-moderator variable distinction in social
  psychological research Conceptual, strategic,
  and statistical considerations. Journal of
  Personality and Social Psychology. 51, 1173
  1182.

33
Moderator-Variable Definition

• MODERATOR
• A Variable, that changes the direction or
  strength of the relationship between two other
  variables
• A moderator is a third variable that affects the
  zero-order correlation between two other
  variables
• Baron Kenny, 1986

34
Moderator
Intention
35
Moderator
Outcome (e.g., Intention)
Predictor (e.g., subjektive norm)
36
Moderator
37
(No Transcript)
38
(No Transcript)