Transcript Powerpoint Presentation

The Divided Self:
How a Better Understanding of the Human
Mind Could Transform Society
Professor Geoff Beattie
Department of Psychology,
Edge Hill University
iafor ‘European Conference on Psychology and the
Behavioral Sciences’
7th July 2015
We are currently faced with a
number of major challenges:
1.
Relationship with the planet
(climate change, sustainability, increasing demand on
finite resource).
2.
Relationship with each other
(prejudice, inequality and conflict).
I want to suggest that some new ideas in
psychology may help us reframe both sets of
challenges and thereby act more effectively.
Defining the issues, imagining
possible solutions

I think that there is genuine frustration that these
issues have proven to be so intractable.

World leaders, economists, social scientists, the
commercial sector have all offered solutions….that
failed. Why is this?

I’m going to suggest that it’s us that’s the problem
and how we think and act.

Firstly, climate change.
Climate change
 The

scientific evidence is
overwhelming.
Stern Review (2006): ‘Climate change
threatens the basic elements of life for
people around the world.’
IPCC report (2014):
A rise in global temperature will have:

severe and widespread impacts

major risks to global food production

more extreme fluctuations in weather,
including droughts, flooding and
storms.’
But there is something else:

Stern (2006):

‘Human activities are a major driver of this
rapid change in our climate…

particularly patterns of consumption and
energy use, driven by consumer demand for
higher standards of living.’
Some governments, NGO’s and
leading multinationals have got the
message
 Unilever

(turnover £40 billion; products sold in 190 countries; 2
billion times a day somebody uses one of their
products).
 KPIs:
primarily financial: 5% growth
But they see the problem
(‘there’s no profit from a dead planet’)

‘Business must be part of the solution.’

‘in order to live within the natural limits of the
planet we will have to decouple growth from
environmental impact’.

KPI: ‘Our Commitment’: ‘Halve the
greenhouse gas impact of our products
across the lifecycle by 2020.’
What Unilever did:
1.
Reduced greenhouse gas emissions from
manufacturing chain, reduced deforestation.
2.
More environmentally friendly sourcing of raw
materials.
3.
Doubled their use of renewable energy.
4.
Produced concentrated liquids and powders.
5.
Reduced greenhouse gas emissions from
transport and refrigeration.
6.
Reduced employee travel.
The end result:

‘Our greenhouse gas footprint impact
per consumer has ………..
 increased by around 5% since
2010.’
The problem

‘We have made good progress in
those areas under our control but…

the big challenges are those areas not
under our direct control
like…..consumer behaviour.’
Why don’t people get the
message?




The effects are less personal.
It’s primarily about future generations.
Easy to shift responsibility; it’s China’s fault.
Uncertainty about time course (powerful lobbies
behind this uncertainty, cf. the tobacco industry in the
past).
It’s complicated.
 Emotional valence of ‘global warming’.
 Too catastrophic to contemplate……….


Hard but not impossible/we just need to ‘read’
people better and design better initiatives.
So how do we promote
sustainable consumption/living?
 Step
1: we need to be able to
read the mind of the consumer.
 To
ascertain what’s possible and
what’s not.
We need:
1.
To understand consumers’ motives
and intentions.
2.
Assess how likely they are to change
their patterns of consumption.
3.
Decide whether they can break long
established habits and routines.

This mind reading might seem to be
very difficult,

but an extraordinary number of
influential people/organizations think it
is indeed possible.
I will argue…

It is more difficult than it might appear
because consumers don’t have a mind
 they
have two.
 They
have two distinct cognitive
systems each with its own
properties and mode of operation.
 One
of these systems is not open
to introspection.
But some great mind readers
have tried

‘Customers want to do more in the
fight against climate change if only we
can make it easier and more
affordable’ (Leahy, 2007).

Numerous market research surveys
supported his conclusion.
Surveys of pro-environmental
attitudes revealed:

‘70% of people agree that if there is no
change in the world, we will soon
experience a major environmental crisis’

‘78% of people say that they are prepared
to change their behaviour to help limit
climate change.’
Downing and Ballantyne (2007)
This view and philosophy shared by others,
including government agencies (Defra)

They also engaged in a spot of
mind-reading-

‘Many people are willing to do more to limit
their environmental impact, they have a much
lower level of understanding about what they
can do and what would make a difference.’
One major initiative by Tesco:
carbon labelling
Huge undertaking
1.
The start of Leahy’s ‘Green Revolution’, to be
led by consumer demand to drive the market.
2.
What consumers said they wanted – info.
3.
It had worked with health info on food.
4.
Tesco planned to label all 70,000 own brand
products.
5.
Several months to calculate the carbon
footprint of each individual product.
6.
Validated by the independent Carbon Trust.
7.
Consumers should now choose the low
carbon footprint alternatives.
8.
But how did consumers actually behave?
They didn’t do what Tesco thought
they would.
(Talk about frustration!).
And here’s something else they didn’t do.
Analyzing gaze fixations

Small black mark denotes where
participants are looking.

Each gaze fixation scored every 40
ms.

Calculate proportion of time spent
fixating on different areas.
Distribution of gaze fixations
(first 5 seconds)
Carbon
footprint
Carbon footprint
information
Other features

But was this mind-reading ever really
justified?

Both Terry Leahy and Defra based
their strategy on data on consumer
attitudes.

Understanding ‘attitudes’ is the key
The definition of an attitude

‘a mental and neural state of readiness
organised through experience, exerting a
directive or dynamic influence upon the
individual’s response to all objects and
situations with which it is related.’
(Allport, 1935)
How do we measure attitudes?
The mainstream approach
 Likert

Scale:
You simply introspect and report preferences (for
example, towards high or low carbon footprint products).

Following Allport, who wanted to escape
‘psychoanalytic excess’ (following his visit to Freud when
he was a student).
But Allport also said:

‘Often an attitude seemed to have no
representation in consciousness
other than a vague sense of need, or
some indefinite or unanalyzable feeling
of doubt, assent, conviction, effort, or
familiarity’ (Allport 1935).
The question is….

Can we build a science (and effective
policy) on these vague senses and
unanalyzable feelings of doubt, which
may be crucial for the prediction of
behaviour?
That therefore is the challenge…

One advance – recognition that there are two
systems of the human mind.
Nobel Laureate Daniel Kahneman:

System 1 (automatic, fast, impulsive, unintended,
unconscious).

System 2 (intentional, slow, controlled, reflective,
conscious).
System 1
(automatic and fast; little or no
effort):
1. Detect
emotion in the human
face
(from 6 basic emotions: happiness, sadness,
anger, fear, disgust, surprise).
System 1
2.
Orient to the source of a sudden
sound.
3.
Complete the phrase ‘bread and …’
4.
Answer 2 + 2 = ?
System 2 is very different.
System 2
What is:
17 x 24=
(My guess is that whatever you are now doing is conscious,
controlled, reflective, intentional and slow).
System 2
(conscious and controlled; requires attention)

Look for the a friend in the audience.

Tell someone your phone number.

Check out the validity of a complex
logical argument.
Relationship between System 1 and
Systems 2 is critical
1.
System 1 continuously generates suggestions for
System 2 (impressions, intuitions, feelings).
2.
If endorsed by System 2, these turn into voluntary
actions.
3.
When System 1 runs into difficulties it calls on
System 2 for more detailed processing.
4.
when System 1 has no answer, System 2 is
mobilised.
So far so good.
As quickly as possible, answer
the following question:
A bat and ball costs £1.10.
The bat costs £1 more than the
ball.
How much does the ball cost?
What does this tell us?

People tend not to check

(more than 80% of university students give the wrong
answer).

System 2 endorsed an intuitive answer that it
could have rejected with a small investment
of effort.

System 2 can be very lazy

(according to Kahneman this judgment is ‘harsh but
not unfair’).
How System 1 works:
Please look at the following two
words.
BANANAS
BANANAS
VOMIT
The Associative Machine
What has just happened to you:

You experienced unpleasant images/feelings.

Your mind assumed a causal connection
between the two words.

You will have experienced a temporary
aversion to bananas.

The state of your memory has changed.
Complete the following word:
‘s-ck’

You are more likely to recognise objects and
concepts associated with ‘vomit’ (this is
called ‘priming’).

And recognise words associated with
‘bananas’ (‘fruit’, ‘yellow’ etc.).
 You

want to survive.
Actions (not just thoughts) can be primed by
events of which you are not even aware.


Holland et al. (2005)
Participants given a crumbly
two conditions
Lemon Scented
Room
biscuit to eat in one of
Room with no
Scent
The nature of the human mind
1.
Comprised of two systems that work on
different principles.
2.
The interaction between the two is critical.
3.
One system is conscious (and we think
reflects our ‘true self’) but lazy.
4.
The other is unconscious and hidden from
us (and a bit of a workaholic).
But what about the long-term
legacy of this (especially System 1)?
1.
How do these associations work
across time?
2.
How do these shape our attitudes and
behaviour?
3.
Do we have unconscious ‘implicit’
attitudes (bases on this associative
activation over time)?
But how can we measure attitudes that elude
conscious introspection? (funding from Tesco)
By measuring associations directly in a
computerised classification task.
 For example, how quickly can you associate
‘low/high carbon footprint’ with the concept of
‘good’ or ‘bad’?
 Harder to associate certain categories rather
than others.

Low versus high carbon footprint.
Low carbon footprint
High carbon footprint
Good versus bad.
Good
Bad
Glorious
Good or High Carbon Footprints Vs.
Bad or Low Carbon Footprints.
Good
or
High carbon
footprint
Bad
or
Low carbon
footprint
Good or Low Carbon Footprints Vs.
Bad or High Carbon Footprints.
Good
or
Low carbon
footprint
Bad
or
High carbon
footprint
This test yields a ‘D’ score’

The difference in time to react to one set of
combinations rather than the other.
Measures of implicit attitudes and the
prediction of behaviour

Both predict behaviour in different domains, and in
different circumstances.

IAT is a better predictor of spontaneous behaviours
when behaviour is under cognitive, emotional or time
pressure (meta-analysis of over 100 studies).

IAT is a better predictor of behaviour in sensitive
domains (including racial discrimination and
environmental issues).
30% of participants
demonstrated a strong
preference for products
with low carbon
footprints.
•
40% of participants
demonstrated a moderate
preference for products
with low carbon
footprints.
•
26% of participants
demonstrated no
preference.
•
4% of participants
demonstrated a
preference for products
with high carbon
footprints.
Percentage
•
Percentage
Likert results in area
of sustainability (Beattie, 2010).
Likert Score
Implicit attitudes to high and low carbon
footprint products (‘D’ scores)
How do explicit and implicit attitudes
connect?
• No significant correlation in this domain.
• Explicitly stated attitudes and implicit associations
appear to be statistically dissociated.
• There were many ‘surface greens’ with a reported
positive attitude to low carbon but actually a
positive implicit attitude to high carbon.
The psychology of the ‘surface
greens’

Understanding these conflicted individuals may be
critical.

They may be very common.

They have not been identified as a group thus far.

After all they report that they prefer low carbon products,
are prepared to adapt their behaviour etc.
Can we identify ‘surface
greens’?
Study the detailed way in which people talk about
relevant issues.
But not just focussing on the language/speech itself
(cf. questionnaires, interviews, focus groups).
Rather by micro-analysis of co-verbal behaviour.
A focus on iconic gestures,
which are:

Spontaneous
 Multi-dimensional
 Meaningful (without the benefit of a lexicon)
 Unconscious (and therefore less open to
editorial control)
They:
 Often back-up speech
 But sometimes contradict it
 These are called gesture-speech mismatches.
‘True Green’
Explicit/implicit
attitudes: convergent
Pro-low carbon
explicit attitude; prolow carbon implicit
attitude
Speech and
gesture: matching
‘Surface Green’
Explicit and implicit
attitudes: divergent
Pro-low carbon explicit
attitude; pro-high carbon
implicit attitude
Speech and
gesture:
mismatching
Is it surprising?

We probably do care about the environment
(when we think consciously and carefully)
and yet for years we have been socialized to
think and feel……….
Implications?

Immediately raises issues about methods to
reveal attitudes.

Some methods (questionnaires, interviews,
focus groups) may provide misleading
results.

Suggests that many people are
psychologically ‘dissociated’ when it comes
to the environment/climate change
Implications for modelling the
human mind

How do these two systems influence
behaviour?

How do these two systems interact (if at
all)?

How do you change these two systems?

How do you appeal to them?
Some preliminary investigations

Implicit attitude to carbon footprint predicts choice of
low carbon products under time pressure.

Implicit attitude to carbon footprint predicts
unconscious eye fixations on climate change
images.
So what about attitudes towards
ethnicity and race?

Could there be some sort of dissociation between
explicit and implicit attitudes here?

Explicit attitudes to race changed dramatically in a
relatively short period (35% acceptable to have a Black
neighbour in 1942, 64% in 1963).
 But
what actually happened?
Is there any actual evidence of ethnic/racial
bias today?
(but avoiding asking people to merely report it)
Take the domain of employment.
First stage of any process: shortlisting
One technique - the (experimentally
manipulated) C.V. test: 2 versions of identical
CV: White/non-White.
Possible discrimination in
the U.K.

Wood et al., (2009): matched CVs with White and nonWhite sounding names.

Sent to a range of jobs (IT, Accounts, H.R.) in 7 major
British cities.

Results: BME had to send 74% more applications that
non-Whites to secure an interview.
Conclusions
1.
There seems to be some
sort of bias
2.
Does not identify type of bias
or processes involved
Could there be ethnic bias in
universities?

11% of all White academics are professors compared
to just 5% of BME academics in the U.K.
 50
Black professors out of a total of 14,385 in
Britain

0.35% of Professors are Black, but 2.8% of the
population.

Not a single Black Vice Chancellor in the United
Kingdom.
Note: all statistics correct at the start point of the research in
2009-10.
Journal of Blacks in Higher
Education (2009):
‘Blacks in faculty ranks (in the U.S.)
will not reach parity with the Black
percentage of the overall American
workforce for another 140 years.’
Attempts at solutions
1.
Operational Performance Reviews
with close monitoring of data on
ethnicity.
2.
Career development advice.
3.
Mentoring.
Still minimal change at senior
level.
Could there be an implicit bias related
to race or ethnicity operating in
society?
There does seem to be significant
evidence of an implicit racial bias
(Greenwald)
(see, for example, Project Implicit: Harvard University
website).
Where does this implicit bias
derive from?

Emotional response to out-group
versus own-group members (more activity
in amygdala).

The psychological pull of the familiar.

Significantly affected by early
socialization experiences.

Significantly affected by our culture as
a whole.
Implications of this implicit bias?
But the effects on relevant
behaviours are less clear.
This was the starting point for my
own research.
Measuring implicit attitudes to
race/ethnicity

There are limitations with many previous
versions of the IAT (including the Race IAT)
used in the literature:
Project Implicit: Race IAT Stimuli
First step: creation of new Ethnic IAT
Likert Scale

Simple self-report measure of underlying attitude
along a 5-point scale (1 to 5).
IAT: Critical trial
Good
or
White
Bad
or
Non-White
IAT: Critical trial
Good
or
Non-White
Bad
or
White
Distribution of Likert scores for White
participants (5: ‘prefer Whites’; 1: ‘prefer non-Whites’; 3: ‘no
preference).
Results: self-reported and implicit
attitude scores
Likert score
Mean
Attitudinal Preference
3.18
3.0 = Neutral
Medium preference for White
(N.B. 0.80 = Strong preference for
D score
0.73
White)
D score of White participants:
0.87
But how do implicit attitudes influence
actual behaviour (shortlisting)?


Two genuine job adverts.
One an academic position, the other an
administrative post.

Four C.V.s created for the lectureship post and four for
the admin post.

Each C.V. contained good and bad information.

Good information: 2.1 or above at university,
publications in high-ranking journals, relevant
experience.

Bad information: 2.2 or lower, publications in lowranking journals, gaps in employment history.
Publications on one of the C.V.s

Participants read through CVs, and then given a
further minute to make decision.

This ‘critical minute’ formed the basis of our
analysis as during this time period participants
scan rather than read the C.V.s.

After seeing all four applicants for a given post,
participants were asked to shortlist two candidates.

Eye movements were recorded for a sample.

Afterwards, explicit and implicit attitudes were
measured.
Eye tracking clip
Example of eye movements

Red dot = fixation point (with time sampling)
Shortlisting decisions:
Our participants were not colour blind.
1.
White Participants: 77.1% of shortlisted candidates (academic
post) were White.
2.
60.4% of White participants shortlisted 2 White candidates.
3.
White participants were ten times more likely to shortlist two White
candidates than two non-White candidates.
1.
Non-White Participants: 61.5% of shortlisted candidates
were non-White.
2.
68.8% chose one White and one non-White candidate (only 4.2% chose 2
White candidates).
Explicit attitudes do not predict
which candidates were shortlisted
(Likert)
Academic Job
2 Non-White
candidates
shortlisted
2 White
candidates
shortlisted
3.00
3.00
Implicit attitudes (‘D’ scores) of White
participants predict shortlisting decisions
Shortlisted two
White applicants
Shortlisted one
White, one nonWhite applicant
Shortlisted two
non-White
applicants
Mean D score:
1.17
Mean D score:
0.71
Mean D score:
-0.22
(strongly pro-White)
(moderately pro-White)
(slightly pro-non-White)
Implicit attitudes (‘D’ score) of non-White
participants predict shortlisting decisions
Shortlisted two
White
candidates
Shortlisted one
White, one nonWhite candidate
Shortlisted two
non-White
candidates
Mean D score:
1.01
Mean D score:
0.52
Mean D score:
0.10
(strongly pro-White)
(moderately pro-White)
(neutral)
Effects on gaze fixations:
White participants viewing candidates for
the academic post
Proportion of time participants spent looking at good/bad
information of White/non-White candidates
D Score of Good info.
participants White
candidates
Bad info.
White
candidates
Good info.
non-White
Bad info.
Non-White
candidates
candidates
High
(1.32)
58.91
41.09
47.25
52.76
Low
(0.30)
52.19
47.81
50.86
49.15
Conclusions

Ethnicity of person making the
judgment seems to influence who is
shortlisted for academic posts but
not non-academic posts.

Self-reported attitude has no effect
on who is shortlisted.

Implicit attitude does seem to
affect who is shortlisted.
Conclusions

Strong pro-White bias spent more time
looking at the good information of White
candidates.

Our implicit (and unconscious) attitude
direct our unconscious eye movements
when we consider their C.V.s.

Our ‘rational’ decisions about the suitability
of candidates are based on this biased
pattern of fixation.
Encouraging the ‘Green Revolution’

We need to recognize that people (unfortunately) have two
minds not one.

We need to understand that many people have implicit and
explicit attitudes that are dissociated.

These different attitudes direct behaviour in different
domains, as a function of different variables (time, cognitive
load, emotion).

We need to find new ways of identifying these individuals
(moving away from self-reports).
The ‘Green Revolution’

We need to understand how those with dissociated
attitudes process incoming information.

What do these individuals see? What do they attend
to?

We need to find new ways of changing implicit
attitude in the area of sustainability.

Emerging methods for measuring implicit attitudes
will tell us if these methods are working or not.
Tackling implicit racism

We need to introduce System 2 thinking.

We need to interfere with the ‘natural connection’ between
implicit processes and behaviour (no ‘first thoughts’ in
shortlisting).

We need to interfere with the connection between
unconscious and conscious behaviour (e.g. implementation
intention, ‘if…then’ formulations).

We need to change contextual factors (no time pressure in
shortlisting, lower the emotional/cognitive load in shortlisting).
But most importantly of all:
We need to work on changing our
implicit attitude in this domain by
influencing our underlying
associative networks.
Why do I think it is possible to change
associations in this way?
(Melinda Gates says that we need to learn hard lessons from the
commercial world)

Smoking.

The one positive legacy of the tobacco
industry!

Some of the best psychologists of a
previous generation worked on this.

If we can make people smoke, we can help
them do anything.
Finally:

I called this talk ‘The Divided Self’.

I hope that I’ve convinced you that this is an
accurate description

The subtitle of the talk was: ‘How a Better
Understanding of the Human Mind Could Transform
Society’.

I believe that some positive transformation is
possible (these are challenges that we could
meet head on) but that depends upon all of
us, both inside and outside the academy.