Values, attitudes and behaviours: Which box would you tick?

September 3, 2012 by
Filed under: Social Brain 

Here is an important and rather difficult multiple choice question for people interested in behaviour change. Which of the following best represents your understanding of what makes people change how they think and act?

1) Attitudes Drive Behaviour

i.e. You need to change what people think before you can change what they will do. e.g. I  have come to really value the environment, therefore I have started behaving in a more environmentally friendly way.

2) Behaviour Drives Attitudes

i.e. You need to change what people do before they will change what they think e.g. I started recycling and now I find I really care about not wasting stuff.

3) Both 1 and 2 apply, but in different contexts.

i.e. It depends on the definitions and contexts of the terms in question. e.g. Your attitude to a specific behaviour e.g. smoking, might be a strong predictor of your tendency to smoke, but this doesn’t mean attitudes drive behaviour in general.

4) Both are false, there are never just these two discrete factors in play (and what on earth does ‘drives’ mean anyway…)?

i.e. attitudes and behaviours are just conceptual constructs with fuzzy edges, and they are both always influenced by numerous factors other than each other e.g. I may think voting is a good idea and this may or may not be because I enjoyed voting in the past, but if I don’t vote on a particular day it might be because of my an idiosyncratic mood or the weather, not because of general attitudes or behaviour.

5) To understand behaviour and attitudes, and their relationship, you really need a deeper understanding of values.

i.e. whatever the relationship between behaviours and attitudes, perhaps both are driven by or somehow underpinned by deeply help but somewhat unconscious values. e.g. If I steadfastly use reduce power and water consumption at home after receiving information on the subject, this might be related to wanting to save money, or wanting to save the planet, but you’ll never know which- and what follows for how to achieve similar impact- until you have a better sense of how my values underpin what I do and say.

None of these five options are strictly wrong, which is why it would make a terrible multiple choice question!

In my view, one and two are equally right and wrong because they are partial, which means three must be true, but on reflection that is also only partial. At a deeper level, the philosopher inside knows that four must be closer to the truth, but it’s a bit harsh and unhelpful because we need some heuristics to work with, people care about behaviour and attitudes and you don’t make much practical progress by rejecting all available working theories and concepts out of hand.

If I had to tick one box, I think I would therefore now opt for number five, but let me take this chance to make better sense of options 1-4 in passing:

On the attitudes/behaviour link(briefly):

Through Twitter (Hat tip Jim Mintz) I came upon an excellent overview on the behaviour-attitudes relationship. I distilled this source into the following points:

‘Attitudes’ are not the same as ‘reported attitudes’ because people cannot be relied upon to say what they really think. (Though the ‘bogus pipeline‘, where feasible, can be a good way round this problem).

However, stated attitudes do predict behaviour quite well when

1) Other influences are minimized.

i.e. the behaviour is not heavily related to contextual or situational features.

2) The attitude is specifically related to the particular action.

i.e. your attitude to jogging is a good predictor of whether you jog, while your attitude to fitness may not be.

3) The attitude is particularly potent.

i.e. Your attitude to your health-related habits may change if your life is suddenly at imminent risk because of them.

On the other hand, behaviour is more likely to drive attitudes when one of the following theories applies:

1) Self-presentation theory
People who care about how they are thought of will adapt their attitude reports to appear consistent with their actions to others.

2) Dissonance theory:  We feel tension after acting contrary to our attitudes and to reduce this discomfort we internally justify our behaviour. This is not just about self-presentation but about our own experience of internal coherence. Moreover, the less external justification we have for an undesirable action, the more we feel responsible for it, and thus the more dissonance arises and the more attitudes change.

3) Self-perception theory: When our attitudes are weak, we simply observe our behaviour and its circumstances and infer our attitudes – ie sometimes our behaviours do not so much cause our attitudes as create them. Within this perspective, the “overjustification effect” also known as ‘crowding out‘ can be relevant- it suggests that rewarding people to do what they like doing anyway can make them less likely to do it (e.g. showing up on time, or giving blood.)

So far, so tentative, which has brought me to think more about values and how they relate to behaviours and attitudes.

On Values(briefly):

Three main sources: Common Cause/Values and Frames, The Righteous Mind by Jonathan Haidt and Values Modes all make major contributions to understanding how our values shape our attitudes and our behaviour. However,  they don’t all agree with each other- not by a long way- on what exactly values are, how exactly values have their influence and what it means for acting on that understanding.

That’s where I am now on this important background issue- trying to figure out which aspects of each of these perspectives makes the most sense, and then thinking of how that knowledge might be best applied to the major behaviour-change imperatives of our day. Any thoughts on the above, or where to look next, much appreciated.



  • Paul Newton

    The most powerful principle I see is that Behavior drives Behavior – there is no stronger force than habit.

    • Jonathanrowson

      Thanks Paul- that’s a great line. I have been thinking about habit for ages, but never thought of it as ‘behaviour driving behaviour’ until now!

  • Hank Sohota

    What if the answer is
    ‘none of the above’?

    We are
    social creatures and our deepest fear is to lose membership of what
    we consider to be the groups we belong to (or want to belong to). Of
    course, in reality, things are made considerably more complex by
    issues relating to how we want to relate to others – both individuals
    or groups – not to mention how others are simultaneously doing the
    same thing with us. Further more, recent research indicates that
    this idea could be working across multiple relationship networks. In
    other words, our behaviour is not only affected by our network of
    direct relationships but also indirect relationships such as friends
    of friends because everyone is working on anticipating

    There is
    a theory, based on mirror neurons, that our brains are heavily
    deployed as anticipatory intentionality engines. So we allocate a
    lot of resources to figuring out the intentionality of others. The
    more confident we are about being able to anticipate someone’s
    intentionality the more likely we are to trust them if we choose to
    and the more likely we are to behave in a way that will maintain our
    relationship with them. The same idea can be applied to group
    preferences, biases or tendencies. The better we can anticipate
    these the more we are able to use compliance to maintain our group
    membership – or even non-compliance or partial compliance to
    influence the group.

    all of this processing and decision-making is non-conscious to the
    point that we will post-rationalise our behaviour rather than risk
    group membership. We can even appear to hold conflicting values or
    attitudes or exhibit conflicting behaviours in different groups
    and/or contexts. In this way we can also see why our so called
    values and attitudes may not be as stable as they are often assumed
    to be. I would suggest that it is more accurate to regard them as
    being in a perpetually dynamic state contingent on social
    (conversational) interaction which can exhibit black swan and
    butterfly effects.

    In my
    experience as a management consultant I have found the application of
    values and attitudes and their relationship to behaviours highly
    problematic – no matter how they were defined. I think I am
    reaching the conclusion that, taken as reality approximations, we can
    no longer regard them as sufficiently accurate – much like Newton’s
    thinking on gravity until Einstein came along. In fact, neuroscience
    seems to be opening up a radically new direction of thinking in which
    our notions of self, others, group dynamics, behavioural motivations,
    change, management, leadership, governance etc. are all being radically

    I have found such thinking to be just as unpopular as the theory of
    relativity when it was first presented – especially amongst
    executives and consultants.

    • Jonathanrowson

      Hi Hank, Thanks for that. I can’t speak for executives and consultants, but many are looking very actively at the implications of neural and behavioural sciences. Have you seen this?

      • Hank Sohota

        … indeed I have and wonderful it is too. It’s the reason I follow your web presence. I am hugely enthusiastic about the RSA’s Social Brain project.

        I am more than encouraged by all the work being done in this area but by the same token I am disturbed by how much it is effectively being ignored (or worse misinterpreted) by many traditional institutions in particular business schools and management consultancies (especially in relation to management, leadership and governance), corporations, the financial sector, governments, political parties, think tanks (especially policy and economics), the list is endless.

        However, one hope I cherish is the notion of this kind of thinking infiltrating and underpinning the future development of the social enterprise sector.

        So keep up the good work.

  • MatthewMezey

    Hi Jonathan,

    Have you come across the the Values Modes guys’ work on ‘VBCOP – A Unifying Campaign Strategy Model’?

    One paper is here:

    Thery write:

    VBCOP stands for Values, Behaviour, Consistency, Opinion, Politics and the model is basically this:

    Define an action that resonates with the values of a target
    audience [V]

    Secure the behaviour [B]

    Utilise consistency heuristic (ie my opinions adjust to
    match my behaviour) [C]

    Reveal the resulting opinion (what I believe in or am in
    favour of) [O]

    Deploy that to change politics [P]

    “Ideally, campaign planners should start by using their knowledge of the dynamics of the issue and the role of politics, to plan backwards, for example, through the sort of opinion required: whose opinion, which ways it is measured, which channels and messengers are used to relay it, etc. The behaviour would also ideally be one that was desirable ‘in itself’ [ie contributed incrementally to a solution]. However it is more important that the behaviour is strategic [1] – it might have an indirect effect rather than any additive direct solution effect – and in this model it is assumed that the end delivery is actually political.”

    Good luck getting to the bottom of it all!

    I suspect that each person responds differently – depending on a whole range of genetic/situational/psychological factors. Simple rules are probably always wrong.

    Mischel though he proved in 1968 that personality disposition was nothing when it came to predicting behaviour, and situation everything. It slowly turned out that Mischel was fairly wrong about this… (I wrote something about Mischel on one of Ben’s recent blog posts…)


  • Steven

    Fantastic, now where is the market research industry and

  • Steven

    Where is the market research industry and business on all this? Why are they still pumping out correlates dressed as fact. Interestingly there are tools of complexity analysis that help; maybe leaders just want fraudulent facts to justify their jobs

  • Anfal

    personality contributes to attitude, and it may in some cases change values