When Do Our Attitudes Actually Predict Our Behaviour?

We tend to assume that what people believe shapes what they do. Someone who values health should exercise regularly. Someone opposed to littering should pick up after themselves. But social psychology has long grappled with an uncomfortable truth: attitudes and behaviour don’t always line up. Since the late 1960s, researchers have tried to untangle exactly when and why our attitudes actually drive our actions – and the answer turns out to be surprisingly conditional.

The gap between attitude and behaviour: a brief history
True attitudes vs. expressed attitudes
Uncovering true attitudes: the bogus pipeline
The Implicit Association Test (IAT)
The aggregation principle: looking at patterns, not single acts
Specificity: matching the level of attitude to the level of behaviour
Self-awareness and the attitude-behaviour link
Attitude strength: how firmly held views drive action
Attitude accessibility: what comes to mind in the moment
Situational constraints: when the context overrides the attitude
Putting it together

The gap between attitude and behaviour: a brief history

The question of whether attitudes reliably predict behaviour first gained serious attention in 1934, when Stanford psychologist Richard LaPierre found that hotels and restaurants across the United States – 184 of which had stated they would refuse service to Chinese patrons – turned away a Chinese group in practice only once. Words and actions were drastically misaligned. Decades later, in 1969, Allan Wicker reviewed 42 studies and concluded that attitudes were, at best, weakly correlated with behaviour (rarely above r = .30), prompting many researchers to question the value of studying attitudes altogether.

The tide turned as better methods emerged. A 1995 meta-analysis by Kraus, drawing on 88 studies, found a more respectable average correlation of r = .38 – and crucially, showed that several variables moderated how strong that link was. The consensus today is not that attitudes are useless predictors, but that they predict behaviour under specific conditions.

True attitudes vs. expressed attitudes

One of the most fundamental distinctions in this area is between true attitudes – what a person genuinely feels or believes – and expressed attitudes – what they report publicly. Social norms, peer pressure, and the desire to appear a certain way can all cause people to voice attitudes they don’t actually hold.

This gap matters enormously for prediction. Research consistently shows that true attitudes are far better predictors of behaviour than expressed ones. When someone’s stated view is shaped more by what they think is socially acceptable than by what they actually believe, that stated view tells us very little about what they will do.

Uncovering true attitudes: the bogus pipeline

To get around the problem of socially desirable responding, social psychologists have developed techniques aimed at revealing genuine attitudes. One early example is the bogus pipeline, first used by psychology professor Harold Sigall in 1971. Participants are told they are connected to a machine capable of detecting their true feelings – in reality, no such detection occurs – but the belief that they might be “caught” reduces their motivation to give a flattering or socially acceptable answer. Studies using this technique have found, for instance, that people report higher levels of racial bias than they do in standard surveys, suggesting the technique does surface attitudes that participants would otherwise conceal.

Though the bogus pipeline has faced ethical criticism for its use of deception, it was largely supplanted by more sophisticated implicit measurement methods, including the Implicit Association Test.

The Implicit Association Test (IAT)

Introduced by psychologists Anthony Greenwald and Mahzarin Banaji in 1998, the Implicit Association Test (IAT) measures attitudes that people may not even be consciously aware of. The test requires users to rapidly categorise two target concepts with an attribute, such as pairing “male” or “female” names with words associated with “career” or “family.” The underlying logic is that faster response times indicate stronger mental associations. Because the task is performed quickly and automatically, it is harder to consciously control or manipulate results.

IAT measures have been shown to predict behaviours such as friendliness, resource-sharing, and discriminatory decisions toward members of different groups – often more accurately than explicit self-reports. This makes the IAT a valuable tool for studying attitudes that people are either unwilling or unable to report directly.

The aggregation principle: looking at patterns, not single acts

Another key reason why attitudes sometimes appear not to predict behaviour is that researchers often compare a general attitude to a single, specific action – and that’s an unfair test. Ajzen demonstrated that part of the reason early analyses failed to find strong links was the failure to aggregate behaviours and the use of broad attitude measures to predict specific actions.

The aggregation principle resolves this by looking at behaviour across multiple instances and over time. A person’s attitude toward health may not predict whether they go for a run on any given Tuesday, but it will predict their overall pattern of health-related behaviour across weeks and months. When attitude and behaviour are measured at the same level of breadth, the predictive relationship becomes substantially stronger.

Specificity: matching the level of attitude to the level of behaviour

Closely related to the aggregation principle is the issue of specificity. One study found that a general attitude toward birth control predicted very poorly which women would actually use birth control pills, while a specific attitude toward using the pill produced good predictive power. Similarly, attitudes about recycling behaviour predict recycling better than attitudes about the environment in general.

The theory of planned behaviour, developed by Fishbein and Ajzen, formalises this insight. It proposes that the best predictors of deliberate behaviour are: (a) the person’s attitude toward the specific behaviour in question, (b) subjective norms – what they believe important others expect – and (c) perceived behavioural control – how confident they are that they can actually perform the behaviour. Together, these factors predict intention, which is the most reliable immediate predictor of action.

Self-awareness and the attitude-behaviour link

Sometimes, people simply aren’t thinking about their attitudes when they act – and that alone can break the connection. Research by University of Minnesota psychologist Mark Snyder found that favourable attitudes toward psychology experiments did not predict students’ volunteering unless they were explicitly reminded of those attitudes when the call for volunteers was made. In another study, mock jurors were more likely to render verdicts consistent with their attitudes when they were given time to reflect beforehand.

This suggests that self-awareness – actively drawing attention to one’s own attitudes – acts as a trigger that brings those attitudes into play. People higher in private self-awareness, meaning those who habitually monitor their own thoughts and feelings, show greater consistency between their attitudes and actions. In contrast, high self-monitors – people who are highly attuned to social cues and adjust their behaviour to fit situational expectations – tend to show weaker attitude-behaviour consistency, because their actions are guided more by context than by internal states.

Attitude strength: how firmly held views drive action

Not all attitudes are equal in their power to guide behaviour. Attitude strength refers to how strongly an attitude is held – how certain, stable, and personally important it is. Some attitudes, such as those about politics or religion, are deeply embedded in a person’s sense of self; others, such as preferences between ice cream flavours, are far more peripheral. Strong, personally significant attitudes are considerably more likely to translate into consistent behaviour than weak or ambivalent ones.

Attitude strength is also connected to how attitudes are formed. Attitudes formed through direct personal experience with the attitude object are more likely to predict future behaviour – partly because they tend to be held with greater confidence, and partly because they are more stable over time. Hearing about something second-hand simply doesn’t produce the same kind of firm, behaviour-guiding attitude as experiencing it first-hand.

Attitude accessibility: what comes to mind in the moment

Attitude accessibility refers to how quickly and easily an attitude comes to mind when a person encounters the attitude object. When accessibility is high, an attitude surfaces automatically whenever the relevant object is encountered – making it far more likely to shape spontaneous behaviour. When accessibility is low, the attitude may never even enter consciousness during the relevant situation.

Direct experience is again a key driver here. The more a person has interacted with an attitude object, the more accessible their attitude toward it becomes. Attitudes are also more accessible – and more predictive – when people report them frequently, or when they form them on the basis of information that is directly relevant to the behaviour in question. Accessibility, in short, determines whether an attitude is even “in the room” when a decision is being made.

Russell Fazio’s MODE model (Motivation and Opportunity as Determinants of behaviour) captures this well: when people are both motivated and have the opportunity to deliberate, they can act on carefully considered attitudes. But when either motivation or opportunity is low – when decisions are quick or automatic – only highly accessible, strongly held attitudes will reliably shape what people do.

Situational constraints: when the context overrides the attitude

Even a strong, specific, accessible attitude can be overridden by situational factors. A meta-analysis of 797 studies confirmed that social pressure and perceived difficulty significantly weaken the attitude-behaviour relationship. When it is socially costly or practically difficult to act in line with one’s attitude, people often don’t – regardless of how genuinely they hold that view.

Behaviours are more likely to be consistent with attitudes when the social situation in which the behaviour occurs resembles the situation in which the attitude was expressed. This is why people often act very differently at home than at work, or among close friends versus acquaintances – the same underlying attitude meets very different situational pressures.

Putting it together

The relationship between attitudes and behaviour is real but conditional. Attitudes are most predictive when they are: genuine rather than expressed for social approval; strong, stable, and formed through direct experience; specific enough to match the behaviour being predicted; easily accessible at the moment of action; and supported by a social situation that doesn’t impose competing pressures. Techniques like the bogus pipeline and the IAT help researchers access the attitudes that are most likely to predict real-world behaviour – those that people may not always voice openly.

Understanding these conditions doesn’t just matter for academic research. It has direct implications for public health campaigns, policy design, and any effort to understand the gap between what people say they value and what they actually do.

What do you think? Think about a time when your own behaviour didn’t reflect an attitude you genuinely held – what conditions do you think made the difference? And if our most predictive attitudes are often unconscious ones, what does that suggest about how well we actually know our own motivations?

References

PDF 📄

Table of Contents

The gap between attitude and behaviour: a brief history 🔗

True attitudes vs. expressed attitudes 🔗

Uncovering true attitudes: the bogus pipeline 🔗

The Implicit Association Test (IAT) 🔗

The aggregation principle: looking at patterns, not single acts 🔗

Specificity: matching the level of attitude to the level of behaviour 🔗

Self-awareness and the attitude-behaviour link 🔗

Attitude strength: how firmly held views drive action 🔗

Attitude accessibility: what comes to mind in the moment 🔗

Situational constraints: when the context overrides the attitude 🔗

Putting it together 🔗