Book summary “Attitudes and attitude change”: summary

Introduction

You often struggle to predict what people will do and find it difficult to change stubborn minds. Human behavior feels erratic because underlying opinions remain hidden behind polite words and shifting social contexts. You need a reliable framework to understand how opinions form, why they resist persuasion, and how they actually drive choices.

Attitudes are evaluative integrations of affect and cognition that drive behavior, resist persuasion, and adapt to social and cognitive contexts.

William D. Crano explains the mechanics of human evaluation in Attitudes and attitude change. You will discover how opinions form, how persuasion succeeds or fails, and why strong beliefs endure against counterattitudinal attacks. The book maps a clear path from basic measurement techniques to complex social identity dynamics.

You will find answers to several core questions as you read:

• How do you accurately measure internal evaluations that people try to hide?
• What triggers the formation of a brand new attitude in your mind?
• Why do some persuasive messages alter behavior while others fail completely?
• How do strong attitudes protect themselves against opposing arguments?
• When do personal opinions actually translate into predictable actions?

1. Nature and Measurement of Attitudes

Attitude models trace back to Thurstone scaling and Osgood semantic differential methods. These foundations established evaluation as the core of the construct.

Attitudes are fundamentally evaluative integrations of cognitions and affects experienced in relation to an object, varying in strength.

Traditional views treat attitudes as representational things residing in memory that are polled when queried. More recent constructionist views argue that attitudes are judgments formed on the spot. Your internal state and the immediate social context determine the final outcome.

Explicit and Implicit Evaluations

Explicit attitudes reflect conscious, deliberate evaluations. You typically assess them using self-reported evaluations and semantic differential scales.

Implicit attitudes represent automatic, unconscious traces of past experience. You measure them via response times linking an attitude object to evaluative categories like good or bad.

Researchers use methods like the Implicit Association Test to assess automatic attitudes toward objects or groups, such as President Bush or racial categories. Implicit approaches attempt to capture evaluations unadulterated by a desire to paint a flattering self-portrait. Validating these measures presents a challenge because researchers often lack a definitive external criterion. They must compare implicit results with manifest response methods or behavioral outcomes.

Context Dependency and Measurement

Direct attitude self-reports are profoundly context dependent. Minor changes in question wording, format, or order alter your results.

Because respondents rarely search their memory comprehensively, they truncate information retrieval once enough data is found. Your judgments become sensitive to whatever information is temporarily accessible due to preceding context.

Accessible information can lead to assimilation effects or contrast effects:

• assimilation pushes judgments in the direction of the target representation when included;
• comparison based contrast triggers if information is excluded or used as a standard.

Minor shifts in survey design or question sequencing dramatically change statistical correlations. Asking about marital satisfaction before general life satisfaction raised their correlation from r = .32 to r = .67. Adding three other life domains intervened to yield r = .43. A joint lead-in to evoke conversational separation lowered the correlation to r = .18, while a reworded control condition yielded r = .20.

Respondents edit their private attitude judgments to present themselves favorably. Social desirability biases are more pronounced in face-to-face interviews than in self-administered questionnaires.

White respondents frequently mute negative sentiments about African Americans when interviewed by a black interviewer rather than a white interviewer. Researchers combat response editing using confidentiality tools and techniques like the bogus pipeline.

2. Attitude Formation and Origins

Attitudes emerge from distinct psychological mechanisms. You form evaluations through associative conditioning or reasoned calculations. These processes combine feelings, beliefs, and past actions into summary impressions.

Attitudes originate from affective conditioning, cognitive expectancy-value calculations, genetic predispositions, and evolutionary pressures.

Associative and Affective Roots

Evaluative conditioning enables attitude formation and change without conscious awareness of contingencies. It produces responses resistant to extinction.

This process operates via affective conditioning through the mere contiguity of neutral and affective stimuli. A neutral stimulus acquires the valence of a liked or disliked stimulus simply by being presented together.

Olson and Fazio found that co-occurring a fictitious character like a Pokémon with negative words or pictures leads to negative reactions toward the character.

Attitudes can form via classical conditioning without conscious awareness, through repeated pairings of neutral stimuli with positive or negative items. In experiments by Olson and Fazio in 2001 and 2002, participants performed a surveillance task while Pokémon cartoon characters were systematically paired with positive words, puppies, or negative items. This procedure involved twenty total pairings of each conditioned stimulus with paired associates, leading to unconscious preferential evaluations.

Evaluative conditioning transfers affective responses from one stimulus to initially neutral stimuli regardless of presentation order. The US-revaluation paradigm supports stimulus-stimulus learning. Changing the evaluation of a source object changes attitudes toward associated stimuli even if those stimuli were never paired with the new information. Walther and colleagues paired positive unconditioned stimuli with negative statements and negative unconditioned stimuli with positive phrases. They observed a corresponding shift in the affective quality of the associated conditioned stimulus.

Cognitive and Genetic Origins

Attitudes serve as summary evaluations composed of three elements known as the ABCs of attitudes: affect, cognition, and behavior. The tripartite approach organizes research by viewing attitudes as summaries combining feelings, beliefs, and past actions toward an attitude object. These components do not need to be internally consistent with one another.

An individual's attitude toward safe sex can be derived by evaluating their positive beliefs about disease prevention, their positive affect toward the practice, and their actual safe sex habits.

According to Fishbein and Ajzen's expectancy-value model, cognitive attitudes are formed by multiplying the perceived likelihood of an attribute occurring by the evaluation of that attribute. You sum the products across all relevant attributes to derive the overall evaluation.

When forming an attitude toward safe sex, a person assesses the probability that safe sex prevents sexually transmitted diseases and multiplies it by how much they value that positive health outcome.

Twin studies reveal that a surprising variety of social, political, and personal attitudes have a strong genetic component and high heritability indices. Genes do not link directly to specific attitudes. They express themselves indirectly via physical, sensory, personality, and intelligence variables.

Martin and colleagues demonstrated in 1986 that attitudes toward the death penalty, jazz music, censorship, and religion have high heritability indices. Patriotism and belief in the veracity of the Bible show low heritability by comparison.

3. Persuasion and Information Processing

Dual-process models differentiate between effortful, systematic message processing and effortless processing of superficial cues. Prominent frameworks like the elaboration likelihood model and the heuristic-systematic model distinguish between effortful information parsing and effortless focus on extraneous cues like source attractiveness. Both models require motivation and ability as conditions for effortful processing, though they conceptualize effortless processing differently.

Persuasion occurs through dual-process models, heuristic-systematic reasoning, and functional matching that aligns messages with psychological needs.

Dual-Process and Unimodel Frameworks

The elaboration likelihood model proposes two antagonistic routes to persuasion depending on elaboration likelihood: the central route via effortful message elaboration and the peripheral route via superficial cues. Motivation and ability to process determine the position on the elaboration likelihood continuum. Variables can play multiple roles in persuasion, such as serving as peripheral cues, message arguments, or factors influencing the amount and direction of elaboration.

Undergraduates watching a pen commercial were influenced by attractive music when they made no decisions, but factual information became more influential when they expected to make a choice.

The heuristic-systematic model distinguishes between an effortless heuristic mode and a more effortful systematic mode, allowing both modes to co-occur simultaneously. Unlike the elaboration likelihood model, it does not assume strict route antagonism. It proposes four co-occurrence hypotheses and operates under the sufficiency principle, where people process information to bridge the gap between actual confidence and their sufficiency threshold.

Participants with high importance varied source credibility and argument ambiguity for an answering machine and showed that they assimilated ambiguous message evaluations to the credibility cue.

The unimodel of persuasion argues that cues and message arguments are functionally equivalent pieces of evidence that do not require qualitatively distinct processing modes. Apparent differences in dual-process research often stemmed from confounding informational length, complexity, and ordinal position. Once these parameters are controlled, cues and arguments function as equivalent evidence evaluated through prior beliefs and reasoning.

Priming the belief cream equals fat or cream equals taste via an unrelated procedure shifted attitudes toward a new ice cream brand.

Functional Matching and Involvement

Daniel Katz deductively categorized attitudes into four distinct functions based on psychological needs. He analyzed Gestalt, learning, and psychoanalytic theories to identify the instrumental, knowledge, ego-defensive, and value-expressive functions.

Housewives listened to daytime radio serials because the shows provided information and advice, serving a knowledge function.

The functional matching hypothesis posits that persuasive messages are significantly more effective when they match the specific psychological function served by the recipient's attitude. Attitude change occurs to meet a functional need, making messages that target relevant concerns far more persuasive than irrelevant ones.

The Liberal Party of Australia won an election by campaigning heavily on housing loan interest rates, directly addressing the electorate's underlying economic security concerns.

Individual differences in self-monitoring dictate whether you are more persuaded by social-image appeals or quality-based arguments. High self-monitors adjust their behavior to fit social circumstances and are more responsive to social-adjustive and image-based appeals, whereas low self-monitors base behavior on internal values and respond better to quality or expert appeals.

Participants were exposed to Canadian Club whiskey advertisements featuring either an image-based slogan or a quality-based slogan. High self-monitors preferred the image ads and indicated they would pay more money for them.

4. Resistance and Attitude Strength

Attitude strength and resistance appraisals determine whether evaluations persist over time, withstand counterattitudinal attacks, and guide behavior.

Because seemingly identical attitudes result in widely varying behaviors, researchers developed the construct of attitude strength. Strength features include meta-attitudinal indicators based on self-impressions and operative indicators originating from underlying judgmental processes.

Attitude Strength and Ambivalence

Krosnick and Petty defined strong attitudes using durability and impact to avoid confusion from the multiplicity of historical labels. Researchers focus on four primary strength properties:

• extremity;
• importance;
• accessibility;
• ambivalence.

Attitude accessibility reflects how easily an attitude comes to mind, with response latency serving as its most common measure. Accessible attitudes are activated automatically from memory upon encountering the attitude object and guide behavior. They are more stable and resistant to persuasion.

Bassili found that for every second longer it took a respondent to express a voting intention, the probability of casting a matching ballot dropped by about 8%.

Attitudinal ambivalence occurs when individuals simultaneously hold positive and negative evaluations of an attitude object, challenging the traditional unidimensional view of attitudes. While traditional models view attitudes as purely bipolar, the bidimensional model recognizes that you can evaluate an object as both positive and negative at the same time.

Citizens often experience reservations about foreign policy while remaining patriotic, or view drinking alcohol as both sociable and inappropriate.

Ambivalence is measured through two distinct approaches: potential ambivalence, which combines separate positive and negative reactions, and felt ambivalence, which measures subjective emotional conflict. Potential ambivalence uses unipolar scales for positive and negative thoughts and combines them mathematically, whereas felt ambivalence asks you to rate your subjective discomfort or felt conflict.

Priester and Petty asked participants to rate subjective felt ambivalence on 11-point scales anchored from feel no conflict at all (= 0) to feel maximum conflict (= 10).

Resistance Appraisals Framework

Successful resistance to persuasion is not merely a passive state but has active meta-cognitive consequences for targeted attitude strength.

The resistance appraisal framework challenges the assumption that resistance leaves attitudes untouched. Instead, perceptions of resistance and the context in which it occurs combine via meta-cognitive factors to influence post-resistance attitude strength.

According to the attributional logic of the resistance appraisals hypothesis, individuals observe their own resistance behavior and form inferences about their attitudes. Positive appraisals of resistance performance increase attitude certainty, whereas negative appraisals undermine it, even when actual resistance behavior is identical.

Researchers have demonstrated that people infer the validity of their attitudes by evaluating how well they handled persuasive appeals.

When you believe you have resisted an exceptionally strong persuasive attack or an expert source, you view your resistance as more diagnostic of your attitude's validity. This acts as an augmenting factor that boosts attitude certainty, making the attitude more predictive of behavior and more resistant to future attacks.

In experiments by Tormala and Petty, participants who were led to believe they resisted a strong message or an expert source became significantly more certain of their initial attitudes.

5. Behavior Prediction and Social Identity

Behavioral consistency depends on attitude accessibility, the theory of planned behavior, cognitive dissonance, and group-normative social identities.

Because global attitudes do not specify action, context, or time, they fail to predict single specific actions. However, they successfully predict multiple-act behavioral aggregates that generalize across various contexts and times. A strong correlation between attitude and behavior requires that both measures match precisely in their action, target, context, and time elements.

In a study on religiosity by Fishbein and Ajzen, global attitude measures showed a mean correlation of .14 with individual behaviors, but strong correlations ranging from .61 to .71 with an aggregate measure across 100 behaviors. The cross-study correlation between degree of compatibility and the magnitude of the attitude-behavior relation reaches .83. Kraus's meta-analysis found a .54 mean correlation when predicting behavior from attitude toward the behavior.

Behavioral Models and Cognitive Dissonance

The theory of planned behavior states that human action is driven by behavioral intentions, which are shaped by attitudes toward the behavior, subjective norms, and perceived behavioral control. Instead of broad global dispositions, the model focuses on specific behaviors. Attitudes stem from behavioral beliefs, subjective norms stem from normative beliefs, and perceived behavioral control stems from control beliefs.

Perceived behavioral control can also influence behavior directly. Sheeran reported a .53 mean correlation between intention and behavior across diverse behavioral domains. Perceived behavioral control accounted for 12% extra variance in smoking behavior over and above intentions. Conner, Sherlock, and Orbell studied club members' intentions to use ecstasy, finding that participants with positive intentions differed significantly from those with negative intentions in their behavioral, normative, and control beliefs.

Leon Festinger published his book "A Theory of Cognitive Dissonance" in 1957. When a person's behavior contradicts their attitudes or beliefs, it produces cognitive dissonance, creating an uncomfortable state of psychological tension that motivates them to restore consistency. Inconsistency among cognitive elements generates a negative drive state similar to hunger or thirst. Because behavior cannot always be undone or changed, people often resolve this tension by distorting their perceptions and changing their attitudes and beliefs.

Smokers trying to quit or dieters succumbing to temptation experience dissonance when their actions clash with their knowledge of health risks. Cognitive dissonance theory explains how behavior shapes attitudes through psychological discomfort caused by inconsistencies, moderated by personal self-standards. Cognitive dissonance occurs when actions conflict with internal standards, including failures to act according to standards. The self-standards model explains that variations in dissonance stem from different personal standards and self-esteem levels, which can motivate behavioral changes toward advocated norms.

Social Identity and Minority Influence

Attitudes originate from group prototypes as expressions of social identity. People represent social groups as category prototypes that simultaneously describe and prescribe group-normative attitudes. Category prototypes act as a fuzzy set of attributes capturing similarities within a group and differences from out-groups. These prototypes obey the metacontrast principle by maximizing the ratio of perceived intergroup differences to intragroup differences, which polarizes norms.

Because validity derives from in-group norms, messages from in-group sources receive closer scrutiny and greater influence than out-group sources, balancing the symmetrical approach to social influence. Social identity theory links attitudes to group membership, making in-group communication and minority influence especially powerful. Numerical minorities within an in-group can still trigger significant persuasion.

Crano's context/comparison model introduces the leniency contract, explaining how in-group minorities can persuade majority members on vested attitudes without triggering source derogation. When a message concerns vested or central attitudes, targets resist identifying with a minority but are also reluctant to derogate other in-group members. This leniency contract allows majority members to elaborate openly on an in-group minority's message without defensiveness, leading to indirect attitude change.

10 Key Ideas

1. Explicit attitudes reflect conscious, deliberate evaluations measured via self-reports, whereas implicit attitudes represent automatic, unconscious traces of past experience measured via response times.
2. Evaluative conditioning transfers affective responses from one stimulus to initially neutral stimuli through mere contiguity without requiring conscious awareness of contingencies.
3. According to Fishbein and Ajzen's expectancy-value model, cognitive attitudes are formed by multiplying the perceived likelihood of an attribute occurring by the evaluation of that attribute and summing the products across all relevant attributes.
4. Dual-process models differentiate between effortful, systematic message processing and effortless processing of superficial cues based on motivation and ability.
5. The functional matching hypothesis posits that persuasive messages are significantly more effective when they match the specific psychological function served by the recipient's attitude.
6. Attitude accessibility reflects how easily an attitude comes to mind, serving as a primary strength property that guides behavior and resists persuasion.
7. Attitudinal ambivalence occurs when individuals simultaneously hold positive and negative evaluations of an attitude object, challenging the traditional unidimensional view of attitudes.
8. The theory of planned behavior states that human action is driven by behavioral intentions, which are shaped by attitudes toward the behavior, subjective norms, and perceived behavioral control.
9. When a person's behavior contradicts their attitudes or beliefs, it produces cognitive dissonance, creating an uncomfortable state of psychological tension that motivates them to restore consistency.
10. Attitudes originate from group prototypes as expressions of social identity, where in-group norms validate messages and enable in-group minorities to persuade majority members without triggering source derogation.