PROFESSOR LEE FABRIGAR

PSYC 441

Conditioning and Modeling Approaches to

Attitude Change

1) Mere Exposure

2) Mere Association

3) Mere Observation

Mere Exposure Hypothesis (Zajonc, 1968): the repeated mere exposure of a stimulus is a sufficient condition to enhance an observer's attitude toward that stimulus

Correlational Data

• -Mere Exposure (Zajonc, 1968)

Vegetable Frequency Rating

Corn 227 4.17

Potato 384 4.13

Lettuce 142 4.00

Carrot 96 3.57

Radish 43 3.13

Asparagus 5 2.33

Cauliflower 27 1.96

Broccoli 18 1.96

Leek 3 1.96

Parsnip 8 1.92

• -similar data for word frequency of names of flowers, trees, cities, and countries ( r = .80)

Experimental Data:

- Experimental demonstrations of effects for frequency of exposure to nonsense words, ideographs, and faces

- Field studies demonstrating effects for nonsense words: Zajonc & Rajecki, 1969 presented nonsense words in the Michigan State campus paper over a one-month period at frequencies of 1, 2, 5, 10, and 25; surveys of campus students revealed effects

- Subliminal perception experiments

Moderators of Mere Exposure Effect:

1) complex vs. simple stimuli produce bigger effects

2) brief vs. long stimulus exposure produce bigger effects

3) subliminal vs. supraliminal exposure produce bigger effects

4) long vs. short delays between exposure and attitude ratings produce bigger effects

5) heterogeneous sequence vs. homogenous sequence produce bigger effects

Explanations for Mere Exposure:

• Demand Characteristics - people figure out the hypothesis being tested and act to confirm it



• Subjective Recognition - people become more subjectively familiar with frequent stimuli and this familiarity produces greater liking



• Learning (Classical Conditioning) - stimuli are presented in a pleasant context and come to associate frequent stimuli with the pleasant context which produces liking



• Response Competition - conflicting responses decrease with repeated exposures so that dominant response becomes more prevalent and weaker responses less prevalent (competing responses are unpleasant and thus decreasing competition enhances liking)



• Two-Factor (Habituation/Boredom)/Perceptual Fluency - increased exposure produces easier perceptual processing of stimuli (pleasant) until at some point boredom sets in (unpleasant)

Mere Association

1) Classical Conditioning

2) Operant Conditioning

Classical Conditioning - type of associative learning in which an initially neutral stimulus (conditioned stimulus or CS) is associated with another stimulus that is connected inherently or by prior conditioning to some response (the unconditioned response or UCR)

Types of Attitude Classical Conditioning:

1) Standard Classical Conditioning (Staats & Staats)

• -attitudes are formed by associating the attitude object with a stimulus that is inherently positive or negative (e.g., food, electric shock)

2) Higher-Order (Second-Order) Classical Conditioning (Staats & Staats)

• -attitudes are formed by associating the attitude object with a stimulus that the person has previously been conditioned to have a positive or negative response to (e.g., the word "good")

Experimental Evidence:

- Original Staats & Staats Experiments

• -paired neutral words with unpleasant stimuli such as electric shocks or loud noises (standard classical conditioning)
• -paired nonsense syllables with positive, negative, or neutral words (higher-order conditioning)

Explanations for Classical Conditioning Effects:

• Demand Characteristics - people are aware of the pairing, figure out the hypothesis, and act to confirm the hypothesis
• Conditioning Without Awareness - people are unaware of the pairing
• Conditioning With Contingency Awareness - people are aware of the pairing, but do not figure out the hypothesis

• -Berkowitz and Knurek (1969) "George & Ed" Experiment

• -conditioned negative attitude toward the name "George" or "Ed"
• -had participants interact with a person named George or Ed in an ostensibly different study in another room
• -participants conditioned to dislike the name George or Ed acted in a less friendly manner to the person if they had been conditioned to dislike the name

• -Subliminal Conditioning (Krosnick et al., 1992)
• -paired ambiguous behavior pictures of a woman (e.g., getting into a car, washing dishes) with subliminal positive pictures (e.g., kittens, bridal couple, child with mickey mouse doll) or subliminal negative pictures (e.g., skull, a face on fire, open heart surgery, bucket of snakes)
• -subliminal pictures influenced ratings of person
• -consistent with classical conditioning, but could be a mood effect

• -Flexor/Extensor (Cacioppo, Priester, Petty)

• -somatic responses could be associated with evaluative processes
• -flexing: approach, extending: avoidance
• -had people evaluate ideographs or nonsense words doing either flexing or extending
• -found flexing created positive attitudes and extending negative attitudes
• -demand characteristics could not account for effects because participants' assumptions did not match the results

• -Familiarity/Knowledge and Conditioning Effects (Cacioppo, Priester, Petty)
• -classical conditioning works better for nonsense words than real words
• -flexor/extensor effects stronger for nonsense words than real words

Operant Conditioning - type of associative learning that occurs when some response becomes more/less likely because of its positive/negative consequences

Experimental Evidence:

1) Essay Experiment (Bostrom, Vlandis, & Rosenbaum, 1961)

• -induced students to write counter-attitudinal essays
• -randomly assigned them to receive grades of A or D
• -students who received an A changed their attitudes to come in line with the essay more than those who received a D

2) Verbal Reward Experiment (Insko, 1965)

• -University of Hawaii students interviewed by telephone about establishing a Aloha Week festival
• -half were rewarded with "good" for endorsing statements in favor of the festival and half for endorsing statements against the festival
• -one week later, students completed survey in class that included a question on the festival
• -reward influenced attitude toward the festival

Two-Factor Theory of Verbal Conditioning:

1) Information

2) Rapport

Insko & Cialdini (1969)

• -interviewed by telephone about pay TV

Information Rapport

Good Moderate High

Good/Humph High Moderate

Humph Moderate Low

• -hmph produced less change than other conditions

Explanations for Operant Conditioning Effects:

1) Demand Characteristics

2) Conditioning Without Awareness

3) Conditioning With Contingency Awareness

Mere Observation

Observational Learning - learn which responses are rewarded and which are not by observing rather than directly experiencing the consequences of a behavior

Bandura Acquisition vs. Performance Distinction:

1) Must believe rewards associated with the model's behavior hold for oneself

2) Must believe outcome worth relative costs of performing behavior

Vicarious Classical Conditioning - classical conditioning that occurs when a neutral stimulus acquires the ability to produce an emotional reaction in a person because they have observed this neutral stimulus being paired with a stimulus that evoked an emotional reaction in another person

Berger (1962) study

• -had participants observe another person being shocked at a time when a buzzer sounded
• -participants developed arm jerk response to buzzer
• -similar effect produced for facial expressions

Message Learning Approach

The Yale Group Included:

Carl Hovland Jack Brehm

William McGuire Milton Rosenberg

Robert Abelson Timothy Brock

Irving Janis David Sears

Harold Kelley Herbert Kelman

Norman Anderson Philip Zimbardo

Muzafer Sherif

Stages of the Persuasion Process (see Figure 3):

Variables Influencing Stages of the Process:

• Source - originator of the persuasive communication
• Message - features of the communication itself
• Recipient - characteristics of who is receiving the message
• Channel - by what mode is the message communicated

Message Learning Model (Figure 4)

Characteristics of Source:

1) Credibility

• -Expertise

• -Trustworthiness
• -Hovland & Weiss (1951)
• -message on feasibility of atomic submarines
• -high credibility (Robert Oppenheimer) vs. low credibility (Pravda)
• -more attitude change for high than low credibility
• -other studies have unconfounded expertise and trustworthiness
• -generally results replicate Hovland & Weiss with some exceptions

2) Attractiveness

• -Similarity
• -Appearance
• -Brock (1965)
• -had experimenter approach people in a paint store about to buy paint and suggest a different brand
• -half were told by the experimenter that he had bought a similar amount and the other brand was better
• -half told by the experimenter that he had bought 20 X as much and the other brand was better
• -more people followed the similar other

3) Power

- Physical

- Social

• -Kelman's (1958) Theory of Source Factors:
• -Power ---> Compliance
• -Attractiveness ---> Identification
• -Credibility ---> Internalization

Message Factors:

1) One-Sided vs. Two-Sided Messages

• - Lumsdaine & Janis (1953)
• -message on whether USSR would produce an A-bomb in 5 years
• -received one-sided or two-sided message
• -those who initially agree were more persuaded by one-sided and those who initially disagreed were more persuaded by two-sided
• -two-sided people later less susceptible to counter-persuasion

• -McGuire's inoculation effect
• -people can be inoculated against persuasion in a way similar to a biological vaccine
• -examined persuasion for cultural truisms (brush teeth after meals, mental illness in not contagious)
• -inoculation condition - received a weak attack on truisms along with refutations
• -supportive condition - gave arguments in support of truisms
• -control condition: no defense
• -two days later received a strong persuasive message
• -both defenses worked, but the inoculation worked better
• -works because people are encouraged to develop arguments to defend their attitudes

2) Order of Messages

• -Hovland, Campbell, & Brock, 1957; Haugtvedt & Wegener, 1994)

• -Primacy Effect: first message works best
• -Recency Effect: second message works best
• -Attitude Strength Perspective: primacy when people are motivated/able to think carefully and recency when people are not motivated/able to think carefully (Haugtvedt & Wegener, 1994)

3) Comprehensibility

• -Eagly (1974)
• -gave people reasoned arguments, sentences cut in half and rematched, same words totally jumbled
• -greatest persuasion in reasoned arguments condition

4) Number of Arguments

5) Rewards Paired with Message

6) Repetition

7) Conclusion Drawing

Recipient Factors:

1) Intelligence (McGuire)

• -Comprehension increases with intelligence

• -Yielding decreases with intelligence
• -Thus, there is a curvilinear effect of intelligence on persuasion with the most persuasion occurring at moderate levels of intelligence (i.e., smart enough to understand but too dumb to resist)

2) Self-Esteem

• -data do not provide clear evidence

3) Gender (Eagly)

• -Socialization hypothesis: woman are more sensitive to social accommodation and thus more likely to be persuaded

• -Knowledge/Interest hypothesis: no gender difference, early work always used topics that men were more likely to have knowledge of, when topics that women know are used, men more persuaded

Channel Factors:

1) Face-to-Face vs. Mass Media

• -face-to-face usually works better

2) Channel Attributes

- complex message better in print

- yielding to what is comprehended is most likely in audio-visual and least likely in print

The Sleeper Effect:

Definition - an effect in which a persuasive message has a greater impact on attitudes after a delay compared to immediately following the message

Conditions Necessary for Sleeper Effect

(Pratkanis, Greenwald, Leippe, & Baumgardner, 1988)

1) Powerful Message

2) Strong Discounting Cue Following the Message (Low Credible Source)

3) Rate Trustworthiness of Source Immediately Following Cue

Differential Decay Hypothesis (Pratkanis, Greenwald, Leippe, & Baumgardner, 1988)

1) Effect occurs when arguments and cue have opposite but near equal impact

2) Cue and arguments not well integrated in memory

3) Impact of cue decays faster than argument

(See Figures 5a and 5b)

Oct. 6, 2004