Week 06
Consumer Behavior · Davood Wadi, PhD
Consumers do not just evaluate products. They feel, think, and act toward them.
An attitude is a lasting evaluation of a person, object, or idea.
Attitudes can be favorable, unfavorable, or somewhere between the two.
They help consumers simplify repeated choices.
They can also make consumers resist new information.
Affect means feelings and emotions toward an object.
Behavior means actions or intentions, such as buying, recommending, or avoiding a brand.
Cognition means beliefs and thoughts about an object.
The three parts often support each other, but they can also conflict.
Each card is one thing a consumer says or does about a brand. Is it affect, behavior, or cognition?
Tap a card, then tap the group it belongs to.
The utilitarian function helps consumers gain benefits or avoid costs.
The value-expressive function communicates identity, values, and self-image.
The ego-defensive function protects self-esteem from uncomfortable threats.
The knowledge function organizes information and makes the world easier to understand.
Match each function of attitudes with the consumer whose attitude serves it.
Function
Consumer
Tap a term, then the example it goes with.
The standard learning hierarchy is think, feel, then act.
The low-involvement hierarchy is think briefly, act, then develop a feeling.
The experiential hierarchy is feel, act, then explain.
The hierarchy depends on the product, the consumer, and the purchase situation.
Each case describes how an attitude formed. Which hierarchy is it?
Standard learning: think, feel, then act.
Low involvement: think briefly, act, then develop a feeling.
Experiential: feel, act, then explain.
Standard learning: beliefs come first, then a feeling, then the purchase.
Experiential: the feeling leads, the action follows, and the explanation comes last.
Low involvement: a brief thought, a purchase, then a feeling.
Which concept does this case show?
Source credibility comes from expertise and trustworthiness.
Expertise makes a source seem knowledgeable.
Trustworthiness makes a source seem honest and dependable.
Source attractiveness comes from familiarity, likability, and similarity.
Each card describes the source of a message. Does its impact come from expertise, trustworthiness, or attractiveness?
Tap a card, then tap the group it belongs to.
Consumers now receive advice from AI systems as well as from people.
Algorithm aversion is the tendency to trust an algorithm less than a person. After people see an algorithm make a mistake, they lose confidence in it faster than in a person who makes the same mistake (Dietvorst, Simmons, & Massey, 2015).
Algorithm appreciation is the opposite tendency. In some tasks, such as numerical estimates, people follow advice more when they are told it comes from an algorithm than from a person (Logg, Minson, & Moore, 2019).
The task matters. Consumers trust algorithms more for tasks that seem objective, such as calculating a loan payment, than for tasks that seem subjective, such as choosing a gift (Castelo, Bos, & Lehmann, 2019).
Credibility still rests on expertise and trustworthiness. An AI source can seem expert but not trustworthy when consumers do not know whose interests it serves.
Each case shows a consumer taking advice from an AI system. Is it algorithm aversion, algorithm appreciation, or an AI that seems expert but not trustworthy?
Algorithm aversion: after one mistake, she loses confidence in the algorithm faster than in the person.
Algorithm appreciation: for a numerical estimate, he follows advice more when it comes from an algorithm.
The AI seems expert but not trustworthy, because Kim does not know whose interests it serves.
Algorithm aversion: the same mistake costs the algorithm more trust than the person.
Algorithm appreciation: in a numerical estimate, she follows the algorithm's advice.
Expert but not trustworthy: credibility needs trustworthiness, and she does not know whose interests the agent serves.
Which concept does this case show?
A positive frame emphasizes gains, benefits, or a desirable result.
A negative frame emphasizes losses, risks, or the problem that may follow inaction.
Effective framing depends on the audience, the goal, and whether the choice feels like a gain or a loss.
Each card is a line from a message. Is it a positive frame or a negative frame?
Tap a card, then tap the group it belongs to.
Researchers measure consumer attitudes with rating scales. A Likert scale asks how much a person agrees with a statement, for example from 1 (strongly disagree) to 7 (strongly agree).
Researchers now give the same scales to AI models, because models that advise and shop for consumers may carry attitudes of their own.
A model does not give one fixed answer. Asked the same question many times, it can give a different rating each time (Wadi & Fredette, 2025).
A single answer can therefore mislead. Reliable measurement asks the question many times, or reads the probability the model assigns to each point on the scale (Wadi, Ghodrat, & Philp, 2026).
Measured this way, models from different developers held different attitudes on the same consumer attitude scale (Wadi, Ghodrat, & Philp, 2026).
An attitude in a model is a pattern in its answers. It is not a feeling that the model experiences.
A brand team wants to know how a shopping assistant's AI model feels about its products. They ask the model once to rate "This brand is trustworthy" on a 1–7 scale, and it answers 6. Which conclusion fits the slide?
Choose an advertisement or product recommendation.
Identify its affect, behavior, and cognition elements.
Identify the source and the message frame.
Discuss whether a different audience would be persuaded by the same message.