Explain choice blindness and mistakes in consumer decisions

Fast and slow thinking describe modes of processing, not two fixed kinds of person

System 1 is a useful label for fast, automatic, associative and low-effort processing. It can produce fluent everyday decisions without a shopper comparing every alternative. It can also lean on an incomplete memory, a misleading frame or a poor match between a cue and the current task.

System 2 describes slower, deliberate and effortful thought, such as checking unit prices or comparing ingredients. It can question an intuitive answer, but it is not guaranteed to be correct: a person can use incomplete information, make an arithmetic error or rationalise a choice after the fact. Effort also has a cost, so shoppers may not engage in detailed analysis for every small purchase.

Shleifer (2012) is a review of Daniel Kahneman’s book and the research behind it, not a new experiment on shoppers. The systems are explanatory frameworks, not literal brain regions or permanent personality categories. A person may use both modes in one decision: notice a familiar package quickly, then deliberately inspect its price and ingredients.

Use the framework to generate a testable question, such as whether a time limit changes the comparison strategy. Do not infer that an individual was ‘using System 1’ solely because the final choice seems mistaken; measure the process or state the inference cautiously.

Choice blindness is a mismatch between an intended choice and the outcome a person receives

In a choice-blindness task, someone makes a choice, the researcher covertly swaps the selected item for the unchosen alternative, then asks the participant to explain the choice they believe they made. Choice blindness occurs when the mismatch goes undetected and the person may defend or explain the substituted option as if it had been chosen.

Hall, Johansson, Tärning, Sikström and Deutgen (2010) set up a tasting venue at a local supermarket. Passers-by sampled two varieties of jam and two varieties of tea, chose a preferred item within each pair, then sampled what they were told was the chosen item. Researchers secretly switched the sample contents and invited participants to explain their preference.

No more than one third of manipulated trials were detected. Even for markedly different items—cinnamon-apple versus bitter grapefruit jam, or mango versus Pernod tea scent—detection reached no more than half of trials. The result shows that a person may not always notice a choice-outcome mismatch in this tasting procedure; it does not prove that people never know what they prefer.

The study combined an experimental manipulation with verbal explanations in a real shopping location. The tasting task makes the choice concrete, but the items and context are specific, and the verbal account may be shaped by the prompt, memory or a wish to give a coherent reason. A study involving food should screen for allergies and dietary needs; deception requires suitable ethical review, a careful debrief and a meaningful opportunity to withdraw data where possible.

Advertising can interfere with remembering a particular brand’s message

Retroactive interference occurs when later information disrupts recall of something learned earlier. In advertising, later exposure to other ads may make it harder to remember a target brand’s message.

Proactive interference occurs when earlier learning disrupts recall or learning of later information. Earlier ads can make it harder to encode or retrieve a newer message. The two labels refer to the direction of interference across time; they are not simply two names for forgetting.

Burke and Srull (1988) reported three experiments on consumer memory for advertising. In Experiment 1, later ads for other products in the same manufacturer’s product line and for competing brands inhibited memory for a brand’s ad—retroactive interference. Experiment 2 found an analogous proactive effect. In Experiment 3, competing-brand ads changed how repetition related to recall: repetition improved recall only when there was little or no advertising for similar products.

The findings suggest that an ad’s memory effect depends on the surrounding competitive message environment, not only its repetition. They do not mean every additional ad erases a prior one. Controlled ad exposure can help isolate interference, while actual viewing, divided attention and real purchase contexts are more complex. A reliable replication needs consistent ad materials, exposure timing, recall instructions and scoring rules.

Individual and situational explanations can be evaluated together

An individual explanation may focus on prior preference, familiarity, memory, attention or a person’s reasons. A situational explanation may focus on a covert switch, the wording of an interview question, recent competing ads, time pressure or which samples were available. A choice-blindness result can involve both the participant’s memory and the task context.

Determinism asks whether these influences help explain a response; free will concerns a person’s capacity to notice, reflect and choose. Evidence that some manipulated trials go unnoticed does not show that preferences are always unconscious or that consumers have no agency. Equally, a confident explanation after a choice does not prove that the stated reason caused it.

Experiments can manipulate a switch or ad sequence and compare responses under controlled conditions. Interviews can capture verbal reasons but may invite demand characteristics or post-hoc rationalisation. Quantitative detection rates and recall scores can be compared, while qualitative explanations need clear coding and, where feasible, inter-rater checks. Reliability means the procedure and scoring are consistent; consistency alone does not prove the interpretation is valid.