# Maps of Bounded Rationality

Daniel Kahneman · 2002

Open the original and follow its chapters or sections with this study guide. Reading tasks are editorial assignments. Use the points to consider to examine your interpretation.

## Scope of source verification

Structure checked against: Public text

The inventory follows the actual headings: sections 1–6 and an unnumbered conclusion. The introductory roadmap differs from those headings. This scope does not substitute the 2003 AER version.

- [Daniel Kahneman — Nobel Lecture essay](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Public text

<a id="guide"></a>
## Read the whole work

Whole-work question: What makes an intuitive answer accessible, and under what conditions can reflective thought revise it?

Edition: NobelPrize.org essay based on the 2002 Prize Lecture

Native units: Reading divisions of the text

Form: Argument

Scope and divisions: Acknowledgments and references are excluded. Descriptive labels are editorial. Original division order is preserved. Descriptive labels, parts and learning units are editorial reading aids.

Checked: 2026-11-17

### Reading paths

Short path: [Distinguishing two modes of thought](#guide-unit-two-modes) → [Separating the target question from the answered one](#guide-unit-attribute-check) → [Comparing a typical impression with total quantity](#guide-unit-prototype-quantity) → [Returning conclusions to testable questions](#guide-unit-limits-conclusion)

Full path: [Mapping the shared research questions](#guide-unit-research-map) → [Distinguishing two modes of thought](#guide-unit-two-modes) → [Comparing equivalent presentations](#guide-unit-framing-comparison) → [Reading value relative to a reference point](#guide-unit-reference-choice) → [Separating the target question from the answered one](#guide-unit-attribute-check) → [When a known rule becomes usable](#guide-unit-corrective-cues) → [Comparing a typical impression with total quantity](#guide-unit-prototype-quantity) → [Returning conclusions to testable questions](#guide-unit-limits-conclusion)

### Parts and reading units

<a id="guide-part-accessibility"></a>
#### Conditions producing an intuitive answer

Native range: Opening: the questions studied together — Section 2: presentation and preferences

Move from the research questions to modes of thought and examine how presentation changes which information comes to mind.

<a id="guide-unit-research-map"></a>
##### Mapping the shared research questions

Native range: Opening: the questions studied together · Read

Map the areas of joint research with Amos Tversky named in the opening and distinguish explaining intuition from evaluating a desirable decision.

Question: What should we observe in this essay besides whether an answer is wrong?

Points to consider:

- Record response speed, accessible information and question wording separately.
- Avoid turning a described tendency into a fixed assessment of an individual.

Concepts: [Accessibility](#guide-concept-accessibility)

Reading tasks to check in the original:

- [Opening: the questions studied together](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open the introduction and locate the research questions about judgment, choice and presentation and the vocabulary linking them.

<a id="guide-unit-two-modes"></a>
##### Distinguishing two modes of thought

Native range: Section 1: intuition and mental accessibility · Close reading

In section 1, compare control, effort, association and rule use instead of dividing thought by speed alone.

Question: To check an answer that arrives immediately, what must a person notice and do?

Points to consider:

- Distinguish an initial impression from the judgment ultimately endorsed.
- Specify rules and attentional conditions rather than assuming deliberation guarantees correctness.

Concepts: [Two modes of thought](#guide-concept-dual-process), [Accessibility](#guide-concept-accessibility)

Reading tasks to check in the original:

- [Section 1: intuition and mental accessibility](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open section 1 and examine the characteristics of the two modes and how its figures change what is accessible.

<a id="guide-unit-framing-comparison"></a>
##### Comparing equivalent presentations

Native range: Section 2: presentation and preferences · Read

Track alternative descriptions in section 2 and separate informational equivalence from psychological equivalence in choice.

Question: Which elements must remain fixed when comparing different wordings of the same options?

Points to consider:

- Check that outcomes and probabilities are held constant and only presentation changes.
- Distinguish the attention suggested by a frame from the preferences actually observed.

Concepts: [Framing](#guide-concept-framing), [Accessibility](#guide-concept-accessibility)

Reading tasks to check in the original:

- [Section 2: presentation and preferences](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open section 2, place corresponding options side by side and inspect whether different features become salient despite equivalent information.

<a id="guide-part-valuation-judgment"></a>
#### Examining what is valued and judged

Native range: Section 3: evaluating changes in choice — Section 4: substituting a judgment attribute

Separate reference points in choice from attributes in judgment and check whether the question being answered changes unnoticed.

Depends on: [Conditions producing an intuitive answer](#guide-part-accessibility)

<a id="guide-unit-reference-choice"></a>
##### Reading value relative to a reference point

Native range: Section 3: evaluating changes in choice · Close reading

Read the account of prospect theory in section 3, distinguish a final state from a change toward it, and examine the preferences it seeks to explain.

Question: How would you examine an explanation in which two people evaluate the same final state differently?

Points to consider:

- Write the initial state, change and final state in separate columns.
- Separate explaining observed preferences from prescribing a desirable choice.

Concepts: [Reference dependence](#guide-concept-reference-dependence), [Framing](#guide-concept-framing)

Reading tasks to check in the original:

- [Section 3: evaluating changes in choice](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open figure 6 in section 3 and compare changes in value for equal-sized gains and losses from the reference point. What difference can you identify, and what limits does the surrounding text state?

<a id="guide-unit-attribute-check"></a>
##### Separating the target question from the answered one

Native range: Section 4: substituting a judgment attribute · Close reading

Use the judgment problems in section 4 to map the target attribute against an easier attribute that might replace it.

Question: How does a person seeming typical differ from a probability judgment about that person?

Points to consider:

- State the target quantity and the intuitive cue explicitly.
- Identify information that may be neglected and a rule for checking it.

Concepts: [Attribute substitution](#guide-concept-attribute-substitution)

Reading tasks to check in the original:

- [Section 4: substituting a judgment attribute](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open section 4 and compare the question with the responses to investigate whether a different attribute is being evaluated.

<a id="guide-part-correction"></a>
#### Conditions of correction and limits of inquiry

Native range: Section 5: thoughts that can correct a judgment — Conclusion: an agenda for research on intuition

Study cues for correction and modes of comparison, then revisit the conclusions together with experimental conditions.

Depends on: [Examining what is valued and judged](#guide-part-valuation-judgment)

<a id="guide-unit-corrective-cues"></a>
##### When a known rule becomes usable

Native range: Section 5: thoughts that can correct a judgment · Read

Study conditions for corrective thought in section 5 and distinguish knowing a rule from retrieving it when needed.

Question: When designing a prompt to reduce errors, what would you change and what would you test?

Points to consider:

- Propose a cue directing attention toward a relevant rule.
- Consider whether the cue changes the task itself and propose an appropriate comparison.

Concepts: [Cues for corrective thought](#guide-concept-corrective-cues), [Two modes of thought](#guide-concept-dual-process)

Reading tasks to check in the original:

- [Section 5: thoughts that can correct a judgment](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open section 5 and compare when statistical knowledge is and is not used, checking the comparison conditions of the experiment.

<a id="guide-unit-prototype-quantity"></a>
##### Comparing a typical impression with total quantity

Native range: Section 6: prototypes and quantities · Read

Use section 6 to record representative features separately from extent, such as item counts or duration.

Question: What additional information matters when evaluating two experiences with similar average impressions as wholes?

Points to consider:

- Choose the relevant quantity among average, total, count and duration.
- Examine the difference between evaluating separately and comparing side by side.

Concepts: [Extension](#guide-concept-extension), [Attribute substitution](#guide-concept-attribute-substitution)

Reading tasks to check in the original:

- [Section 6: prototypes and quantities](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open section 6 and identify quantities missed by a prototype and conditions that change attention to those quantities.

<a id="guide-unit-limits-conclusion"></a>
##### Returning conclusions to testable questions

Native range: Conclusion: an agenda for research on intuition · Read

Read the conclusion and organize the observations supporting the proposed account of intuition alongside questions needing further inquiry.

Question: What comparisons would be needed to examine this account in a new setting?

Points to consider:

- Separate what is likely to be accessible from what actually changed the judgment.
- Specify differences in participants and tasks rather than generalizing original results without conditions.

Concepts: [Accessibility](#guide-concept-accessibility), [Cues for corrective thought](#guide-concept-corrective-cues), [Extension](#guide-concept-extension)

Reading tasks to check in the original:

- [Conclusion: an agenda for research on intuition](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf) — Open the conclusion and connect the remaining research questions with the experiments in earlier sections.

### Key concepts

<a id="guide-concept-accessibility"></a>
#### Accessibility

The ease with which information or an evaluation comes to mind at a given moment; it need not coincide with relevance.

First unit: [Mapping the shared research questions](#guide-unit-research-map)

<a id="guide-concept-dual-process"></a>
#### Two modes of thought

A descriptive distinction between automatic associative processing and controlled processing involving effort and rules.

First unit: [Distinguishing two modes of thought](#guide-unit-two-modes)

<a id="guide-concept-framing"></a>
#### Framing

The possibility that equivalent information elicits different judgments or preferences depending on its presentation.

First unit: [Comparing equivalent presentations](#guide-unit-framing-comparison)

<a id="guide-concept-reference-dependence"></a>
#### Reference dependence

Evaluation in which a change or difference relative to a reference point matters, rather than only the final state.

First unit: [Reading value relative to a reference point](#guide-unit-reference-choice)

<a id="guide-concept-attribute-substitution"></a>
#### Attribute substitution

Answering a difficult assessment by evaluating a related but more accessible attribute.

First unit: [Separating the target question from the answered one](#guide-unit-attribute-check)

<a id="guide-concept-corrective-cues"></a>
#### Cues for corrective thought

Information that directs attention to relevant knowledge or rules when checking an initial impression.

First unit: [When a known rule becomes usable](#guide-unit-corrective-cues)

<a id="guide-concept-extension"></a>
#### Extension

The extent of an object or set, such as count or duration, distinguished from its average impression.

First unit: [Comparing a typical impression with total quantity](#guide-unit-prototype-quantity)

### Misreadings and corrections (editorial examples)

Possible misreading: Intuition is wrong and deliberation is correct.

Correction: Examine the roles of the two modes and the conditions under which correction succeeds.

Check the passage: [Section 1: intuition and mental accessibility](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)

Possible misreading: A strong impression means the target question has been answered correctly.

Correction: Identify the attribute being judged and distinguish the strength of an impression from its relevance to the question.

Check the passage: [Section 4: substituting a judgment attribute](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)

Possible misreading: One experimental example proves that the same judgment occurs in every setting.

Correction: Check participants, wording and comparison design, and state conditions for testing a new setting.

Check the passage: [Conclusion: an agenda for research on intuition](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)

<a id="guide-capstone"></a>
### Capstone

Choose an everyday judgment. Create two informationally equivalent presentations and identify the target attribute, reference point, potentially neglected extent and a corrective cue. Record how the setting differs from the essay’s experiments.

- Are changes in presentation distinguished from changes in information?
- Does the analysis connect intuitive processing, valuation and conditions for correction?
- Is a way of testing the cue provided without claiming its effect has already been measured?

Parts to connect: [Conditions producing an intuitive answer](#guide-part-accessibility), [Examining what is valued and judged](#guide-part-valuation-judgment), [Conditions of correction and limits of inquiry](#guide-part-correction)

<a id="guide-scope"></a>
### Native scope inventory

1. [Opening: the questions studied together](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
2. [Section 1: intuition and mental accessibility](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
3. [Section 2: presentation and preferences](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
4. [Section 3: evaluating changes in choice](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
5. [Section 4: substituting a judgment attribute](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
6. [Section 5: thoughts that can correct a judgment](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
7. [Section 6: prototypes and quantities](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)
8. [Conclusion: an agenda for research on intuition](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)

### Primary references

- [Daniel Kahneman — Nobel Lecture essay](https://www.nobelprize.org/uploads/2018/06/kahnemann-lecture.pdf)


[Return to the learning courses](/en/lab/classics/learn)

[Download Markdown](/classics-data/modern/kahneman.en.md)
