Perception, Cognition and Audience Analysis

Introduction

Leadership communication is not simply the transmission of information from one person to another. It is a process of interpretation. Leaders select facts, frame issues, infer motives, anticipate reactions and adapt messages to people who may have different experiences, interests and expectations. The same message can therefore produce confidence in one audience, resistance in another and confusion in a third.

Effective leaders understand that communication occurs at the intersection of three factors: what the leader intends to communicate, what the audience notices and how the audience interprets it. Perception and cognition influence all three. Perception concerns how people select and organise information. Cognition concerns mental processes such as attention, memory, reasoning, judgement and meaning-making. Audience analysis is the systematic study of the people affected by, or capable of influencing, a communication.

The purpose of audience analysis is not to manipulate people or tell different audiences whatever they want to hear. Its ethical purpose is to make accurate information understandable, relevant and usable while preserving honesty, consistency and respect. This chapter examines attribution, assumptions, bias, framing, mental models and stakeholder expectations, and then applies these concepts to practical audience analysis.

Learning Objectives

By the end of this chapter, you should be able to:

1. Explain how perception and cognition shape leadership communication.
2. Distinguish attribution, assumptions, bias, framing and mental models.
3. Identify how stakeholder expectations influence message reception.
4. Conduct a structured audience analysis.
5. Adapt language, evidence, emphasis and channel without compromising accuracy or integrity.
6. Recognise ethical risks associated with stereotyping, selective disclosure and manipulative framing.
7. Apply audience-analysis principles to a realistic leadership communication problem.

1. Communication as Interpretation

A useful starting point is to distinguish the message sent from the message received. A leader may intend to communicate, “The organisation is changing its reporting process to improve reliability.” Employees may instead hear, “Management does not trust us,” or “My role is at risk.” The words have not necessarily changed; the interpreted meaning has.

Interpretation is influenced by:

• prior experience and professional identity;
• current emotions, stress and workload;
• expectations about leadership;
• trust in the speaker and the organisation;
• cultural and linguistic background;
• perceived consequences of the message;
• the medium, timing and social setting; and
• the amount and quality of available evidence.

This does not mean that meaning is entirely subjective. Some interpretations are better supported by the message and evidence than others. Professional communication therefore requires both empathy for different interpretations and discipline in checking whether interpretations are justified.

A leader should ask three related questions:

1. What do I mean? This concerns purpose, content and intended action.
2. What might the audience notice? This concerns attention, relevance and salience.
3. What might the audience conclude? This concerns inference, attribution and expectations.

These questions prevent the common error of treating transmission as success. Communication is successful when the intended meaning is conveyed sufficiently for the audience to make an appropriate decision or take an appropriate action.

2. Perception and Cognition

2.1 Perception

Perception is the process through which people select, organise and interpret sensory and social information. It is selective because people cannot attend to every available stimulus. It is organised because people classify information into patterns. It is interpretive because people assign meaning using previous knowledge and expectations.

In organisational communication, perception is especially important when information is incomplete. A delayed email, an unexplained meeting or a change in wording may become a signal from which employees infer priorities or intentions. Leaders cannot control every interpretation, but they can reduce avoidable ambiguity by explaining context, decisions and consequences.

2.2 Cognition

Cognition includes attention, memory, learning, reasoning, judgement and problem-solving. Human cognition is capable but limited. People often use mental shortcuts, sometimes called heuristics, to make decisions under time pressure or uncertainty. These shortcuts can be useful, but they can also produce systematic errors.

For example, a person who recently experienced a failed project may give unusually high weight to risks in a new proposal. A manager who remembers a highly visible success may overestimate the reliability of a particular team. These judgements may contain some truth, but the evidence may be unrepresentative.

Leaders should design messages for realistic cognitive conditions. Important information should be structured, signposted and repeated appropriately. Complex decisions should distinguish facts, interpretations, uncertainties and required actions. This is not “simplifying” the audience; it is reducing unnecessary cognitive load.

2.3 Selective attention and salience

People tend to notice information that is relevant to their goals, emotionally striking or inconsistent with expectations. A finance director may focus on cost, a frontline employee on workload and a regulator on compliance. Each person may hear the same briefing through a different attentional filter.

A leader can improve comprehension by making relevant implications explicit. Instead of saying, “The new system will be implemented in July,” a leader might say, “For frontline staff, training begins in June and no employee will be expected to use the system before completing training.” The second formulation connects the general announcement to the audience’s likely concern.

3. Attribution: Explaining Behaviour

Attribution is the process of explaining the causes of behaviour or events. People commonly distinguish between:

• Internal attribution: explaining behaviour through ability, effort, personality or motivation.
• External attribution: explaining behaviour through circumstances, resources, incentives, workload or environmental constraints.

Suppose a team misses a deadline. A leader might conclude that the team lacks discipline, an internal attribution. Alternatively, the leader might investigate whether requirements changed, resources were unavailable or dependencies failed, an external attribution. Sound leadership requires evidence before assigning blame.

3.1 Common attribution errors

The fundamental attribution error is the tendency to overemphasise personal characteristics when explaining another person’s behaviour and to underemphasise situational factors. A supervisor may describe an employee as careless after one error without considering unclear instructions or excessive workload.

The self-serving bias occurs when people attribute success to internal qualities but explain failure through external circumstances. Leaders are not immune. A manager may claim credit for a successful project but blame staff or market conditions for an unsuccessful one.

The actor–observer difference describes a related tendency: people often explain their own behaviour through circumstances while explaining others’ behaviour through personal characteristics. A late manager may say that traffic caused the delay, while describing a late employee as unreliable.

3.2 Attribution and communication practice

Attribution affects language, fairness and action. A blame-oriented message may close inquiry: “The team failed because it was not committed.” An evidence-oriented message opens inquiry: “The target was missed. We need to examine decisions, resources, dependencies and execution before deciding what corrective action is required.”

A practical attribution protocol is:

1. Describe the observable behaviour without exaggeration.
2. Separate facts from interpretations.
3. Identify plausible internal and external explanations.
4. Seek information from people close to the situation.
5. Check whether the same standard is being applied to comparable cases.
6. Communicate responsibility proportionately and specify corrective action.

This approach does not eliminate accountability. It makes accountability more accurate and defensible.

4. Assumptions and Mental Models

4.1 Assumptions

An assumption is an idea accepted as true without sufficient verification. Assumptions are unavoidable because communication always relies on some background knowledge. The risk arises when assumptions remain invisible or are treated as facts.

Common leadership assumptions include:

• “Everyone understands the strategic rationale.”
• “Silence means agreement.”
• “The audience has the same technical knowledge as I do.”
• “People will support the change if the benefits are obvious.”
• “A single communication is sufficient.”

A disciplined communicator labels uncertainty and tests important assumptions. Questions such as “What evidence supports this belief?” and “What would we expect to observe if it were false?” help distinguish knowledge from inference.

4.2 Mental models

A mental model is an internal representation of how something works. It helps people predict outcomes and decide how to act. Employees may hold different mental models of a customer, a performance system, a leader’s role or the purpose of a policy.

For example, senior management may view a performance dashboard as a coaching tool, while employees may view it as a surveillance mechanism. The dashboard is the same artefact, but the surrounding mental models differ. Simply repeating the policy may not resolve the disagreement. The leader must explain purpose, safeguards, decision rules and actual use.

Mental models can be improved through:

• clear definitions and examples;
• explanation of cause-and-effect relationships;
• visual process maps or demonstrations;
• opportunities for questions and feedback;
• comparison between current and desired practice; and
• evidence from experience or reliable data.

Leaders should also recognise that their own mental model may be incomplete. Stakeholders with different roles often notice constraints that decision-makers cannot see.

5. Bias in Leadership Communication

Bias is a systematic tendency that influences judgement or behaviour. Bias may be conscious or unconscious, individual or institutional. It does not always result from ill intent, but it can produce unfair or inaccurate outcomes.

Relevant biases include:

Bias cannot be solved merely by telling people to “be objective.” More effective safeguards include using explicit criteria, seeking diverse perspectives, documenting decisions, inviting challenge and reviewing outcomes for unequal effects.

A leader should distinguish personalisation from stereotyping. Personalisation uses verified knowledge about a particular audience or person, such as their role, information needs and decision authority. Stereotyping assigns assumed characteristics to a group without sufficient evidence. Audience analysis should personalise responsibly, not reduce people to categories.

6. Framing: Selecting Emphasis Without Distorting Reality

Framing is the process of selecting and emphasising certain aspects of an issue so that audiences understand it in a particular context. Every message has a frame because no communication can include every fact. A frame may highlight opportunity, risk, fairness, quality, cost, urgency or learning.

For example, a restructuring can be framed as:

• a cost-reduction programme;
• an effort to improve service quality;
• a response to changing customer needs; or
• a threat to established roles.

These frames may draw attention to different legitimate aspects of the same event. Ethical framing requires that emphasis does not become deception. A leader must not hide material facts, use technically true language to create a false impression or exploit fear and confusion.

A responsible framing process is:

1. Define the decision or issue accurately.
2. Identify the legitimate interests of affected stakeholders.
3. Present the relevant benefits, costs, risks and uncertainties.
4. Explain why particular information is being emphasised.
5. Avoid loaded labels and unsupported predictions.
6. Invite questions and correct material misunderstandings.

Framing can adapt a message without changing its factual core. For example, the same accurate evidence may be presented to engineers with methodological detail and to customers with service implications. The difference is relevance and level of explanation, not truthfulness.

7. Stakeholder Expectations

A stakeholder is an individual or group that affects, is affected by, or perceives itself to be affected by an organisation’s decisions. Stakeholders may include employees, customers, suppliers, owners, regulators, communities, professional bodies and partners.

Stakeholder expectations are beliefs about what an organisation or leader should do, provide or explain. Expectations may concern:

• outcomes, such as quality, safety or financial performance;
• process, such as consultation and fairness;
• relationships, such as respect and responsiveness;
• information, such as transparency and timeliness; and
• values, such as sustainability, inclusion or integrity.

Expectations are not always compatible. Customers may want low prices and extensive service. Employees may want stability and rapid innovation. Regulators may require detailed evidence while customers prefer simple explanations. Leadership communication involves making trade-offs visible rather than pretending that all interests can be maximised simultaneously.

A stakeholder map can classify stakeholders according to influence and level of impact. However, influence should not be the only criterion. Groups with limited formal power may still possess important knowledge or bear substantial consequences. Ethical audience analysis therefore considers vulnerability, rights and voice as well as authority.

8. Conducting an Audience Analysis

Audience analysis is a structured process for understanding who needs to receive a message, what they need to know, how they may interpret it and what action is appropriate.

Step 1: Define the communication purpose

State the desired outcome in observable terms. “Inform staff” is weak because it does not specify what staff should understand or do. A stronger purpose is: “Enable supervisors to explain the new scheduling process, identify exceptions and direct employees to support resources.”

Step 2: Identify audiences and stakeholders

List primary audiences, secondary audiences, decision-makers, implementers, affected groups and potential critics. Consider who receives the message directly and who may encounter it through informal networks or public channels.

Step 3: Analyse information needs

For each audience, determine:

• What do they already know?
• What must they know to make a sound decision?
• What could they reasonably misunderstand?
• What evidence will they consider credible?
• What action or response is expected?

Step 4: Analyse interests, concerns and expectations

Identify likely benefits, costs, risks and emotional responses. Do not treat anticipated resistance as irrational. Resistance may signal a real burden, an unresolved question or a history of broken commitments.

Step 5: Analyse authority and influence

Determine whether the audience decides, recommends, implements, approves, complies with or is affected by the issue. This helps determine the level of detail and the appropriate channel.

Step 6: Select message design and channel

Adapt vocabulary, examples, evidence, length, sequence and medium. Sensitive issues may require direct conversation before written confirmation. Operational instructions may require accessible written procedures. Complex change may require repeated communication across several channels.

Step 7: Test and revise

Use pilot groups, comprehension checks, feedback sessions or questions from trusted representatives. A useful test is to ask recipients to explain the message in their own words and identify what they will do next.

Audience analysis matrix

This matrix is a planning tool, not a substitute for engagement. Actual audiences should be consulted where consequences are significant.

9. Adapting Messages Without Compromising Integrity

Adaptation is legitimate when it changes the presentation while preserving the underlying truth. The following dimensions can be adapted:

• Vocabulary: technical terms for specialists; plain language for general audiences.
• Depth: detailed evidence for decision-makers; essential implications for busy users.
• Examples: role-relevant examples that illuminate the same principle.
• Sequence: lead with strategic rationale for executives; lead with practical effects for employees.
• Channel: meeting, briefing paper, email, dashboard, video or one-to-one conversation.
• Support: glossary, frequently asked questions, training or contact points.

The following should not be altered merely to gain approval:

• material facts;
• known risks and limitations;
• eligibility or compliance requirements;
• uncertainty that could affect decisions;
• evidence of negative consequences; and
• commitments made to other stakeholders.

A message is ethically weak when it uses omission, euphemism or selective statistics to produce a conclusion the full evidence would not support. Leaders should distinguish confidentiality from concealment. Information may legitimately be withheld for privacy, security or legal reasons, but the reason and limits of withholding should be handled transparently where possible.

10. A Practical Message-Design Framework

A robust leadership message can be developed through the following sequence:

1. Purpose: Why is this communication necessary now?
2. Audience relevance: What does this audience need to understand or decide?
3. Core claim: What is the central, accurate message?
4. Evidence: What facts, examples or sources support it?
5. Implications: What changes for this audience?
6. Uncertainty: What is not yet known or guaranteed?
7. Action: What should recipients do, by when and with what support?
8. Feedback: How can recipients ask questions or challenge the interpretation?

A useful quality check is the “same facts, different access” test: if different audiences compare the communications, they should find different levels of detail and emphasis, but no contradiction on material facts.

11. Worked Application: Communicating a New Service-Quality System

Consider an illustrative case. A medium-sized health-support organisation plans to introduce a digital service-quality system. The system will record response times, unresolved cases and customer feedback. Senior leaders expect improved reliability. Employees worry that the system will be used primarily to rank individuals. Customers want faster service and assurance that personal information will be protected.

Background and problem

The initial management announcement states: “The new dashboard will increase accountability and enable performance management.” After the announcement, employees express anxiety, supervisors report rumours and a staff representative claims that consultation has been inadequate.

The problem is not only wording. The announcement activates competing mental models. Management sees accountability as learning and improvement. Employees interpret it as surveillance and punishment. The word “performance” is ambiguous and may imply individual sanctions.

Analysis

The key audiences are:

Attribution errors are also present. Some leaders attribute employee concern to resistance to change. An alternative explanation is that the organisation has not specified safeguards, appeal procedures or the distinction between team-level learning and individual evaluation.

Approach

The leadership team revises the communication without changing the underlying decision. It makes five commitments:

1. The dashboard will initially be used for service improvement, not automatic disciplinary action.
2. Measures will be interpreted with context, including case complexity and staffing conditions.
3. Employees will receive training and have a process to challenge inaccurate records.
4. Customer information will be restricted according to privacy requirements.
5. Results from a pilot will be reviewed before wider implementation.

The message is adapted for each audience. Employees receive a plain-language briefing, examples of data use and opportunities to question the process. Supervisors receive guidance on interpretation and escalation. Customers receive a concise explanation of service benefits and privacy safeguards. Senior leaders receive a risk analysis including implementation costs and indicators of trust.

Outcome

In the illustrative case, the pilot reveals that one response-time measure disadvantages teams handling complex cases. The organisation changes the measure before full implementation. Employee concerns do not disappear, but they become more specific and manageable. The revised system uses several indicators and requires contextual review.

Learning points

• Ambiguous language can activate negative mental models.
• Resistance may contain information about design weaknesses.
• Ethical framing includes relevant benefits, risks and safeguards.
• A single factual core can support several audience-specific presentations.
• Feedback is not merely a courtesy; it can improve the decision itself.

12. Challenges, Limitations and Ethical Governance

Audience analysis has limitations. Leaders may lack reliable information about stakeholder beliefs, audiences may contain internal differences and expectations may change rapidly. A map created at the start of a project can become obsolete. Analysis can also create a false sense of precision if categories are treated as fixed identities.

Important ethical risks include:

• Manipulation: using knowledge of fears or vulnerabilities to secure compliance rather than informed agreement;
• stereotyping: assuming that all members of a group think alike;
• paternalism: withholding information because leaders believe people cannot handle it;
• selective transparency: disclosing positive evidence while hiding material negatives;
• privacy intrusion: collecting personal information about audiences without justification; and
• token consultation: asking for views without a genuine possibility of influence.

Good governance requires proportionality, accountability and traceability. Leaders should document significant assumptions, identify evidence sources, record major stakeholder concerns and explain why feedback was accepted or rejected. Where decisions affect rights, employment, safety, privacy or access to services, communication should align with relevant organisational policies and legal obligations.

13. Technology and AI in Audience Analysis

Digital tools can support audience analysis by identifying frequently asked questions, analysing engagement patterns or segmenting communications by role. AI systems may help summarise feedback or generate alternative plain-language explanations. These tools do not replace human judgement.

Risks include biased training data, inaccurate summaries, privacy breaches, inappropriate profiling and overconfidence in engagement metrics. High click-through rates do not prove comprehension or informed agreement. AI-generated messages should be checked for factual accuracy, hidden assumptions, exclusionary language and inconsistent treatment of audiences. Personal data should be collected only for a legitimate purpose, protected appropriately and not used to infer sensitive characteristics without lawful and ethical justification.

A responsible practice is to use technology for assistance while retaining human accountability for decisions, disclosure, fairness and correction. Leaders should tell audiences when automated systems materially influence communication or evaluation, where transparency is relevant to trust and rights.

14. Best Practices for Leaders

Effective audience-centred communication is supported by the following practices:

1. Begin with purpose, not with a preferred channel.
2. Separate evidence, interpretation, prediction and opinion.
3. Check assumptions before designing the message.
4. Investigate situational explanations before attributing blame.
5. Use audience-relevant examples without stereotyping.
6. State what is known, unknown and conditional.
7. Explain material risks and trade-offs.
8. Adapt depth and language, but preserve factual consistency.
9. Use more than one channel for complex or consequential change.
10. Invite questions and demonstrate how feedback affects decisions.
11. Test comprehension rather than assuming that exposure equals understanding.
12. Review messages for privacy, fairness, accessibility and unintended effects.

Chapter Summary

Leadership communication is an interpretive process shaped by perception, cognition, experience and expectation. Attribution influences how leaders explain behaviour; unchecked attribution errors can produce unfair blame. Assumptions fill gaps in knowledge but should be identified and tested. Mental models shape how stakeholders understand systems, policies and motives. Bias can distort attention, evidence evaluation and judgement, while framing determines which legitimate aspects of an issue receive emphasis.

Stakeholder expectations concern outcomes, process, relationships, information and values. Audience analysis identifies who needs to know what, why the information matters, how it may be interpreted and what action is required. A structured process includes defining purpose, identifying stakeholders, analysing knowledge and concerns, assessing influence, selecting channels, testing comprehension and revising the message.

Ethical adaptation changes vocabulary, detail, examples, sequence and channel without changing material facts or concealing relevant risks. Strong leaders combine accuracy with empathy, transparency, contextual understanding and openness to challenge. Technology and AI can assist analysis, but human leaders remain responsible for privacy, fairness, truthfulness and accountability.

Key Terms

Attribution: The process of explaining the causes of behaviour or events.

Actor–observer difference: The tendency to explain one’s own behaviour through circumstances and others’ behaviour through personal characteristics.

Assumption: An idea accepted as true without adequate verification.

Audience analysis: A systematic study of an audience’s knowledge, interests, expectations, influence and information needs.

Availability bias: Giving excessive weight to information that is vivid or easily recalled.

Bias: A systematic tendency that influences judgement or behaviour.

Confirmation bias: The tendency to favour information that supports existing beliefs.

Cognition: Mental processes including attention, memory, reasoning, judgement and problem-solving.

External attribution: Explaining behaviour through situational or environmental factors.

Framing: Selecting and emphasising aspects of an issue to shape interpretation.

Fundamental attribution error: Overemphasising personal factors and underemphasising situational factors when explaining others’ behaviour.

Halo effect: Allowing one positive characteristic to influence an overall judgement.

Internal attribution: Explaining behaviour through personal characteristics such as ability, effort or motivation.

Mental model: An internal representation of how a system, situation or relationship works.

Perception: The selection, organisation and interpretation of information.

Self-serving bias: Attributing success to internal factors and failure to external factors.

Stakeholder: A person or group that affects, is affected by, or perceives itself to be affected by an organisation’s decisions.

Stakeholder expectation: A belief about what an organisation or leader should provide, explain or do.

Review Questions

5 Short Answer Questions

1. What is the difference between perception and cognition?
2. Distinguish internal attribution from external attribution.
3. Why can assumptions create communication problems?
4. What is meant by ethical framing?
5. List four elements that may be examined in audience analysis.

5 Long Answer Questions

1. Explain how attribution errors can affect fairness and accountability in leadership communication. Use an organisational example.
2. Discuss how mental models influence stakeholder responses to organisational change.
3. Explain the relationship between bias and audience analysis, and describe practical safeguards against biased judgement.
4. Describe how a leader can adapt a message for different audiences without compromising accuracy or integrity.
5. Evaluate the ethical risks of selective transparency, stereotyping and manipulative framing.

3 Analytical / Application Questions

1. A manager announces a new attendance-monitoring system by saying, “The system will identify unreliable employees.” Analyse the likely attribution, framing and mental-model problems. Redesign the communication ethically.
2. A university introduces a compulsory digital learning platform. Students, lecturers, administrators and data-protection officers have different concerns. Construct an audience-analysis matrix and recommend suitable communication approaches.
3. A project team misses a major deadline. Using the attribution protocol in this chapter, analyse the situation and propose a fair leadership response that preserves accountability while testing situational explanations.

Multiple Choice Questions with Answers

Question 1

Which statement best describes audience analysis?

A. Selecting the most persuasive emotional appeal for every audience

B. Systematically examining an audience’s knowledge, interests, expectations and information needs

C. Reducing a message to the shortest possible form

D. Using the same message for all stakeholders to ensure consistency

Correct Answer: B. Systematically examining an audience’s knowledge, interests, expectations and information needs

Question 2

A manager explains an employee’s missed deadline only by saying that the employee is careless, without examining workload or changing requirements. Which error is most evident?

A. Fundamental attribution error

B. Availability bias

C. Anchoring

D. Self-serving bias

Correct Answer: A. Fundamental attribution error

Question 3

Which practice is an example of ethical framing?

A. Hiding known implementation costs to increase support

B. Using fear to prevent stakeholders from asking questions

C. Emphasising relevant benefits while also disclosing material risks and uncertainty

D. Presenting different factual outcomes to different audiences

Correct Answer: C. Emphasising relevant benefits while also disclosing material risks and uncertainty

Question 4

What is a mental model?

A. A formal legal requirement for organisational communication

B. An internal representation of how a system or situation works

C. A numerical measure of audience engagement

D. A communication channel used by senior managers

Correct Answer: B. An internal representation of how a system or situation works

Question 5

Which action best tests whether an audience has understood a message?

A. Counting how many people received the email

B. Asking recipients to explain the message and intended next action in their own words

C. Increasing the number of technical terms

D. Sending the same announcement repeatedly without feedback

Correct Answer: B. Asking recipients to explain the message and intended next action in their own words

Question 6

A leader attributes a project’s success entirely to personal leadership skill but blames external conditions for failure. This illustrates which bias?

A. Self-serving bias

B. Halo effect

C. In-group bias

D. Availability bias

Correct Answer: A. Self-serving bias

Question 7

Which item should generally remain consistent when a message is adapted for different audiences?

A. The amount of technical detail

B. The sequence of information

C. The underlying material facts and relevant risks

D. The examples used

Correct Answer: C. The underlying material facts and relevant risks

Question 8

Why should stakeholder influence not be the only criterion used in audience analysis?

A. Low-influence groups may possess important knowledge or experience significant consequences

B. Influence is always impossible to estimate

C. Only external stakeholders should be analysed

D. High-influence stakeholders never need detailed information

Correct Answer: A. Low-influence groups may possess important knowledge or experience significant consequences

Question 9

Which statement best distinguishes personalisation from stereotyping?

A. Personalisation uses verified audience information; stereotyping assigns unsupported group characteristics

B. Personalisation changes facts; stereotyping preserves facts

C. Personalisation is always emotional; stereotyping is always rational

D. Personalisation is used only with customers; stereotyping is used only with employees

Correct Answer: A. Personalisation uses verified audience information; stereotyping assigns unsupported group characteristics

Question 10

Which is the most appropriate response when a dashboard measure disadvantages teams handling complex cases?

A. Keep the measure unchanged to demonstrate managerial authority

B. Remove all performance measures immediately

C. Review the measure, consider contextual factors and revise it if the evidence warrants

D. Blame the teams for failing to fit the measure

Correct Answer: C. Review the measure, consider contextual factors and revise it if the evidence warrants