How Michael Emery Thinks About Curiosity Before Incident Investigation

Episode 14 of the Headline Podcast, published on February 11, 2026, features Michael Emery discussing coaching, mentoring, questioning, listening, and the role of the safety professional after a deviation. His central argument is that curiosity makes an investigation more useful because it keeps the investigator close to how the work was understood before the event.

That is a demanding standard. The investigator must remain interested in the conditions that shaped the decision without allowing curiosity to become an excuse for weak accountability, vague language, or an endless search for context.

Why curiosity belongs at the start of an investigation

Curiosity belongs at the start of an incident investigation because the first explanation is often built from the most visible action. A worker crossed a boundary, a permit was incomplete, or a check was skipped. Those facts matter, but they do not explain what the person knew, what the task required, which control was available, or why the chosen path made sense in that moment. Curiosity opens those questions before the report closes them.

A narrow investigation asks who failed to follow the rule. A stronger investigation asks what the rule was expected to control, whether the work method supported that control, and how the supervisor responded when production pressure appeared. The difference is not softer language. It is better evidence.

James Reason's work on latent failures supports this distinction. The visible action is one layer in the event, while decisions about design, staffing, planning, maintenance, supervision, and escalation may sit behind it. Curiosity helps the team look across those layers without removing the person's responsibility for the action.

Headline's earlier discussion of evidence chains in safety investigations makes the same point from another direction. A conclusion is only as strong as the connection between the account, the record, the field condition, and the decision that follows.

What Episode 14 changes about the investigator's role

Michael Emery's view moves the safety professional away from the role of compliance policeman and toward a disciplined partner in understanding work. That does not mean agreeing with every decision or replacing a formal investigation with a conversation. It means using questions and listening to uncover the conditions that a checklist cannot see.

Michael states the coaching principle directly: “Coaching is essentially a questioning and listening exercise.” The phrase is useful in investigations because a question can reveal sequence, expectation, pressure, and uncertainty before the interviewee has learned which answer the investigator wants.

Curiosity also changes the investigator's posture. Instead of entering with a finished theory, the investigator carries a provisional theory that must survive contact with evidence. The person who made the decision remains part of the account, yet the investigation also examines the work system that made the decision possible.

This approach protects rigor when it is applied deliberately. The interviewer still records dates, locations, equipment states, documents, roles, and control requirements. What changes is the order of attention. Understanding comes before judgment, and judgment must be tied to evidence.

Which questions reveal the work behind the event?

The best investigation questions ask what the person saw, expected, feared, and could control before the event. They do not invite a polished opinion about safety. They reconstruct the decision in enough detail for another reviewer to compare the account with records, physical evidence, and the work method that was actually available.

Begin with the sequence. Ask the person to describe the task from the first preparation step to the moment the situation changed. Then ask what should have happened, what did happen, and when the two paths separated. This creates a timeline before the interview turns into a debate about fault.

Next, examine the pressure around the choice. Was the team waiting on equipment, a permit, a contractor, a restart, a delivery, or a supervisor's approval? Pressure does not excuse a dangerous act, although it can explain why a control became difficult to use or why an alternative appeared acceptable.

Finish by testing the control itself. Ask who owned it, how the worker knew it was effective, what happened when it failed, and whether anyone had authority to pause the task. A control that exists only in a procedure is different from a control that is available, understood, and checked at the point of work.

How questioning differs from blame-seeking

Blame-seeking looks for a sentence that ends the investigation. Questioning looks for a chain of conditions that explains the event and identifies what must change. The investigator can still identify a poor decision, a missed duty, or a deliberate violation, but the conclusion should show how that finding was established and what surrounding controls did or did not do.

Michael also warns against a relationship in which the safety professional becomes a finger-wagger. When the interview feels like a disciplinary performance, the witness has an incentive to shorten the account, omit uncertainty, and repeat the language that seems safest to management.

That is why the investigator should separate three statements. First, what the person did. Second, what the person believed was happening. Third, what the organization required and made possible. The statements may conflict, and the conflict is often where the useful evidence begins.

Responsibility remains visible in this model. If someone bypassed a known control for convenience, the report can say so. It should also state whether the control was practical, whether the supervisor tolerated similar decisions, and whether previous reports had changed the task. Accountability becomes more credible when it includes the conditions that leaders control.

Comparison: compliance policing or investigative curiosity?

The difference between compliance policing and investigative curiosity appears in the questions each approach asks, the evidence each approach keeps, and the action each approach produces. One approach closes the conversation around the rule. The other uses the rule as a reference point while examining the work conditions that determine whether the rule can operate.

The table does not suggest that retraining or discipline is never appropriate. It shows why neither response should be automatic. If the investigation finds a knowledge gap, training may be part of the action. If the investigation finds an impractical barrier, the work method must change. If the investigation finds intentional disregard after a clear and workable control, accountability must address that finding directly.

Headline's earlier analysis of missing root causes adds a useful test. If the action only tells the next worker to be more careful, the investigation probably stopped at the visible behavior.

How to keep curiosity from becoming investigation drift

Curiosity needs boundaries, or an investigation can collect stories without reaching a decision. The investigator should define the event, the exposure, the decision under review, and the evidence needed to test each explanation. A clear scope protects both fairness and speed.

Use four control questions throughout the review. What is known? What is assumed? What evidence would change the current explanation? Who will verify the proposed action? These questions prevent the team from treating a plausible narrative as a proven cause.

The interview should also have a close. Summarize what was heard, distinguish facts from interpretation, ask what the investigator may have missed, and explain what will happen next. Michael's second principle is important here: “You can be trained to be curious in a particular direction; it's got nothing to do with how you were born or brought up.” Curiosity is therefore a repeatable practice, not a personality lottery.

Within 7 days, the investigation lead should identify unresolved evidence and assign owners for the next checks. Within 14 days, the team should review whether the proposed control is being used. Within 30 days, leaders should test whether the same pressure still produces the same shortcut. Those dates are operating commitments, not claims about a universal investigation standard.

Recommendation

Use Michael Emery's curiosity discipline in the next investigation by delaying the first judgment long enough to reconstruct the task, the pressure, the control, and the information available to the person. Then convert what was learned into one control change, one accountable owner, and one verification date.

Before the first interview, write the decision under review in one sentence. During the interview, ask for the sequence without interrupting, then test the account against documents and field evidence. Before closing the report, record which explanation remains uncertain and what evidence would resolve it.

Do not measure the investigation by the number of pages, interviews, or corrective actions. Measure it by whether the organization changed the condition that made the unsafe decision attractive, difficult to challenge, or easy to repeat.

What Episode 14 changes for safety leaders

Episode 14 changes the standard for a good investigation from “Did we identify the person closest to the event?” to “Did we understand the decision well enough to change the conditions around it?” Curiosity is the method that keeps that question open long enough for evidence to matter.

Michael Emery's contribution is practical because it joins communication skill to operational accountability. Listening is not a soft alternative to control. It is how the investigator learns whether a control was understood, available, reinforced, and realistic under the pressure of the task.

Andreza Araujo's safety editorial approach keeps the same boundary. A learning culture does not erase responsibility, and accountability does not require investigators to stop asking how the system shaped the choice. The report becomes stronger when it can name both the action and the conditions that made the action possible.

Read the Headline companion on training, coaching, and mentoring, compare it with the investigation charter guide, and listen to the full conversation with Michael Emery in Episode 14.