What human error actually is

Human error is too often used as the final line in an incident report. It should be the beginning of the analysis.

A practical human factors view separates human failure into four broad types:

• Slips: attentional errors where the person knows what to do, but does the wrong action, such as selecting the wrong valve, pressing the wrong control or stepping into a line of fire while focused elsewhere.
• Lapses: memory errors where a person forgets a step, loses track of a sequence or misses a check.
• Mistakes: judgement or planning errors where the person believes they are doing the right thing, but the rule, assumption or mental model they are using is wrong.
• Violations: deliberate deviations from a rule, procedure or expected way of working.

This distinction is important because each category needs a different control. A violation may require better supervision, accountability, consultation or redesign of unrealistic work expectations. A mistake may require clearer rules, better decision support or improved training. A slip or lapse requires a different response again, because the worker may already be competent, trained and motivated.

Most safety training still leans heavily towards preventing violations: follow the procedure, comply with the rule, stop the job, escalate the issue. Those messages matter, but they do not control the most common day-to-day reliability problem: the unintentional failure of attention.

Slips and lapses are particularly important in high-risk industries because they occur during familiar work. They are not limited to new workers. They often happen to experienced people doing routine tasks in environments they know well. The HSE makes the same point: errors can occur regardless of how well trained and motivated people are, and training alone will not remove slips and lapses.

That means the problem is not simply knowledge. A person can know the procedure and still fail to execute it if attention is elsewhere, memory is overloaded or fatigue has narrowed the field of awareness.

The three conditions that produce human error

Human error in mining and other high-risk industries is usually produced by three overlapping conditions: attentional drift, cognitive overload and fatigue. These are not soft issues. They are operational risk factors.

1. Attentional drift: the mind is elsewhere when the risk occurs

Attentional drift occurs when the person is physically in the task, but mentally not fully with the risk. The worker may be present, trained and experienced, but their attention has shifted to a production pressure, a conversation, a personal concern, a radio call, a familiar routine or the next job in the sequence.

This is not laziness. It is how attention works. Attention is selective and limited. The brain prioritises some signals and filters out others. In repetitive high-exposure environments, workers can also become less attentive to hazards they see every day. Anderson and colleagues (2024) describe how repeated exposure to a hazard, particularly when it does not produce an immediate negative consequence, can reduce attention to that hazard over time.

In practice, attentional drift shows up as a worker looking at a task but not registering that the conditions have changed; a supervisor hearing a risk cue but not recognising its importance; or an operator scanning a screen while missing a weak signal.

This is why experienced workers can be vulnerable. Familiarity creates efficiency, but it can also create blindness. In high-risk work, the hazard often appears briefly. If attention is elsewhere in that moment, the control may exist on paper but not in practice.

2. Cognitive overload: too many demands competing for limited capacity

Cognitive overload occurs when the demands of the task exceed the person’s available mental bandwidth. This is common in high-risk operations because workers and leaders are often monitoring changing conditions, processing information, coordinating with others, managing time pressure, following procedures and responding to interruptions at the same time.

SafeWork NSW (2024) identifies cognitive task demands such as attention, perception, memory, active listening and decision-making as important parts of psychosocial task analysis. The same guidance recommends examining how workers solve problems, how they notice abnormal or dangerous situations, and why hazards might be missed.

In mining and other high-risk environments, cognitive overload can be produced by radio traffic, alarms, conflicting priorities, unclear work packs, contractor interfaces, shift handovers, production pressure, poor system design or complex permits used in noisy and time-pressured settings.

When cognitive load rises, people do not simply “try harder” and perform normally. They narrow. They simplify. They default to habit. They become more likely to miss peripheral cues, skip checks, misread information or apply the wrong rule. Cognitive overload should therefore be treated as a safety hazard, not a personal weakness.

3. Fatigue: the flashlight narrows and the floodlight dims

Fatigue is often managed as a roster issue. It is more than that. Fatigue depletes the executive functions that make safe work possible: attention control, working memory, judgement, inhibition and flexible decision-making.

A useful way to explain this is that fatigue narrows the mental flashlight and dims the broader floodlight. The tired worker may still focus on the immediate task, but the broader field of situational awareness becomes weaker.

NIOSH describes work-related fatigue as commonly associated with non-standard schedules that disrupt or shorten sleep, including night shift work and extended hours (National Institute for Occupational Safety and Health [NIOSH], n.d.). OSHA similarly notes that long hours and irregular shifts may increase the risk of injuries and accidents (Occupational Safety and Health Administration [OSHA], n.d.).

In operational terms, fatigue means the pre-start is heard but not absorbed. The critical control is checked but not challenged. A change in conditions is seen but not escalated. A permit step is completed automatically rather than consciously.

Fatigue management should therefore be understood as human reliability management. It is not simply about whether someone can stay awake. It is about whether they still have enough cognitive capacity to notice, decide, inhibit, remember and recover.

Why procedures do not prevent human error

Procedures are essential. They define the expected method, make risk controls visible and support consistency across teams. But procedures do not execute themselves.

A procedure assumes that the person applying it is cognitively present, adequately resourced and able to notice when the situation has changed. In real operations, that assumption is not always safe.

There is a gap between knowing a procedure and being mentally available to apply it under pressure. This is where many serious incidents live.

SafeWork NSW (2024) makes a similar systems point in its work design guidance: many psychosocial and physical hazards arise from the same underlying work system issues, and effective work design is less reliant on human behaviour than lower-order controls. The SafeWork NSW Code of Practice also states that role overload can include excessive time pressure, cognitively difficult work, frequent task switching and safety-critical tasks where decisions may have serious consequences (SafeWork NSW, 2021).

This is why the phrase “they did not follow the procedure” is rarely a sufficient finding. It may be factually correct and still analytically shallow.

A better investigation asks:

• Was the procedure usable under the conditions?
• Was the worker cognitively available to apply it?
• Were workload, time pressure and fatigue controlled?
• Were weak signals easy to detect?
• Was the system tolerant of a foreseeable slip or lapse?
• Were leaders and workers consulted early enough to design the work properly?

Risk management literature makes the same broader point: risks are easier and cheaper to influence when they are identified early in the design stage rather than late in execution. Miladi Rad and Yamini (2017) argue that starting risk management late creates limitations, increases costs and reduces the effectiveness of available responses. That principle applies directly to human error. Once the roster, task design, staffing model, equipment interface and production plan are already set, the remaining control often becomes “be careful”. That is not enough.

What the evidence says works

The answer is not to choose between systems and people. The answer is to manage both.

Work design remains the first-order control. Organisations should remove unnecessary complexity, reduce competing demands, make critical information easier to detect, improve handovers, build recovery into rosters, and involve workers in designing how work is actually done. SafeWork NSW (2024) is clear that work design should address underlying root causes, not just visible symptoms.

However, there is also a strong case for training attention directly. In high-risk industries, attentional training should not be dismissed as a wellbeing add-on. It can be framed as a human performance control because it strengthens the capability that many safety-critical controls depend on: the ability to notice the moment before the mistake becomes the event.

The research base is relevant here. Jha and colleagues (2017) examined short-form mindfulness training and relaxation training with Division I college football players during a high-demand pre-season interval. Both groups received structured training, but greater practice and adherence in the mindfulness group was linked with more stable attention and fewer attentional lapses. Related work in military cohorts found that mindfulness practice may promote cognitive resilience during high-stress periods by reducing attentional lapses (Jha et al., 2016).

The practical lesson is not that a short daily practice is magic. The lesson is that attention can be trained, and that repeated practice appears to matter. A realistic threshold, such as 12 minutes of daily attention practice, gives teams a practical entry point. It is short enough to fit into operational life and long enough to build a recognisable habit of noticing, resetting and returning to the task.

For high-risk industries, this matters because risk often appears in transition moments: a restart, a handover, the final hour of a night shift, a changed isolation point, a delayed work pack, a contractor interface, a call from the control room, or the moment after a stressful interaction.

This is where attentional training enters the risk control picture — not as a wellness add-on, but as a direct support to human reliability. Training that builds the capacity to notice drift, recover focus and re-engage situational awareness gives workers a practical capability for the moments that matter most: the restart, the handover, the final hour of a night shift, the task that has been done a hundred times before.

That kind of training works best when it is short, repeatable and embedded in operational routine — close enough to the work that the skill transfers. And it should sit inside a broader control framework, not replace the engineering controls, workforce planning, fatigue management and task redesign that address risk at the source.

Practical takeaway: what HSE managers can do now

1. Audit attention as a hazard

Do not only audit whether the procedure exists. Audit where the task depends on attention being available at exactly the right moment.

Start with safety-critical work and ask: where does this task rely on a person noticing a weak signal? Where is the work repetitive enough to create hazard familiarity? Where are workers switching rapidly between tasks? Where are the visual, radio, alarm or paperwork demands competing? Where would a slip or lapse create serious harm?

SafeWork NSW (2024) recommends task analysis that considers cognitive demands, safety-critical decisions, dependencies, workarounds and non-compliance with procedures. For HSE managers, this creates a practical pathway: treat attention as a finite control. Design for it, verify it and do not assume it is always available.

2. Train cognitive recovery

Workers need more than hazard awareness. They need practical recovery skills for the moment attention has narrowed, drifted or overloaded.

These skills can be embedded into pre-starts, field leadership, critical control verification and supervisor routines. Examples include a 30-second attention reset before critical work, a stop-breathe-scan routine after interruption, a deliberate “what has changed?” prompt before restarting, or a peer check that includes attention state rather than only task status.

The goal is not relaxation for its own sake. The goal is operational readiness: notice the drift, recover the mind and re-enter the task with clearer situational awareness.

3. Design rosters and work for attentional depletion

Fatigue controls should not stop at hours worked. They should also account for the type of attention required by the task.

A person performing night-shift driving, complex isolation, repetitive monitoring, emergency response, control-room supervision or high-consequence maintenance is carrying cognitive load, not just physical load. SafeWork NSW (2024) recommends scheduling complex tasks, where possible, for times when workers are less likely to be fatigued and time pressured.

For operations leaders, this means reviewing consecutive night shifts, long commutes after extended shifts, critical tasks placed late in the roster cycle, high-risk work during circadian lows, and whether breaks genuinely restore attention. It also means making sure supervisors have the capacity to think, not just react.

The standard should be simple: do not design work that requires perfect attention from a depleted brain.