Neurodiversity at Work: The Real Barrier Is the System

• Neurodivergent means a person thinks, learns and processes information in a way that differs from what most people consider typical.
• Common types include dyslexia, Attention Deficit Hyperactivity Disorder (ADHD), autism, dyspraxia, dyscalculia and Tourette’s syndrome.
• Under the Equality Act 2010, a neurodivergent condition may often be considered a disability where it has a substantial and long-term adverse effect on an individual’s ability to carry out normal day-to-day activities, but this is always assessed case by case.
• Employers must avoid unlawful discrimination and, where the legal definition applies, make reasonable adjustments to reduce workplace disadvantage.

Neurodiversity describes natural variation in human cognitive functioning across the population. Neurodivergent refers to people whose brain functioning differs from what is statistically typical, rather than being defined as a deficit (Vargas-Salas et al., 2025).

In practice, neurodivergent is often used as an umbrella term that can include:

• Autism
• ADHD
• Dyslexia (Vargas-Salas et al., 2025)

In workplace terms, neurodivergence can influence how someone:

• Sustains attention and shifts between tasks
• Processes sensory input such as noise, light, or interruptions
• Communicates and interprets information
• Handles information load and pressure (Vargas-Salas et al., 2025)

Research summaries in practitioner literature suggest that in some studies up to around 20 percent of people are described as neurodivergent, although estimates vary depending on definitions and methods. (Thompson and Miller, 2024). 

Neurodiversity has rarely been considered in the design of workplace processes, management practices, environments, or in how work is organised (Thompson and Miller, 2024).

Most organisations start with training and manager guidance. It is a visible step. It is also where progress often stalls.

CIPD notes that neurodiversity has rarely been considered in the design of workplace processes, management practices, environments, or in how work is organised. Typical norms may suit some people but not others (Thompson and Miller, 2024). If the work design stays typical, training becomes a layer over the same conditions.

This is how it fails:

• Inclusion is positioned as knowledge and attitudes, not as work design.
• Operating conditions stay the same when workload rises and work becomes more fragmented.
• Support becomes inconsistent between teams because the system does not build adjustments into normal ways of working.

In many roles, interruption is not an exception. It is the workflow. When focus is not designed as capacity, it survives only in quiet moments. Those moments disappear first when demand rises. (González and Mark, 2004)

Common consequences include:

• Planned work is displaced by incoming requests and “quick” asks.
• Complex tasks restart repeatedly, with time lost rebuilding context.
• Deep work is deferred, shortened, or pushed outside core hours.
• Quality control becomes harder because thinking happens in fragments.
• Some people pay a higher cost to re-enter focus after interruption, so the same environment creates an uneven load.

Many organisations miss the early failure because delivery continues. Throughput can stay stable while the effort required to sustain it increases. (Mark, Gudith and Klocke, 2008)

Common consequences include:

• Stress and time pressure rise without showing up in headline performance metrics.
• “Everything is fine” becomes the default story until errors or attrition appear.
• Rework increases quietly because tired teams rely on short-term fixes.
• Managers respond to visible delivery, not the invisible load behind it.
• Support arrives late because the system notices outcomes, not rising strain.

When priorities and decisions are distributed across email, chat, meetings, and documents, the organisation relies on individuals to integrate the picture. That integration work becomes a private burden rather than an owned process. (Kern et al., 2024; Nastjuk et al., 2022)

Common consequences include:

• People spend time reconstructing decisions, priorities, and intent across threads.
• The same updates are repeated in multiple channels to keep others aligned.
• Personal tracking systems replace shared operational visibility.
• Handoffs degrade because context is scattered, not bundled.
• Errors trigger more reminders and more messaging, which increases load further.

Task switching has reliable performance costs. It is not a motivation issue. Preparation helps, but it does not remove the cost. (Pashler, 2000; Monsell, 2003)

Common consequences include:

• Jobs assume continual switching, but performance expectations assume continuity.
• Switching tolerance becomes a proxy for competence and readiness.
• People who show strain are framed as disorganised or “not coping,” even when the design drives the strain.
• Support focuses on individual strategies while the switching demand stays fixed.
• The system repeatedly “solves” the person while preserving the same conditions.

Reasonable adjustments are a core mechanism for support, but access often depends on disclosure. That creates a predictable gap between support on paper and support in practice. (Acas, 2025)

Common consequences include:

• Help is available in principle, but it is not activated unless someone self-identifies.
• Disclosure becomes a repeated task across managers, projects, or reorganisations.
• People delay or limit disclosure when the social or career risk feels high.
• Support becomes uneven because the system serves those who can safely ask and keep asking.
• Work design barriers remain in place for those outside the adjustments pathway.

Awareness training can improve conversations. It does not redesign work. If interruption, overload, ambiguity, and visibility-based signals remain, the same barriers persist. (Thompson and Miller, 2024)

Common consequences include:

• Training raises intent, but does not change volume, pace, or channel load.
• Managers become more empathetic, but still rely on the same fragile workflows.
• The organisation treats improved awareness as a control, even when conditions drive failure.
• People are coached to adapt while the environment continues to demand constant switching.
• The system reassures itself instead of changing how work is planned, owned, and protected.

Good neuroinclusion is visible in everyday work conditions, not policy statements. It shows up in how attention is protected, how decisions are documented, and how support is triggered without relying on personal disclosure.

Interruptions do not just slow work down. They change the conditions people have to think in.

Research on interrupted knowledge work found that people often compensate by working faster, but at the cost of higher stress, frustration, time pressure, and effort (Mark, Gudith and Klocke, 2008).

In practice, protecting focus means:

• Triage rules for “quick asks” and walk-ups.
• Meeting-free windows for roles that require deep work.
• Escalation criteria, so interruptions are intentional rather than habitual.

This control fails when:

• Demand spikes and urgency bypasses triage.
• Focus time is treated as optional and is the first thing removed.
• Responsiveness is rewarded more than completion.

Digital work often creates information overload through volume, fragmentation, and constant switching between tools and channels (Arnold et al., 2023).

From a human factors perspective, communication systems are part of the work environment that shapes performance, especially under time pressure (HSE, 2024).

In practice, this means:

• One authoritative location for decisions and requirements.
• Defined channels by purpose (requests, decisions, approvals).
• Defaults that reduce non-essential notifications.

This control fails when:

• Decisions live in side chats and never make it into the system of record.
• Handoffs happen without bundled context.
• Multiple “latest versions” exist, so people revert to whatever is fastest.

People can be well-trained and highly motivated and still make predictable errors, especially when work is complex, rushed, or ambiguous (HSE, 2024).

If the organisation relies on people to infer priorities and “what good looks like” in real time, it is designing for drift.

In practice, operational clarity means:

• Visible work queues and explicit priority rules.
• Clear acceptance criteria (a “definition of done”).
• Agreed trade-offs when priorities conflict.

This control fails when:

• Ambiguity becomes socially enforced (“just figure it out”).
• Priorities are renegotiated informally and silently.
• Work is judged by visible busyness rather than completion against criteria.

Support should not depend on whether someone has a diagnosis or on how willing they are to disclose. Acas is explicit that employers should offer support whether or not a worker has a diagnosis and that a diagnosis is not required for someone to be considered disabled under the Equality Act 2010 (Acas, 2025).

In practice, “default adjustments” look like:

• Written follow-ups after verbal instructions.
• Agenda-first meetings and clear actions in writing.
• Predictable routines for task assignment.
• Access to quieter spaces or reduced sensory load where feasible.

This control fails when:

• Adjustments only appear after repeated self-advocacy.
• Implementation varies by the manager’s discretion.
• Support disappears during reorganisations, role changes, or peak workload periods.

Psychological safety is not a culture slogan. It is the condition that determines whether people will ask questions, admit uncertainty, and surface early risk. Edmondson defines it as a shared belief that the team is safe for interpersonal risk-taking and links it to learning behaviour in work teams (Edmondson, 1999).

In practice, this means managers:

• Normalise clarification as part of doing the job well.
• Invite pushback on workload and sequencing.
• Treat early signals as operational data, not attitude problems.

This control fails when:

• Performance management penalises questions or “slowness.”
• Leaders treat uncertainty as incompetence.
• People mask difficulty until it shows up as errors, burnout, or exit.

Output can stay high while effort and strain rise. That is one reason inclusion failures are often recognised late.

HSE guidance notes that fatigue is a decline in mental and/or physical performance and is linked to factors such as working time, sleep loss, and workload (HSE, 2025).

Interruption research shows the same pattern. Work may still get done, but the “cost” appears as time pressure and effort (Mark, Gudith and Klocke, 2008).

In practice, leading indicators include:

• Rework and repeated clarification loops.
• Missed handoffs and dropped actions.
• After-hours load and “work about work” growth.
• Frequent priority churn and constant escalation.

This control fails when:

• Leaders only track throughput and deadlines.
• Signals are treated as individual resilience problems.
• Strain remains invisible until sickness absence or performance issues appear.

Neurodivergence cannot be identified reliably from appearance or behaviour alone. There is no single set of traits that defines it. Some individuals have a formal diagnosis, while others do not.

Managers should not attempt to diagnose or label anyone. Instead, focus on where an employee may be experiencing difficulty at work and consider adjustments that reduce barriers and support effective performance.

Indicators that an employee may benefit from adjustments can include:

• Difficulty following complex or unclear instructions, particularly where information is provided only verbally.
• Difficulty organising tasks, prioritising work, or managing time (for example, missed steps or deadlines).
• Taking longer to read, write, or process written information, especially under time pressure.
• Sensitivity to noise, lighting, or busy environments may affect concentration.

Needs vary between individuals, and these indicators are not diagnostic. The most effective approach is to discuss what support is helpful and review whether adjustments are working.

Not all employees will disclose neurodivergence or have a formal diagnosis. Many people avoid disclosure because it creates extra work, repeated explanations, and perceived career risk.

Support is most reliable when it does not depend on disclosure. The aim is to reduce barriers through normal working practices and reasonable adjustments that make work clearer, calmer, and easier to control under pressure.

In practice, adjustment requests should be treated as system signals. They point to friction in how tasks are specified, how information moves, how the environment is controlled, and what performance is rewarded.

Noise, lighting, movement, and frequent interruptions can increase cognitive strain and make concentration unreliable. Sensory needs vary, so adjustments should be agreed case by case.

Where appropriate, adjustments may include:

• Access to a quieter workspace or quieter areas of the workplace.
• Noise-reducing equipment.
• Greater control over desk lighting, where feasible.

System signal: Requests for quieter space usually indicate that the environment is not controlling interruption, noise, or visual distraction as a normal condition of work. Concentration is being treated as personal resilience rather than a managed capability.

This fails when: Quiet arrangements depend on informal agreements that collapse during peak periods, hot-desking, shared coverage, or reactive work. The risk is underestimated because the workspace can look “fine” on a calm day, while the real disruption appears when demand rises.

Long, dense, unstructured communication increases the effort needed to extract meaning, identify actions, and track priorities. Some people may need more time to read and process information, especially under time pressure, and lengthy verbal instructions can be unreliable when the environment is busy.

Where appropriate:

• Provide instructions in more than one format (for example, a short written summary alongside verbal guidance).
• Use clear headings and bullet points.
• Confirm key actions, deadlines, and priorities in writing.

System signal: A need for clearer structure often exposes weak information design. Critical decisions, requirements, and “what good looks like” are not owned, not standardised, or scattered across channels.

This fails when: Summaries become “extra work” done by the employee or a conscientious manager rather than the standard way tasks are assigned. Under time pressure, teams revert to fast messages and verbal handoffs, and the clarity control disappears first.

Traditional interviews can disadvantage candidates when selection relies on social signalling, rapid improvisation, or ambiguous questioning. In those cases, capability relevant to the role may be better demonstrated through practical work.

Where appropriate:

• Use skills-based tasks, work samples, or structured interviews linked to role requirements.
• Provide questions in advance where possible and make evaluation criteria explicit.

System signal: If interviews reward confidence under ambiguity, the process is testing interview performance rather than job performance. That is a design choice, not a neutral standard.

This fails when: Panels revert to unstructured interviews due to time pressure, inconsistent training, or habit. The risk is underestimated because interviews still “feel” rigorous, even when they select for presentation rather than capability.

Standard working patterns may not suit every role or individual. Prolonged cognitive effort, sensory demands, and fatigue can reduce performance consistency over the day.

Where the role allows:

• Offer flexibility so complex tasks can be done during periods of highest concentration.
• Protect that time by scheduling routine meetings, admin, and low-cognitive-load work elsewhere.

System signal: A need for flexibility often indicates that the job is organised around meeting cadence and responsiveness norms, not around task load, recovery, and deep work requirements.

This fails when: Flexibility exists only as an informal agreement that disappears during workload spikes, staffing gaps, or management changes. It is underestimated because flexibility is viewed as a personal preference rather than a control that stabilises performance.

Some employees may find it difficult to decide where to start, maintain focus across multiple tasks, or manage changing priorities. This can increase the risk of missed steps, rework, and deadline drift.

Where appropriate:

• Break work into smaller tasks with defined outcomes and timeframes.
• Use a shared task board or project management system to make priorities visible.
• Run short, regular check-ins to confirm sequencing, trade-offs, and deadlines.

System signal: Difficulty “getting started” often reflects poor task definition, unclear priorities, or too many concurrent work packages. The system is requiring people to perform constant prioritisation privately.

This fails when: Priorities churn daily, task boards are not treated as the decision system, and escalations override planned work without explicit trade-offs. The risk is underestimated because the team still appears busy, while the hidden cost shows up as rework, dropped actions, and after-hours catch-up.

Neurodiversity awareness training builds the baseline competence needed to recognise neurodiversity, understand legal duties, and make informed decisions about adjustments. It provides a shared reference point so decisions are made on a consistent and credible basis, rather than through individual interpretation or assumption.

For those decisions to translate into meaningful work improvement, they must be supported by organisational systems that enable them to be applied, reviewed, and sustained. When training is used alongside clear work design, defined decision processes, and mechanisms for implementing and embedding change, it plays a valuable role in improving how organisations respond to barriers in everyday work.

The point is not to train people to cope better. It is to help the organisation see where work design is creating avoidable load, then fix the conditions that make support unreliable.