Few concepts have travelled as far in leadership thinking as the idea of mental models. The term appears in leadership training decks, coaching conversations, transformation programmes, even performance reviews. We are told that our mental models shape how we see the world, that old ones hold us back, and that better ones will help us lead. The message is simple. If you can identify the model in your mind, you can adjust your behaviour. Shift the model, shift the outcome.

This narrative has a strong appeal. It promises that complexity can be managed by editing the contents of our heads. It turns ambiguity into a personal development task. Leaders do not need to stay close to the messy interplay of people, history, power and context. They need to update their mental software. It is reassuring. It offers the idea of depth while avoiding the discomfort of engagement.

Yet behind this appeal lies a problem. Mental models are not things. They are not discrete structures that can be discovered, labelled and improved. They have no clear boundaries, no stable form, no agreed definition. They are a convenient metaphor that has been mistaken for a mechanism. And when leaders treat metaphors as mechanisms, the result is distortion. The world becomes smaller than it is.

This article does not deny that humans interpret experience. It challenges the assumption that those interpretations exist as coherent internal models. The trouble is not that mental models simplify reality. All tools simplify. The problem is that the mental model story encourages a particular posture. It tells leaders they can understand their organisation primarily by understanding themselves. It asserts that clarity begins within their own minds. And it quietly invites them to see complexity as something that can be tidied.

What neuroscience says about mental models

If leadership theory treats mental models as internal structures, neuroscience points in a very different direction. The brain does not store neatly defined templates that govern behaviour. It does not house beliefs and assumptions as stable objects awaiting discovery. Instead, cognition is constructed moment by moment. The brain works less like a library of ideas and more like a prediction engine that continuously generates hypotheses about what might be happening, based on memory, sensory cues and the demands of the context.

These predictions are provisional. They update rapidly. They do not solidify into anything resembling a fixed internal model. They are temporary acts of sensemaking, assembled and dissolved in fractions of a second. This undermines the idea that leaders can locate and adjust a mental model as if it were a faulty component in a machine. There is no such component. There is only a fluid and adaptive process responding to what the system presents.

Cognitive science strengthens this point. Andy Clark’s Extended Mind thesis argues that thinking is not confined to the head at all. Our environment, tools, conversations and social rhythms function as active parts of cognition. We think with our surroundings as much as with our neurons. Ed Hutchins’s work on distributed cognition takes this further. By studying real teams in action, he showed that what appears to be reasoning is a process distributed across people, artefacts, and individuals, often involving collective interactions. The intelligence of the system does not reside inside any single person. It is enacted between them.

Kahneman adds another layer. His research shows that the mind constructs explanations after the fact. Much of what we describe as a mental model is an interpretive story created to justify behaviour that emerged from context, habit and interaction. The narrative feels coherent, but coherence is a feature of the story, not of the underlying process. The model is not the cause. It is the retrospective account.

Taken together, these insights reveal a critical flaw in the mental model metaphor. Human cognition is not a set of internal templates. It is a dynamic, relational, situated activity. Our thinking is shaped by the tools we use, the people we engage with, the pressures we face and the histories we inherit. The mind is not a container of content. It is a process embedded in a system.

This matters for leadership. If thinking is distributed and emergent, then the idea of upgrading one’s mental models becomes largely symbolic. Reflection still matters, but it cannot do the heavy lifting we often attribute to it. What shifts behaviour is not the refinement of internal assumptions but the restructuring of relationships, constraints and experiences. Insight is helpful, but interaction is what changes things.

Neuroscience does not deny interpretation. It simply shows that interpretation is fluid, collective and context-bound. Which means the work of leadership is not to edit the contents of the mind, but to participate differently in the systems that shape thought in the first place.

How mental models distort leadership practice

The idea of mental models would be harmless if it simply offered language for reflection. The trouble begins when leaders treat these metaphors as tools for action. Once mental models are assumed to be the root of behaviour, leaders begin seeking personal adjustments when the system requires relational, structural, or contextual shifts. The focus moves inward, even when the work lies elsewhere.

This distortion shows up in several predictable ways:

First, mental models individualise systemic problems. When performance stalls, leaders are encouraged to ask which assumptions are getting in the way. When collaboration breaks down, the default explanation becomes a clash of internal models. This reframes collective dynamics as private shortcomings. It sends people inward when the issue may lie in workload, clarity, incentives or power. The system disappears behind the story of mindset.

Second, mental models oversimplify the factors that drive behaviour. Leaders begin to imagine that actions flow neatly from beliefs, as though employees act from internal logic rather than from constraints, expectations and risks. Yet most organisational behaviour is pragmatic. People do what feels safe, what is rewarded, what is tolerated and what appears consistent with how others behave. Treating behaviour as a direct expression of a mental model neglects all the forces that actually shape it. [See the article on the Iceberg Model]

Third, the mental model lens breeds a subtle moralism. Once beliefs become the presumed source of problems, people are judged for thinking incorrectly. Teams acquire labels. Leaders talk about colleagues who have an old model, a fixed model, a negative model. These labels harden quickly. They become shorthand for dismissal rather than invitations to understanding. A complex person gets reduced to a cognitive caricature.

Fourth, the metaphor inflates the role of self-insight. Leaders begin to imagine that greater introspection will unlock better decisions. They spend time naming, mapping and reframing their own assumptions. Reflection is valuable, but not when it becomes a substitute for engagement. The most significant leadership shifts often come not from inner clarity but from altered relationships, new constraints, redistributed authority or different rhythms of work. Insight without interaction rarely moves a system.

Finally, mental models distort the nature of change. If leaders believe that behaviour flows from internal models, then change becomes a matter of diagnosis and correction. Find the faulty assumption, replace it, and improvement will follow. This is an attractive storyline, yet it bears little resemblance to how complex systems evolve. Change emerges from what people try together, not from what they believe alone. It begins at the edges, not at the centre of someone’s mind.

The danger is not that leaders think. It is that they treat thinking as the primary lever of change. When mental models become the main lens of leadership, attention shifts away from patterns of interaction, from conditions that shape behaviour, and from the lived experience of the organisation. Leaders end up polishing the mirror rather than stepping into the room.

In this way, the myth of mental models does not expand leadership capability. It narrows it. It encourages self-examination when what is required is shared inquiry. It privileges the inner landscape over the organisational terrain. It promises a false control.

A more useful shift: from introspection to interaction

If mental models turn leaders inward, complexity asks them to reorient outward. The work is less about refining internal assumptions and more about changing how we participate in the system. The shift is not from thinking to action. It is from isolated reflection to collective sensemaking, from self-improvement to shared movement. What follows are five practices that support this shift in a practical, grounded way.

Look for signals, not stories: Leaders often search for the narrative that explains what is happening, yet narratives tend to tidy what the system is still unfolding. A more useful approach is to notice the signals that are already visible. Pay attention to small changes in behaviour, unexpected interests, tensions that keep resurfacing or invitations that appear at the margins. These signals reveal where the system is leaning before people can articulate why. They offer direction without demanding a diagnosis. Sensemaking begins with signals, not conclusions. Stories matter, but they come later, through conversation. When leaders rush to explanation, they close down the very inquiry that allows shared meaning to emerge.

Expand who is involved, not who is analysed: Instead of examining the mental models of individuals, widen the circle of participation. Bring together people who do not usually share space. Include perspectives that feel inconvenient or peripheral. Complex systems shift when new combinations of people interact, not when leaders refine their private assumptions. The quality of the conversation often matters more than the clarity of the insight. Inclusion itself becomes an intervention.

Shape the conditions, not the cognition: When behaviour stagnates, the temptation is to interpret it as a mindset problem. But patterns often persist because the conditions that surround them have not changed. Rather than focusing on beliefs, examine the constraints, permissions, incentives and risks at play. Small adjustments to expectations, decision rights or rhythms of work can shift behaviour more effectively than asking people to reconsider their assumptions. Change the conditions, and the thinking tends to follow.

Test in practice, not in theory: Complexity does not reward perfectly articulated plans. It responds to experiments that generate real feedback. Create small, low-risk tests designed to learn what the system is ready for. Try a new meeting structure in one team. Adjust a workflow in a single project. Invite a different stakeholder group into a routine conversation. Each experiment sends a signal. The response reveals more than introspection ever can. The goal is not to be right, but to be in motion.

Strengthen relationships, not explanations: Many leadership challenges attributed to mental models are, at their core, relational. Conflicts persist because trust is thin. Coordination falters because expectations are unclear. Innovation stalls because risk is unevenly shared. When leaders invest in relationships, they create the conditions for different behaviour to emerge naturally. This is not soft work. Strong relationships increase psychological safety, broaden perspective and create the connective tissue through which change travels.

Letting go of the mental model myth

Mental models are not the enemy. They have helped many leaders name their assumptions and reflect on how they interpret the world. The difficulty lies in how literally the metaphor has been taken. It promises that behaviour can be traced to inner templates, that clarity sits inside the mind, that change begins with correcting our private beliefs. It turns the organisation into a psychological puzzle that can be solved with enough introspection.

Complex systems do not work that way. They move through interaction, not intention. They respond to conditions, not cognition. They are shaped by relationships, constraints, timing and the many small adjustments people make as they respond to one another. Belief plays a part, but it is rarely the starting point. What shifts a system is not the model in someone’s head, but the pattern that emerges between people.

Letting go of the mental model myth is not a rejection of reflection. It is an invitation to place it in its proper context. Reflection becomes one move among many, not the central mechanism of change. Clarity becomes something created with others, not something excavated from within. Leadership becomes less about discovering what we think and more about noticing how we participate.

This shift asks leaders to stand less as internal archaeologists and more as active contributors to the living system around them. It asks them to work with what is visible, not what is imagined. To shape conditions rather than correct cognition. To move closer to the patterns they hope to shift. To see complexity not as something that can be tidied, but as something that can be engaged.

We do not need better mental models. We need better conversations, better relationships, better experiments and better attention to what is emerging. Depth does not live inside the mind. It lives in the spaces between us.

Questions for reflection

Where are you treating mental models as causes, when they may only be stories created after the fact?

What organisational dynamics are being labelled as mindset issues rather than examined as patterns of interaction?

What would change if you stopped trying to update mental models and focused instead on changing how people engage with one another?

Sources:

Clark, A. (2008) Supersizing the mind: Embodiment, action, and cognitive extension. Oxford: Oxford University Press.

Hutchins, E. (1995) Cognition in the wild. Cambridge, MA: MIT Press.

Kahneman, D. (2011) Thinking, fast and slow. London: Allen Lane.