Valuing Human Cognition in The Time of AI

LISTEN ON SPOTIFY / ON APPLE / ON AMAZON

The case of Dr. P, chronicled in Oliver Sacks' "The Man Who Mistook His Wife for a Hat", provides a powerful metaphor. Dr. P, suffering from visual agnosia due to damage to his right hemisphere, could perceive details but not wholes - he saw features but could not recognise faces. As Sacks observed, "He approached faces - even of close relatives - as if they were abstract puzzles or tests. He did not relate to them, he did not behold them."

This neurological case study offers a striking parallel to the limitations of current generative AI systems. Both represent what happens when analytical, feature-detecting processes (associated with left-brain functions) operate without the integrative, contextual understanding typically provided by right-brain capabilities. This bulletin explores this parallel crisis and its implications for leadership in an increasingly data-driven world.

Mistaking data for reality

When data ceases to represent reality and begins to replace it with abstractions, we enter dangerous territory. Like Dr. P, who "had lost the capacity to see a face as a whole... He saw details, he saw features - but he could not see the whole", our AI systems can analyse components without comprehending the gestalt. This limitation becomes increasingly problematic as we delegate more of our perception, analysis, and decision-making to these systems.

Contrary to popular assumption, the more data there is, and the faster it moves, the more certainty is undermined. The hitherto unbreakable bond between evidence and fact starts to pull apart. This paradox manifests in AI systems that can generate convincing outputs without genuine understanding: content that appears meaningful but lacks authentic connection to truth.

Perhaps the most compelling illustration of why balanced human judgment remains critical comes from examining what can happen, when organisations over-rely on AI's left-brain strengths without compensating for its limitations.

Like Dr. P in Oliver Sacks' famous case study - who could see details but not recognize faces - organisations are in some danger of a form of 'corporate visual agnosia', processing unprecedented amounts of information, while losing the ability to recognise what matters most.

What sort of thing could happen?

• A healthcare system could implement AI diagnostic systems that improve detection rates for specific conditions, but miss holistic patterns in patient presentations that don't fit established categories - resulting in delayed diagnoses for complex cases that require integrative thinking.
• A financial services firm might rely on AI risk models that excellently analyse historical patterns, but fail to anticipate novel systemic risks, because the models lack the contextual awareness to recognise emerging patterns without precedent.
• A consumer products company could optimise marketing based on AI analytics, but miss profound shifts in consumer sentiment that weren't captured in their quantitative measures - changes that human observers with contextual understanding would have promptly spotted.

Each of these putative cases represents not a failure of technology, but a failure of leadership - specifically, the failure to recognise that AI amplifies our own left-brain bias toward measurable data, while potentially atrophying our right-brain capacity for holistic perception, contextual understanding, and value-based judgment.

Minding the gaps

This limitation in generative AI could significantly reshape how humans interact with reality, potentially creating some concerning shifts:

• Fragmented attention and understanding: As we increasingly rely on AI that excels at analysing details but struggles with holistic meaning, our own cognitive patterns may shift toward a more fragmented, decontextualised understanding of the world.
• Outsourcing of judgment: We risk delegating more decisions to systems that can process vast amounts of information, but lack the integrative judgment that comes from embodied, emotionally-grounded understanding.
• Social cognition impacts: As AI mediates more human interactions, the nuanced, intuitive aspects of social connection could become diminished, with relationships increasingly filtered through systems that don't truly understand human social reality.
• Epistemological distortions: AI's convincing but sometimes hollow interpretations of reality could produce content that appears meaningful, but lacks genuine connection to truth.
• Value alignment challenges: Without right-brain capacities for empathy and contextual moral reasoning, AI systems may optimise for quantifiable metrics while missing qualitative human values.

The most concerning scenario isn't that AI will become super intelligent and hostile, but rather that we'll increasingly adapt ourselves to fit AI's limitations – becoming more analytical and less embodied in our own perception of reality. As we shape our tools, our tools shape us in return.

However, awareness of these limitations also creates opportunities to design systems and practices that intentionally preserve and enhance our distinctly human capacities for holistic understanding.

Essential human capabilities

In response to these challenges, leadership now needs to cultivate and value specific human capabilities that complement AI's left-brain strengths:

• Contextual judgment - The ability to understand situations holistically, considering ethical implications and stakeholder needs that AI cannot fully grasp.
• Empathic understanding - Genuine emotional connection with others that enables leaders to anticipate needs and navigate complex human dynamics.
• Moral wisdom - The capacity to make principled decisions reflecting values and virtues, rather than just optimising for measurable outcomes.
• Systems thinking - Seeing interconnections between seemingly disparate elements and understanding second-order effects across organisational boundaries.
• Creative synthesis - Combining ideas across domains in novel ways, particularly drawing connections between technical possibilities and human needs.
• Embodied intuition - The "gut feeling" that integrates subconscious pattern recognition from years of experience.
• Adaptive resilience - Navigating ambiguity and maintaining equilibrium during change.

The new 'parallel processing'

This way of thinking rests not on rejecting AI and data-driven approaches, but in ensuring they remain tools that serve human flourishing, rather than becoming environments that diminish our humanity. This requires leadership to understand both the power and limitations of these technologies.

Like Dr. P, who "dealt with his students as if they were abstract patterns" and "saw patterns where others saw people", AI systems excel at detecting patterns but struggle with holistic understanding. The organisations that will thrive are those whose leaders consciously develop and invest in the right-brain capacities that complement AI's left-brain strengths.

As we navigate this challenging terrain, it’s helpful to remember T.S. Eliot's lament, from his play The Rock (1934): "Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information?" And perhaps most poignantly, "Where is the Life we have lost in living?"

The challenge we face is not technological but existential - how to maintain our humanity in a world increasingly mediated by technologies that excel at abstraction, but struggle with meaning. The path forward requires leadership that values and develops the distinctly human capabilities that AI cannot replicate.