Toward a Non-Scientistic Neuroscience

Why Neuroscience Needs the Humanities to Avoid Bad Science, False Certainty, and Real-World Harm

One of the most gifted students I have taught at the University of Texas at Austin was a neuroscience major. Like many of our strongest undergraduates, she was drawn to a field seen as empirical, technologically sophisticated, and closely connected to fundamental questions about thought, emotion, and behavior.

Yet for her senior thesis, she chose to write a work of history.

She brought to that project the habits she had learned in the lab—precision, skepticism, and respect for evidence—but also discovered what neuroscience alone could not offer: a way to think about meaning, context, moral complexity, and human experience over time. She learned how ideas shape institutions, how categories evolve, and how lives unfold within social and historical constraints.

She is now at Harvard Law School, pursuing a profession that depends on exactly that combination of analytic rigor and interpretive judgment.

Her path points to both a problem and an opportunity at the heart of contemporary brain science.

When Neuroscience Overreaches

In the mid-2000s, prominent journals published studies claiming to identify neural signatures of criminal propensity. Within a decade, brain scans were appearing in courtrooms, offered as explanations—or excuses—for violent behavior. Judges admitted this evidence even as neuroscientists warned it was being misunderstood and overinterpreted.

This was not an isolated misuse of research. Neuroscience has acquired such cultural authority that its findings rapidly migrate into law, medicine, education, and policy, often before they are conceptually clear or empirically secure. The result is not merely misapplied science, but bad science producing real harm.

Psychiatry offers a clear example. The chemical imbalance theory of depression—specifically, serotonin deficiency—came to dominate late 20th-century thinking. It shaped prescribing practices, insurance coverage, public messaging, and millions of people’s understanding of their own suffering. Antidepressants were prescribed at unprecedented rates, including to children, while psychotherapy lost institutional support.

Yet the theory was never well supported by evidence, and is now widely recognized as flawed. Large meta-analyses show that SSRIs perform only marginally better than placeboes for mild to moderate depression. The theory prevailed not because it was accurate, but because it was simple, fundable, and aligned with pharmaceutical incentives.

Patients were told their brains were broken when the sources of distress were often grief, trauma, isolation, or loss of purpose—experiences that require interpretation and therapeutic care, not medication alone.

Education provides another example. In the 1990s, “brain-based learning” promised to transform teaching. Schools adopted programs built around claims about learning styles, left-brain/right-brain differences, and critical developmental windows. Billions were spent. Almost none of these claims were supported by solid neuroscience. The programs failed—but only after a generation of educators had been trained in neuromyths.

Neuroscience’s prestige allows neurobiological explanations to crowd out other forms of understanding—even when those explanations are premature, incomplete, or wrong. The cost is not merely intellectual. It is clinical, educational, legal, financial, and moral.

Neuroscience has become so powerful that it can make us less, not more, thoughtful about ourselves.

The problem is not neuroscience itself, but scientism: treating a powerful method as a complete worldview, and assuming that what can be measured captures what most deeply matters. If we are to avoid reducing the mind to the brain—or the person to neural circuitry—we need not just better neuroscience, but neuroscience genuinely integrated with humanistic inquiry.

The humanities are not optional supplements to brain science. They are essential to getting the science right.

From Localization to Entanglement: What Neuroscience Has Learned—and What It Still Misses

Modern neuroscience began with a reasonable ambition: to map mental functions onto specific brain regions. Early 20th-century work by Korbinian Brodmann and Oskar Vogt and Cécile Vogt-Mugnier exemplified this effort. By dividing the cortex into distinct anatomical areas, they hoped to identify the brain’s functional units—the “organs of the mind.”

This approach rested on a compelling intuition: structure determines function. Some successes followed, particularly in sensory and motor systems. But most higher-order processes resisted tidy mapping. Over time, researchers found that single regions participate in multiple functions, and that any given function involves multiple regions of the brain.

The model shifted from localization to networks.

Contemporary neuroscience now emphasizes dense connectivity across multiple brain regions. Local circuits are embedded in long-range pathways; cortical regions communicate through hubs such as the thalamus; subcortical structures participate in loops shaping perception, memory, emotion, and action.

Luiz Pessoa’s framework of the “entangled brain” captures this shift. Cognition and emotion are not separate systems but deeply intertwined. Neural ensembles form dynamically in response to context, assembling and disassembling as tasks and environments change. What matters is not which region activates, but how distributed patterns coordinate.

This view aligns with broader developments in cognitive science. Karl Friston emphasizes predictive processing—that the brain constantly generates and updates a mental model of the world to predict incoming sensory input.

Michael Gazzaniga argues for distributed cognition—that mental processes are not confined to the individual brain but are distributed across social systems, external artifacts (like tools), and groups of people.

Olaf Sporns focuses on neuroplasticity—on how the brain's complex network structure allows it to adapt and reorganize in response to experience and environment.

Antonio Damasio shows how emotions and feelings is integral to reasoning, decision-making, selfhood, and consciousness.

These approaches undermine simplistic reductionism and portray the brain as a complex, adaptive system shaped by evolution, development, and experience.

And yet, even at this level of sophistication, crucial questions remain unanswered—and often unasked:

▪ How do neural processes give rise to meaning?

▪ How does culture and environment shape the brain’s organization and use?

▪ How should we understand consciousness, normativity, and self-interpretation?

▪ What is the relationship between neural correlation and lived experience?

On these questions, neuroscience is not merely incomplete. It lacks the conceptual tools to address them on its own. And it is here that the absence of humanistic thinking produces not just gaps in understanding, but systematic error.

The Problem of Reification: When Neuroscience Gets the Science Wrong

The most common and damaging mistake in contemporary neuroscience is not crude reductionism but reification: the tendency to turn abstractions into things, processes into entities, correlations into causes, and measurements into explanations.

A familiar example is the claim that “the amygdala is the brain’s fear center.” The phrase is convenient, but it is misleading. The amygdala does play a role in fear responses, yet it is also involved in reward processing, social evaluation, uncertainty detection, arousal, attention, and memory formation. Its function depends on context and network dynamics. To label it a dedicated “fear center” is not just an oversimplification; it is wrong.

The deeper problem is conceptual. Fear itself is not a single, uniform state. It includes sudden fright, chronic anxiety, panic, dread, phobic avoidance, and moral horror. These experiences differ in timing, object, bodily expression, and meaning. Treating them as variants of one neural mechanism flattens distinctions that matter both clinically and scientifically.

Early anxiety research followed a simple model: a threat appears, the amygdala activates, and fear follows. Treatments built on this model—such as exposure therapies aimed at dampening amygdala reactivity—helped some patients but failed many others.

Why? Because anxiety disorders are not all the same.

Panic disorder involves sudden, overwhelming episodes focused on immediate physical danger. Generalized anxiety is diffuse, chronic, and anticipatory. Social anxiety centers on shame and self-evaluation. Moral anxiety involves guilt and concerns about worth.

These experiences unfold over different temporal horizons and relate to the self in different ways. They are not minor variations of a single process; they are distinct forms of human experience, requiring different kinds of intervention.

Progress has come not from sharper brain images, but from attention to lived experience: clinical observation, phenomenology, and philosophical reflection on emotional life. Neuroscience advanced when it learned to respect these distinctions.

The same pattern appears in memory research. For decades, memory was treated as information storage and retrieval—a legacy of the computer metaphor. Memories were “encoded,” “stored,” and “retrieved.” This framework generated important discoveries about the hippocampus, consolidation, and brain plasticity.

But it also created blind spots. It struggled to explain why remembering alters memories, why memory is shaped by narrative, why people avoid or distort recollections, and why remembering is often social—distributed across people, practices, and artifacts.

This research has forced a rethinking of what memory is. Once again, the key insights did not come from better imaging, but from autobiography studies, oral history, psychoanalysis, and literary analysis—fields that had long treated memory as reconstruction rather than storage. Humanists understood this long before neuroscience caught up.

Perhaps the clearest case of reification is the long-standing claim that dopamine is the brain’s “reward chemical.” For decades, dopamine spikes were taken to signal pleasure, and addiction was understood as a hijacking of the brain’s reward system.

This account was elegant—and largely wrong.

Kent Berridge’s research showed that dopamine tracks wanting, not liking. It signals incentive salience, not pleasure itself. People can intensely desire what they do not enjoy, and enjoy what they did not strongly anticipate. Craving, habit, anticipation, and satisfaction are distinct experiences, with distinct neural profiles.

This insight transformed addiction science. And it emerged by taking seriously distinctions philosophers and psychologists had been making for centuries. Aristotle distinguished appetite from flourishing. Schopenhauer distinguished will from satisfaction. Harry Frankfurt distinguished first-order desires from higher-order volitions. These were not neuroscientific concepts—but they proved essential for interpreting what dopamine activity actually measures.

When neuroscience ignored these distinctions, it produced confused theories and ineffective treatments. When it incorporated them, the science improved.

The lesson is not that philosophy can replace neuroscience. It is that neuroscience cannot succeed without conceptual clarity, phenomenological grounding, and careful attention to what is being explained.

Depression, Conceptual Confusion, and the Serotonin Debacle

The chemical imbalance theory of depression offers a cautionary tale. It shows how a weak explanatory model can dominate for decades—and how humanistic insight might have prevented the error.

The idea that depression is caused by low serotonin was never a settled scientific conclusion. It was a simplification—useful for public communication and pharmaceutical marketing, but not a robust causal theory. Yet it became culturally dominant, shaping research agendas, clinical practice, insurance coverage, and millions of people’s understanding of themselves.

The core problem is that depression is not a single phenomenon. It is a diagnostic category that bundles together diverse experiences: loss of pleasure, slowed thought and movement, guilt, a sense of worthlessness, suicidal ideation, disrupted sleep and appetite, and social withdrawal. These features can co-occur, but they need not. They have different causes, trajectories, and meanings.

Depression following bereavement is not the same as chronic dysthymia. Depression centered on guilt differs from depression marked by emptiness. Depression shaped by trauma is not the same as depression with no clear precipitant. Treating all of these as serotonin deficiency is a category mistake.

That mistake had real consequences. Some people benefit greatly from SSRIs; others experience modest relief; many see no benefit at all. The wide variation reflects the heterogeneity of the underlying conditions.

Philosopher Matthew Ratcliffe has shown how first-person accounts reveal structures invisible to third-person measures. Depression is not merely reduced positive affect. It is a transformation in how the world is experienced: time feels stalled, action feels pointless, the future appears closed, and the self feels cut off from meaning and connection.

These are not surface symptoms layered atop a hidden neural defect. They are constitutive of what depression is.

They cannot be captured by neurotransmitter levels or regional activation patterns alone. They require description, interpretation, and conceptual analysis—the kinds of work the humanities specialize in.

Had psychiatry and neuroscience taken phenomenology seriously from the outset, the chemical imbalance theory would never have achieved its dominance. We would have developed more precise categories, more targeted treatments, and fewer false expectations.

The cost of ignoring humanistic insight was not merely intellectual confusion. It was unnecessary suffering and expense.

The Prestige Asymmetry: Why Neuroscience Appears “Harder”

At large research universities, many of the strongest students gravitate toward fields seen as innovative, powerful, and future-oriented: neuroscience, computer science, data science, engineering, and economics. These disciplines carry cultural prestige and institutional prioritization, promising clear pathways to funding, influence, and visible impact.

The humanities, by contrast, are often viewed as reflective rather than generative, backward-looking rather than forward-driving. They may be admired, even respected, but they are rarely seen as essential to understanding how the world actually works.

This hierarchy is deeply embedded in university culture, but it rests on shaky intellectual foundations. It is sustained by three persistent mistakes.

Mistake 1: Confusing Technological Sophistication with Explanatory Depth

Brain images look like pictures of thought. Functional MRI scans are colorful, precise, and reproducible, giving the impression that we are watching the brain think or feel in real time. This visual power explains much of their appeal—and much of their danger.

In fact, fMRI does not measure neural activity directly. It tracks changes in blood oxygenation as a proxy for metabolic demand. The signal is slow, unfolding over seconds rather than milliseconds; spatially coarse, averaging across millions of neurons; and heavily processed through statistical models that compare conditions, average across subjects, and correct for noise.

The bright spots on a scan are not thoughts. They are inferences about where metabolic activity may have increased under specific assumptions, after extensive data processing.

More important, identifying brain activation is not explanation. Showing that a region becomes active during a task tells us little about what it is doing, why it matters, or how it contributes to experience. Localization is not understanding.

Neuroscience derives much of its prestige from technological and mathematical sophistication. But technical power is not the same as explanatory depth. A field can be rigorous and high-tech—and still be conceptually confused.

Mistake 2: Treating Empirical Problems as Harder Than Conceptual Ones

Neuroscience confronts formidable empirical challenges: how neurons coordinate across distance, how memories consolidate, how the brain integrates sensory information. These problems demand ingenuity, advanced tools, and careful analysis.

But the deepest problems in neuroscience are not empirical. They are conceptual.

We do not lack data about consciousness; we lack clarity about what consciousness is. We do not need more scans to understand free will; we need clearer accounts of agency, responsibility, and choice. The binding problem is not solved by finding more neurons; it concerns how distributed processes yield a unified experience.

Philosophy has long grappled with concepts such as intentionality, selfhood, temporality, normativity, and meaning. These are not pseudo-problems discussed in obscure language; they are genuinely difficult questions about how experience is structured and how understanding is possible.

Literature and history confront similarly demanding issues: how people make sense of suffering over time, how identities form and change, how narratives shape what individuals and societies regard as possible, and how we interpret ambiguous evidence. These inquiries require rigor and discipline no less than laboratory science.

The difference is not that humanistic problems are “soft.” It is that they concern meaning rather than measurement—and meaning resists quantification.

Mistake 3: Assuming Science Answers Questions the Humanities Merely “Contextualize”

A common view holds that science discovers facts while the humanities provide context or interpretation. This division of labor is false.

Humanistic inquiry is not external to science; it is internal to it. Every experiment depends on:

▪ Conceptual clarity about what is being measured

▪ A grounded understanding of what needs explaining

▪ Awareness of how categories are shaped by culture and history

▪ Ethical reflection on which questions matter and for whom

▪ Interpretive skill in making sense of complex results

These are preconditions of good science.

When neuroscience proceeds without them, it does not become more rigorous. It becomes more precise about poorly defined phenomena. It generates results that fail to replicate because the underlying concepts were never clear.

What the Best Neuroscientists Already Know

The claim that neuroscience needs the humanities is not a plea from outside the field. Many of its most thoughtful practitioners have already reached this conclusion.

Iain McGilchrist argues that today’s reductionist approach to neuroscience reflects a broader cultural tendency to break things into parts and try to control them. He suggests that modern Western culture favors a narrow, mechanical way of understanding the world, while neglecting a more holistic, context-sensitive way of seeing things.

Georg Northoff argues that the self cannot be understood through neural data alone. Patterns of brain activity associated with self-reference are not the same as the lived experience of being a self. Without first-person, phenomenological accounts, neuroscience lacks a clear grasp of the very phenomenon it claims to explain.

Francisco Varela’s neurophenomenology argues that first-person experience and third-person scientific methods must be brought together. Conscious experience is the very phenomenon neuroscience exists to understand.

Vittorio Gallese’s work on mirror neurons reshaped social neuroscience by drawing on phenomenology. His idea of “embodied simulation” is based on the insight that we understand other people not by abstract reasoning alone, but by directly reflecting upon their actions and emotions—an idea philosophers had articulated long before neuroscience provided neural evidence for it.

Lisa Feldman Barrett’s theory of constructed emotion draws on insights from cultural psychology and social construction. Emotions are not hardwired reflexes but learned ways of interpreting experience—an idea long familiar to the humanities, and one that has enabled more precise scientific inquiry.

These figures point toward neuroscience at its best: integrative rather than isolated, conceptually sophisticated rather than technologically dazzling.

The Institutional Problem

The remaining obstacle is institutional. Universities reward specialization more than synthesis. Funding agencies favor technical methods. Journals privilege quantification over interpretation. Graduate training produces technical experts rarely taught to reflect philosophically on their work.

If we want neuroscience that explains rather than merely measures, these structures must change. The goal is give it the conceptual partners it needs to succeed.

Five Roles the Humanities Can Play in Brain Science

The humanities contribute to neuroscience in distinct but overlapping ways. Each role addresses a different kind of problem, and together they explain why humanistic inquiry is not optional but essential.

Role 1: Conceptual Clarification (Philosophy)

Philosophy guards against category errors and clarifies levels of explanation. It distinguishes:

▪ Neural activity from mental states

▪ Correlation from causation

▪ Mechanisms from meanings

▪ First-person experience from third-person measurement

When neuroscience claims that “the amygdala is fear,” philosophy asks: What is fear? Is it a brain state, a bodily response, an evaluative judgment, or some combination of these? Are we explaining fear itself, or merely identifying one of its neural correlates?

This is not pedantry. It is the difference between careful explanation and conceptual error.

Role 2: Phenomenological Grounding (Philosophy, Literature)

Phenomenology provides first-person descriptions of what actually needs explaining. Everyday psychological language often collapses experiences that differ in structure and meaning.

Fear and anxiety are not the same. Fear has a clear object and an immediate temporal structure: a threat appears, the body responds. Anxiety is more diffuse and anticipatory, oriented toward what might happen rather than what is happening.

Guilt and shame are also distinct. Guilt concerns actions and can, at least in principle, be repaired. Shame is more global, attaching to identity rather than behavior, and is therefore harder to escape.

Boredom is not simply low stimulation. One can be busy and still bored. Boredom reflects a breakdown of meaning, a sense that time itself has become empty.

Depression, likewise, is not just intense sadness. It alters the structure of experience, flattening the future, draining motivation, and changing how the self relates to the world.

These distinctions matter because mental life cannot be captured by blunt categories or single biological measures. Each experience has its own internal logic, temporal shape, and relationship to meaning.

Literature functions as a laboratory for exploring these dimensions. Jane Austen and George Eliot clarify self-deception and moral judgment. Fyodor Dostoevsky probes guilt and freedom. Virginia Woolf captures inner time. Henry James dissects perception and misinterpretation. These works do not replace science—they sharpen what science needs to explain.

If neuroscience ignores phenomenology, it risks measuring the wrong thing.

Role 3: Cultural and Historical Context (History, Anthropology)

Culture is constitutive of neural development and mental life.

Lev Vygotsky argued that higher mental functions are internalized social relations. Literacy reshapes neural circuits. Music trains timing and emotion. Religious ritual organizes attention and affect. Digital media reconfigures memory and distraction. Trauma is culturally mediated: whether it is recognized or denied profoundly shapes its course.

Anthropology shows that emotional categories, self-concepts, and even perception vary across cultures. Joseph Henrich has demonstrated that much psychological research generalizes from WEIRD populations—a narrow and unrepresentative slice of humanity.

To ignore culture is not to be objective. It is to universalize a parochial baseline and mistake it for human nature.

Role 4: Historical Reflexivity (History, Sociology of Science)

The history of science reveals that neuroscientific categories are themselves historically constructed. “Childhood,” “adolescence,” “trauma,” “ADHD,” and “burnout” are not timeless categories but concepts that emerged under specific social and institutional conditions.

Ian Hacking’s idea of “looping effects” shows how classifications applied to people reshape how they understand themselves—and thereby reshape the phenomena being classified. Neurobiological explanations do not merely describe mental illness; they alter people’s identity, sense of responsibility, and expectations.

Nikolas Rose has shown how the rise of the “neurochemical self” encourages people to view emotions and behavior as brain states to be managed or optimized, often pharmacologically. This is not neutral science. It is a transformation of selfhood with ethical and political consequences.

Role 5: Ethical and Normative Critique (All Humanities)

Finally, the humanities ask questions that science cannot answer on its own:

▪ What counts as mental illness, and what counts as normal variation?

▪ Who benefits from neurobiological explanations of social problems?

▪ How should moral responsibility be understood in light of neural causation?

▪ What is lost when suffering is treated as a technical malfunction rather than meaningful response to an event or a loss?

These questions are central to what neuroscience is for.

What This Means in Practice

A skeptical neuroscientist might accept these arguments in theory and still ask: What difference would this actually make to my research?

The answer is straightforward: better experiments, clearer results, stronger theories, and more effective applications.

For Basic Research

Better experimental design. Define phenomena before measuring them. “Emotion” is not a single construct. Fear, anxiety, anger, shame, and disgust differ in structure and function. Fear, panic, phobia, and dread are not the same thing. Phenomenology supplies the distinctions that experiments require.

More interpretable results. Know what you have measured. An fMRI finding showing amygdala activation during “negative emotion” means little unless the emotion, context, and subjective meaning are specified.

Stronger theories. Avoid conceptual confusion. If a theory claims consciousness is an illusion, it must explain what is being excluded. If agency is reduced to neural causation, the theory must still account for responsibility and normativity.

For Clinical Practice

More effective treatment. Medication can alter neurochemistry, but recovery often requires meaning-making, narrative repair, social connection, and renewed purpose. Trauma therapy is not just exposure to reduce reactivity; it is the integration of experience into a coherent life story.

Better diagnosis. Track lived experience, not just biomarkers. Diagnosis should be grounded in phenomenology, development, and culture—not only symptom checklists or scans.

Reduced iatrogenic harm—suffering caused by medical intervention itself—by resisting the urge to treat complex human experiences as simple dysfunctions. Grief is not a broken brain. Feelings of defeat are not serotonin deficiency. Some forms of suffering call for social, moral, or existential responses, not medical ones.

For Policy

More realistic expectations. Neuroscience can inform education, criminal justice, and social policy—but it cannot design schools, justify punishment, or define responsibility on its own.

Resistance to neuro-determinism. Brain differences do not excuse crime, predict behavior, or justify coercion.

Recognition that social problems need social solutions. Inequality, poverty, racism, and isolation affect the brain—but the remedy is justice, not neural intervention.

For Public Understanding

Less fatalism. Neural mechanisms do not eliminate agency. They are part of how agency is exercised.

A clearer mind–brain relationship. The brain is necessary for mental life, but not sufficient. Experience is shaped by culture, relationships, history, and self-interpretation.

Preservation of moral categories. People are not brains. They are persons—beings who act for reasons, live in time, and hold one another accountable.

The Choice We Face

The question is not whether neuroscience should study the brain. It should. The question is whether it can afford to proceed without the humanities.

The answer is no.

Neuroscience that ignores conceptual clarity, phenomenology, culture, and history does not become more scientific. It becomes more confused—measuring precisely while misunderstanding fundamentally, finding correlations and mistaking them for explanations.

The choice is not between neuroscience and the humanities. It is between good neuroscience—conceptually rigorous, empirically grounded, and interpretively informed—and bad neuroscience that confuses measurement with understanding.

The brain may be entangled. So is knowledge. Neural processes are embedded in histories, cultures, practices, and forms of life. Studying the brain in isolation studies an abstraction, not a mind.

That student who moved from neuroscience to history to law understood something essential: the most powerful empirical tools, joined to interpretive judgment and ethical reflection, produce not just better scholars—but better science.

If we want to understand minds—not just brains—we will need that combination.