Cognitive Neuroscience

Cognitive neuroscience is the attempt to explain how the brain makes the mind: how tissue, electricity, and chemistry produce perception, memory, language, attention, and the sense of self.

13 frameworks, from 1800 to today · Neuroscience · Natural Sciences

still a working view ended · each leads to its entry below

Overview

Cognitive neuroscience is the attempt to explain how the brain makes the mind: how tissue, electricity, and chemistry produce perception, memory, language, attention, and the sense of self1. Its practitioners combine three kinds of evidence—what happens to people when specific parts of the brain are damaged, what healthy brains do while people think, and models borrowed from computing and psychology—to ask where and how mental operations are implemented in the nervous system1. The field's history is a long argument between two intuitions: that different parts of the brain do different jobs, and that thought is spread across the whole organ. Most progress has come not from picking a winner but from discovering the levels at which each is true.

Before the Disciplines: Localization, Equipotentiality, and the Birth of Neuropsychological Inference

Serious debate about where mental functions live began in the early nineteenth century. Franz Joseph Gall proposed around 1800 that specific mental faculties—language, wit, combativeness—resided in specific cortical organs, and that their strength could be read from skull shape (phrenology)2. The skull mapping was wrong, but the core claim of localizationism—that distinct functions have distinct neural seats—survived2.

It gained clinical footing in 1861, when Paul Broca described a patient, Leborgne, who had lost articulate speech while understanding language and whose brain showed a lesion in the left inferior frontal region. Broca's area, as it came to be called, made the case that language production depends on a particular piece of cortex13. Carl Wernicke extended the argument in 1874, identifying a posterior temporal region whose damage impaired comprehension, and proposed that the two regions were linked by white-matter pathways. This disconnectionism—that complex abilities depend on connections between specialized centers—predicted new syndromes from pathway damage, a testable idea that shaped neurology for a century and that modern tractography has revived.

The opposing camp formed around 1870s experiments by Gustav Fritsch and Eduard Hitzig, who electrically stimulated dog cortex and elicited movements, and by David Ferrier's mapping of motor and sensory areas in monkeys. But it was Karl Lashley, working from the 1910s into the 1950s, who gave the anti-localizationist view its strongest statement. Training rats on mazes and then removing cortical tissue, Lashley found that the amount of tissue removed predicted performance loss better than its location: he called this mass action. He concluded that memory was not stored in one place but distributed across cortex—equipotentiality—while still believing some localization must exist. Lashley's student Donald Hebb supplied the mechanism in The Organization of Behavior (1949): co-active neurons strengthen their connections ("cells that fire together, wire together"), so memories could be physically distributed yet specific.

A crucial methodological insight emerged from this clinical tradition. If damage to region A impairs function X but not function Y, while damage to region B does the reverse—a double dissociation—one can argue that X and Y rely on distinct, partially separable systems3. This dissociation logic, sharpened by cognitive neuropsychology in the later twentieth century, turned case studies into evidence rather than anecdote3. It also required caution: because the brain is a network, single lesions often suffice to make the point, and interpreting patterns of spared and impaired performance became an art in itself.

Cognitive Psychology Supplies the Questions

Cognitive neuroscientists needed more than anatomy: they needed a theory of what the brain computes. That came from cognitive psychology. Beginning in the 1950s, researchers such as George Miller, Jerome Bruner, and Ulric Neisser treated the mind as an information-processing system, using reaction times and errors to infer internal stages of processing. Miller's 1956 work on the limits of short-term memory, and later the Sternberg paradigm (1966), in which reaction time grows with the number of items scanned in memory, showed that mental operations could be measured indirectly with precision. By the 1970s, researchers like Michael Posner and Roger Shepard were dissecting attention, imagery, and mental rotation into distinguishable processes.

The field's name was coined in 1976, when Michael Gazzaniga and George Miller launched what became the Cognitive Neuroscience Society, and the journal Cognitive Neuroscience appeared around 1991. The intellectual bet was that these process models could be mapped onto brain systems.

Two Roads to the Brain: Connectionism and Functional Imaging

In parallel, a computational strand took shape. Warren McCulloch and Walter Pitts had shown in 1943 that idealized neurons could compute logical functions45. The connectionist program—most famously Frank Rosenblatt's Perceptron (1958), a network that learned to classify patterns by adjusting connection weights—argued that cognition emerges from distributed activity in networks of simple units4. The idea faced a setback when Marvin Minsky and Seymour Papert's 1969 critique showed single-layer networks had fundamental limits, but it revived in the 1980s with back-propagation (popularized by Rumelhart, Hinton, and Williams, 1986), and parallel distributed processing models explained phenomena like graceful degradation after damage and content-addressable memory. Connectionism made specific, testable claims about how cognitive deficits should look—and they often matched.

The decisive methodological breakthrough was functional neuroimaging. In the late 1980s and early 1990s, positron emission tomography (PET), building on tracer work by Louis Sokoloff and colleagues in the 1970s, and then functional magnetic resonance imaging (fMRI) allowed researchers to measure blood oxygenation changes—the BOLD signal—while a person performed a task. Seiji Ogawa and colleagues demonstrated the BOLD contrast in 1990; Marcus Raichle's group and others showed it tracked local neural activity; and by the late 1990s, fMRI had become the field's workhorse. For the first time, researchers could see which regions were active during memory retrieval, language, or mental rotation in healthy people, not just infer from lesions. Electroencephalography (EEG), event-related potentials, and later magnetoencephalography (MEG) added millisecond timing to the picture16.

Networks, Embodiment, and the Predictive Turn

Functional imaging revealed a puzzle: almost any demanding task activated many regions simultaneously, and the same regions appeared across tasks. Roger Sperry's split-brain studies (from the 1960s) had already shown that the two hemispheres could operate independently, with specific lateralized functions; now whole-brain analyses suggested that cognitive functions emerge from networks rather than isolated modules. Network neuroscience, formalized from the 2000s onward by researchers such as Olaf Sporns and Ed Bullmore, applied graph theory to structural and functional connectivity, identifying hubs, modules, and the default mode network—a set of regions active during rest and deactivated during focused tasks, described by Raichle's group in 2001. The brain's small-world architecture—dense local clustering with efficient long-range connections—became a framework for thinking about both cognition and disease.

A second challenge came from embodied cognition. Critics argued that treating the brain as a disembodied computer ignored how body, action, and environment shape thought. Early proponents like Francisco Varela, Evan Thompson, and Eleanor Rosch (1991) argued that cognition is enacted, not just computed7. Embodied cognitive neuroscience turned this into experiment: work by Giacomo Rizzolatti's group in the 1990s on mirror neurons—cells that respond both to performing and to observing an action—became a touchstone, though its scope and interpretation remain debated. Studies of sensory and motor cortex showed that these areas represent actions and sensations in ways that influence judgment, and that the body's structure constrains how the brain represents space and tools.

A third strand, Bayesian brain and predictive processing, recast perception and action as inference. The idea, articulated by Hermann von Helmholtz in the nineteenth century as unconscious inference and formalized by David Knill, Whitman Richards, and others in the 1990s, holds that the brain combines prior expectations with noisy sensory data to estimate the causes of its inputs. Karl Friston's free energy principle (from 2010) extended this into a general theory: the brain minimizes prediction error by updating beliefs or acting on the world. Predictive processing reframed phenomena from illusion to attention to psychosis in terms of hierarchical prediction and error signals, and became one of the field's most active theoretical frameworks.

Today: Networks, Computation, and the Open Questions

Contemporary cognitive neuroscience operates at the intersection of all these traditions. Human neuroimaging has grown into large-scale consortia and resting-state connectivity studies; lesion and brain-stimulation methods—including transcranial magnetic stimulation (TMS)—continue to test causal claims that imaging alone cannot settle. Network science supplies both the data analysis toolkit and the conceptual vocabulary. Connectionist and Bayesian ideas have merged with deep learning, which now models visual recognition and language in ways that increasingly overlap with neural data.

The central questions remain open: How do distributed networks give rise to coherent experience? Which features of brain activity carry information versus merely correlate with it? To what extent are cognitive functions localized, and at what grain? And how do models built from computation relate to the biological machinery of cells and synapses? The field's history suggests the answers will be plural—different functions organized at different levels—rather than a single verdict for localization or holism.

References

  1. Cognitive neuroscience (introduction, Historical origins, Combining neuroscience and cognitive science), Wikipedia
  2. Phrenology (introduction, History), Wikipedia
  3. Cognitive neuropsychology (History, introduction), Wikipedia
  4. Connectionism, Wikipedia
  5. Cognitive science (History), Wikipedia
  6. Event-related potential, Wikipedia
  7. Enactivism, Wikipedia

The frameworks of Cognitive Neuroscience

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1800-Present

    Localizationism

    Do specific mental functions have specific brain locations?

    Fly to its tower in Nousfarer

  2. 1874-Present

    Disconnection syndrome

    Do complex abilities depend on connections between brain centers?

    Fly to its tower in Nousfarer

  3. 1910-1970

    Equipotentiality

    Is memory stored in one place or spread across the brain?

    Fly to its tower in Nousfarer

  4. 1950-Present

    Neuropsychological Dissociation Logic

    How can damaged brains reveal separate mental systems?

    Fly to its tower in Nousfarer

  5. 1956-Present

    Information-Processing Cognitive Psychology

    What stages of processing does the mind perform?

    Fly to its tower in Nousfarer

  6. 1958-Present

    Connectionism

    Can cognition emerge from networks of simple units?

    Fly to its tower in Nousfarer

  7. 1960-Present

    Cognitive Neuropsychology

    What do patterns of deficit reveal about cognitive architecture?

    Fly to its tower in Nousfarer

  8. 1960-1990

    Hemispheric Specialization

    Do the two hemispheres perform different mental jobs?

    Fly to its tower in Nousfarer

  9. 1985-Present

    Lesion and Brain-Stimulation Causal Methods

    Does damaging or stimulating a region cause a specific deficit?

    Fly to its tower in Nousfarer

  10. 1990-Present

    Functional Neuroimaging

    Which brain regions are active during a mental task?

    Fly to its tower in Nousfarer

  11. 1990-Present

    Predictive coding

    Does the brain infer causes from noisy sensory input?

    Fly to its tower in Nousfarer

  12. 1991-Present

    Embodied Cognition

    How do body and environment shape thought?

    Fly to its tower in Nousfarer

  13. 2001-Present

    Network Neuroscience

    How do brain regions form networks that support cognition?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

LocalizationismDisconnection syndromeEquipotentialityNeuropsychological Dissociation LogicInformation-Processing Cognitive Psyc…ConnectionismCognitive NeuropsychologyHemispheric SpecializationLesion and Brain-Stimulation Causal M…Functional NeuroimagingPredictive codingEmbodied CognitionNetwork Neuroscience

Beyond the field

Where Cognitive Neuroscience's ideas came from in other fields, and what they led to there: the strongest 12 of 86 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Cognitive Neuroscience is a planet in Nousfarer

Its city has a tower for each of these 13 frameworks. Land, walk in, and climb one concept at a time.

Land on Cognitive Neuroscience

Related fields

Fields its ideas are linked to

More in Neuroscience

See where Cognitive Neuroscience sits on the Map of Human Knowledge