Navigational Mind

00 — The ‘seat’ of memory

Before 1953, memory was not associated with any one area of the brain. Lashley spent decades looking for the trace and though this is a long story and lots of debate, the idea was sort of that memory is everywhere and nowhere. But epilepsy and surgeons removing the medial temporal lobes of patients brought scientists to focus on the hippocampus. As part of this, Henry Moalison, a 27-year-old man with intractable epilepsy, had his hippocampi revoved and then his seizures improved but he could not form any new memories, or could not attach anything new to the memory he already had.

Note: The operation was experimental, performed on a patient in distress, and it destroyed a man's capacity to build a life in exchange for knowledge the rest of us now have. Scoville came to regard it as a mistake. H.M. spent fifty-five years being tested by researchers he met, each time, for the first time.

01 — The cognitive map

A cognitive map is often defined as a type of mental representation used by an individual to order their personal store of information about their everyday or metaphorical spatial environment, and the relationship of its component parts. The concept was introduced by Edward Tolman in 1948. He tried to explain the behavior of rats that appeared to learn the spatial layout of a maze, and subsequently the concept was applied to other animals, including humans The term was later generalized by some researchers, especially in the field of operations research, to refer to a kind of semantic network representing an individual's personal knowledge or schemas.

What is happening with the navigataional approach is that the map is becoming a ‘mapping’ and there are not stored internal representations but rahter different actions of communication at many scales, from the way we communicate with our own bodies as those bodies to the ways we communicate with all we encounter.

02 — Seeing cells fire

03 — Concepts and structure

If the grid code represents relations, it should appear wherever a task has map-like structure — including tasks with nothing spatial in them.

04 — Cognitive spaces: Doeller lab at Max Planck (where i did my master thesis) and others

The human-imaging side, pushed into theory: the hippocampal formation supplies a geometric format that any set of dimensions can be mapped into.

05 — Sound, smell, faces, status

The dimensions usually called "non-spatial."

Note: the hexadirectional fMRI signature is an indirect, population-level proxy for grid coding, not a recording of grid cells. Productive, and contested (like everything else).

06 — When the map is lost

07 — Imagining is continuous with remembering?

Amnesic patients couldn't remember their past. Then someone asked them to imagine lying on a white sandy beach, and that was also a struggle and most couldn't do that either, so there is a link here. Memory isn't a recording you play back; it's a construction, and imagination runs on similar ‘habits’ and ‘machinery’.

08 — Bodily action

Towards stronger claims: that remembering and imagining aren't just neurally similar but enacted — carried out by eyes, motor systems and place-cell sequences doing versions of what they do when you actually move and look.

Note: evidence is strong for shared ‘machinery’ — overlapping networks, re-enacted scanpaths, motor reactivation, sequences running both directions. Remebering and imagining are not the same but have operational coherence, is how I would put it.

09 —Selective attention; the worlds choose and the worlds chosen for us and the worlds we miss

10 — Attention writing to the hippocampus

11 — Rereading the map

If one system maps rooms, concepts, pitches and people, either "space" was the wrong word, or thinking was always more like moving than we assumed.

Some conversations about some of this

‍ ‍

1. Foundational constructs (classic cognitive science & theory)

These define the ideas the neuroscience later tried to localize. Mostly books/older papers; could not always find links.

  • Bartlett, F. C. (1932).Remembering: A Study in Experimental and Social Psychology. Cambridge University Press. — Introduced "schema" as an active, reconstructive structure: memory is rebuilt to fit expectations rather than replayed verbatim.

  • Piaget, J. (1926/1952).The Origins of Intelligence in Children. — Schemas as assimilation/accommodation structures through which knowledge is built and updated during development.

  • Schank, R. C., & Abelson, R. P. (1977).Scripts, Plans, Goals and Understanding. Lawrence Erlbaum. — The origin of "script": stereotyped event sequences (the famous restaurant script) used to understand language and predict how situations unfold.

  • Miller, G. A. (1956). "The Magical Number Seven, Plus or Minus Two." Psychological Review 63:81–97. — The original "chunking" idea: we compress information into a handful of packaged units to beat capacity limits. Later generalized to action.

  • Tolman, E. C. (1948). "Cognitive maps in rats and men." Psychological Review 55:189–208. https://doi.org/10.1037/h0061626 — Proposed that animals build internal "maps" of relationships, not just stimulus–response chains — the seed of the modern cognitive-map program.

  • O'Keefe, J., & Nadel, L. (1978).The Hippocampus as a Cognitive Map. Clarendon Press. — Tied Tolman's map to the hippocampus after the discovery of place cells; the anatomical anchor for everything downstream.

  • Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). "Learning representations by back-propagating errors." Nature 323:533–536. — The connectionist backdrop in which schemas emerge from distributed networks rather than living as discrete symbols.

2. Rodent studies — schemas and rapid systems consolidation

causal, cellular-level evidence.

  • Tse, D., et al. (Morris lab) (2007). "Schemas and memory consolidation." Science 316:76–82. https://www.science.org/doi/abs/10.1126/science.1135935 — Landmark study: once rats had a flavor–place schema, brand-new pairings learned in a single trial became neocortical and hippocampus-independent within ~48 hours, showing schemas can hugely accelerate consolidation. (Follow-up: Tse et al., 2011, Science showed matching immediate-early-gene activity in medial prefrontal cortex.)

  • McKenzie, S., et al. (Eichenbaum lab) (2014). "Hippocampal Representation of Related and Opposing Memories Develop within Distinct, Hierarchically Organized Neural Schemas." Neuron 83:202–215. https://www.cell.com/neuron/fulltext/S0896-6273(14)00405-X — Hippocampal ensembles organize related events hierarchically by shared dimensions and keep opposing contexts separate — a neural instantiation of a relational schema. (See also the McKenzie & Eichenbaum, 2011 review.)

  • McKenzie, S., et al. (2013). "Learning Causes Reorganization of Neuronal Firing Patterns to Represent Related Experiences within a Hippocampal Schema." Journal of Neuroscience 33:10243–10256. https://www.jneurosci.org/content/33/25/10243 — New goal memories are first assimilated into existing goal-related firing patterns, then gradually differentiate — consolidation as assimilation-then-separation.

  • Zhou, J., et al. (Schoenbaum lab) (2019). "Complementary task structure representations in hippocampus and orbitofrontal cortex during an odor sequence task." Current Biology 29:3402–3409. https://doi.org/10.1016/j.cub.2019.08.040 — Hippocampus and OFC carry complementary representations of abstract task structure (schema), extending schema coding beyond space.

  • Wang, S.-H., & Morris, R. G. M. (2010). "Hippocampal–neocortical interactions in memory formation, consolidation, and reconsolidation." Annual Review of Psychology 61:49–79. https://www.researchgate.net/publication/26645633 — Authoritative review situating the schema/rapid-consolidation findings within hippocampal–neocortical dialogue.

3. Human neuroimaging — schemas, congruency, and event scripts

  • van Kesteren, M. T. R., Ruiter, D. J., Fernández, G., & Henson, R. N. (2012). "How schema and novelty augment memory formation." Trends in Neurosciences 35:211–219. — The SLIMM model (Schema-Linked Interactions between Medial prefrontal and Medial temporal regions): mPFC detects congruency with existing schemas and modulates hippocampal encoding accordingly. (Locate by title/DOI 10.1016/j.tins.2012.02.001.)

  • Baldassano, C., Hasson, U., & Norman, K. A. (2018). "Representation of Real-World Event Schemas during Narrative Perception." Journal of Neuroscience 38:9689–9699. https://www.jneurosci.org/content/38/45/9689 — The key "script" imaging study: mPFC, posterior medial cortex, and superior frontal gyrus carried restaurant-vs-airport script patterns that generalized across stories and modalities; mPFC uniquely tracked event order.

  • Baldassano, C., et al. (2017). "Discovering Event Structure in Continuous Narrative Perception and Memory." Neuron 95:709–721. https://doi.org/10.1016/j.neuron.2017.06.041 — Introduced the Hidden Markov Model method for finding event boundaries in continuous experience — the methodological backbone of the script work.

  • Masís-Obando, R., Norman, K. A., & Baldassano, C. (2022). "Schema representations in distinct brain networks support narrative memory during encoding and retrieval." eLife 11:e70445. https://elifesciences.org/articles/70445 — Dissociated story-specific vs. general schematic representations across cortical and hippocampal regions during memory, not just perception.

  • Song, H., et al. (2023). "Top-down attention shifts behavioral and neural event boundaries in narratives with overlapping event scripts." bioRxiv.https://www.biorxiv.org/content/10.1101/2023.08.08.552465 — Tackles the realistic case where multiple scripts overlap (e.g., a birthday at a restaurant), showing attention reshapes where event boundaries fall.

  • Greve, A., Cooper, E., Tibon, R., & Henson, R. N. (2019). "Knowledge Is Power: Prior Knowledge Aids Memory for Both Congruent and Incongruent Events, but in Different Ways." Journal of Experimental Psychology: General.https://pmc.ncbi.nlm.nih.gov/articles/PMC6390882/ — Confirmed SLIMM's predicted U-shaped memory curve: both highly schema-congruent and highly incongruent events are remembered better than neutral ones, via different mechanisms.

  • Sommer, T., et al. (2022). "The Assimilation of Novel Information into Schemata and Its Efficient Consolidation." Journal of Neuroscience 42:5916. https://www.jneurosci.org/content/42/30/5916 — Human evidence on vmPFC–hippocampal coupling when new facts can be related to prior knowledge, testing competing schema-consolidation accounts.

  • Study on self-generated learning (2024). "Recognizing ideas generated in a creative task: the roles of the hippocampus and medial prefrontal cortex." Cerebral Cortex 34:bhae219. https://academic.oup.com/cercor/article/34/5/bhae219/7682115 — Applies the schema/SLIMM framework to how prior knowledge scaffolds novel idea generation.

  • Post-encoding connectivity & durable memory (2023). "Effects of schema on the relationship between post-encoding brain connectivity and subsequent durable memory." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10229577/ — Shows schema-consistency reconfigures which post-encoding networks (vmPFC–hippocampal vs. sensory–vmPFC) predict lasting memory.

  • Spatial schemas & one-shot navigation (2018). "Learned Spatial Schemas and Prospective Hippocampal Activity Support Navigation After One-Shot Learning." Frontiers in Human Neuroscience 12:486. https://www.frontiersin.org/articles/10.3389/fnhum.2018.00486/full — Human analogue of Tse: a learned spatial schema supports rapid, one-shot integration of new locations with reduced hippocampal dependence.

4. Theoretical frameworks and reviews

  • Ghosh, V. E., & Gilboa, A. (2014). "What is a memory schema? A historical perspective on current neuroscience literature." Neuropsychologia 53:104–114. https://www.sciencedirect.com/science/article/abs/pii/S0028393213003990 — Attempts to give "schema" a rigorous, componential definition, cutting through inconsistent usage across the field.

  • Gilboa, A., & Marlatte, H. (2017). "Neurobiology of Schemas and Schema-Mediated Memory." Trends in Cognitive Sciences 21:618–631. https://www.sciencedirect.com/science/article/abs/pii/S1364661317300864 — The major synthesis; positions vmPFC, hippocampus, angular gyrus and posterior cortex as the schema network and details their interactions.

  • Sekeres, M. J., Winocur, G., & Moscovitch, M. (2017/2018). "Details, gist and schema: hippocampal–neocortical interactions underlying recent and remote episodic and spatial memory." Current Opinion in Behavioral Sciences / related Trace Transformation work. https://www.sciencedirect.com/science/article/abs/pii/S2352154616302777 — Trace Transformation Theory: memories shift from detailed to gist-like/schematic over time, with the entorhinal cortex bridging anterior hippocampus and mPFC.

  • Kesner, R. P., & Rolls, E. T. / "Revisiting a unified theory" (2018). "Coordinating what we've learned about memory consolidation: Revisiting a unified theory." Neuroscience & Biobehavioral Reviews.https://www.sciencedirect.com/science/article/abs/pii/S0149763418302021 — Integrates schema findings (Tse) and engram findings (Kitamura/Tonegawa) into a unified account of cellular + systems consolidation.

  • Preston, A. R., & Eichenbaum, H. (2013). "Interplay of hippocampus and prefrontal cortex in memory." Current Biology 23:R764–R773. https://doi.org/10.1016/j.cub.2013.05.041 — Widely cited framework for how hippocampus and PFC divide labor in encoding, schema use, and retrieval.

  • McClelland, McNaughton & O'Reilly (1995) / Kumaran, Hassabis & McClelland (2016). Complementary Learning Systems theory. Psychological Review 102:419–457; Trends in Cognitive Sciences 20:512–534. — The backdrop: a fast hippocampal learner and a slow neocortical generalizer must be separate to avoid catastrophic interference — the reason schemas exist at all.

5. Entorhinal–hippocampal cognitive maps & structural knowledge

most specific to the hippocampal–entorhinal area — schema as structural code in a spatial-like format.

  • Constantinescu, A. O., O'Reilly, J. X., & Behrens, T. E. J. (2016). "Organizing conceptual knowledge in humans with a gridlike code." Science 352:1464–1468. https://pmc.ncbi.nlm.nih.gov/articles/PMC5248972/ — When people navigated an abstract 2D concept space, entorhinal cortex and vmPFC showed the same hexagonal grid signal as in spatial navigation — evidence that grid codes organize non-spatial knowledge.

  • Whittington, J. C. R., et al. (Behrens lab) (2020). "The Tolman-Eichenbaum Machine: Unifying Space and Relational Memory through Generalization in the Hippocampal Formation." Cell 183:1249–1263. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7707106/ — A model in which medial entorhinal cells encode reusable structure and hippocampal cells bind it to sensory content, reproducing grid/place cells and generalizing schemas across environments.

  • Behrens, T. E. J., et al. (2018). "What Is a Cognitive Map? Organizing Knowledge for Flexible Behavior." Neuron 100:490–509. https://www.researchgate.net/publication/328509715 — The conceptual manifesto arguing the hippocampal–entorhinal map encodes relational structure of any domain, enabling inference and generalization.

  • Bellmund, J. L. S., Gärdenfors, P., Moser, E. I., & Doeller, C. F. (2018). "Navigating cognition: Spatial codes for human thinking." Science 362:eaat6766. https://doi.org/10.1126/science.aat6766 — Proposes that hippocampal–entorhinal geometric codes map "cognitive spaces," letting us reason about concepts as if navigating them.

  • Peer, M., Brunec, I. K., Newcombe, N. S., & Epstein, R. A. (2021). "Structuring Knowledge with Cognitive Maps and Cognitive Graphs." Trends in Cognitive Sciences 25:37–54. https://www.cell.com/trends/cognitive-sciences/abstract/S1364-6613(20)30250-3 — Distinguishes continuous "map" codes from graph-like relational codes as two formats the brain uses to structure knowledge.

  • Hernández-Frausto, M., et al. (2024). "Entorhinal cortex–hippocampal circuit connectivity in health and disease." Frontiers in Human Neuroscience 18:1448791. https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2024.1448791/full — Up-to-date review of the EC–hippocampal circuitry (including its breakdown in Alzheimer's) that underlies memory and map formation.

  • Bellmund, J. L. S., et al. "Mapping sequence structure in the human lateral entorhinal cortex." bioRxiv.https://www.biorxiv.org/content/10.1101/458133 — Human lateral entorhinal cortex represents the temporal/sequence structure of events, extending map coding into time.

  • Working-memory & the EC–hippocampal circuit. "Successful working memory linked to theta connectivity patterns in the hippocampal-entorhinal circuit." bioRxiv.https://www.biorxiv.org/content/10.1101/2022.09.08.507081 — Theta-band EC–hippocampal coupling supports holding information online, linking the circuit to active maintenance.

  • Eichenbaum, H. (2017). "On the Integration of Space, Time, and Memory." Neuron 95:1007–1018. https://www.sciencedirect.com/science/article/pii/S0896627317305603 — Argues the hippocampus maps memories across space, time, and abstract relations via one generalized organizing mechanism.

6. Automatization — habits, model-based vs. model-free control, action chunking

The procedural face of the same efficiency principle.

  • Daw, N. D., Niv, Y., & Dayan, P. (2005). "Uncertainty-based competition between prefrontal and dorsolateral striatal systems for behavioral control." Nature Neuroscience 8:1704–1711. https://www.nature.com/articles/nn1560 — The formal bridge: a flexible model-based (prefrontal) controller and a cheap model-free (dorsolateral striatal) controller, arbitrated by which is more accurate given uncertainty. Model-based control ≈ using a cognitive map.

  • Dolan, R. J., & Dayan, P. (2013). "Goals and habits in the brain." Neuron 80:312–325. https://doi.org/10.1016/j.neuron.2013.09.007 — Definitive review of the goal-directed/habitual (model-based/model-free) distinction and the classic "driving to your old house" illustration of automatized control.

  • A tutorial-style primer (2020). "Model-based decision making and model-free learning." Current Biology 30:R860–R865. https://www.sciencedirect.com/science/article/pii/S0960982220309039 — Accessible walkthrough of outcome-devaluation logic and the dorsomedial-vs-dorsolateral striatal dissociation.

  • "Model-based learning protects against forming habits" (2015).Cognitive, Affective, & Behavioral Neuroscience 15:523–536. https://link.springer.com/article/10.3758/s13415-015-0347-6 — In humans, stronger model-based control is associated with reduced slipping into rigid stimulus–response habits.

  • Cortico-striatal compartments model (2023). "Distinct cortico-striatal compartments drive competition between adaptive and automatized behavior." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10030038/ — Computational model of how dorsomedial (goal-directed) and dorsolateral (habitual) striatum compete, and how weakened prefrontal control tips behavior toward rigidity.

  • Model-based control in psychiatry (2017). "Model-Based Control in Dimensional Psychiatry." Biological Psychiatry.https://www.sciencedirect.com/science/article/abs/pii/S0006322317314816 — Frames compulsion across disorders as an imbalance favoring model-free/habitual over model-based control — the pathological edge of automatization.

  • Parkinson's & habit control (2021). "Impaired Formation and Expression of Goal-Directed and Habitual Control in Parkinson's Disease." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8574955/ — Dopamine loss in sensorimotor striatum disrupts the normal goal-directed↔habitual balance, linking the framework to disease.

  • Graybiel, A. M. (1998). "The basal ganglia and chunking of action repertoires." Neurobiology of Learning and Memory 70:119–136. https://pubmed.ncbi.nlm.nih.gov/9753592/ — Proposes the striatum "chunks" action sequences into performance units, explicitly generalizing Miller's information-chunking idea to action control.

  • Smith, K. S., & Graybiel, A. M. — "task bracketing." QnAs / review: https://www.pnas.org/doi/10.1073/pnas.1315012110 — As behavior becomes habitual, sensorimotor striatal neurons fire mainly at the start and end of a routine, "bracketing" it as one packaged chunk.

  • Graybiel, A. M., & Grafton, S. T. (2015). "The Striatum: Where Skills and Habits Meet." Cold Spring Harbor Perspectives in Biology.https://pmc.ncbi.nlm.nih.gov/articles/PMC4526748/ — Reviews how skills and habits share striatal machinery, including the intrastriatal reorganization behind task-bracketing.

  • Martiros, N., Burgess, A. A., & Graybiel, A. M. (2018). "Inversely Active Striatal Projection Neurons and Interneurons Selectively Delimit Useful Behavioral Sequences." Current Biology. Explainer: https://mcgovern.mit.edu/2018/02/08/distinctive-brain-pattern-helps-habits-form/ — Identified specific striatal neurons that mark the beginning and end of chunked, learned routines.

  • Sensory compression & chunking (2025). "Sensory Compression as a Unifying Principle for Action Chunking and Time Coding in the Brain." bioRxiv.https://www.biorxiv.org/content/10.1101/2025.09.05.674585 — Recent proposal that compression links stereotyped action chunks and elapsed-time coding in the same dorsolateral-striatal circuits.

7. Perception and prediction — "how we see and move through the world"

The perceptual face: schemas as priors in a predictive brain.

  • Bar, M. (2007). "The proactive brain: using analogies and associations to generate predictions." Trends in Cognitive Sciences 11:280–289. https://doi.org/10.1016/j.tics.2007.05.005 — Argues the brain is fundamentally predictive: it uses stored associations/schemas to anticipate incoming input rather than passively receiving it.

  • Clark, A. (2013). "Whatever next? Predictive brains, situated agents, and the future of cognitive science." Behavioral and Brain Sciences 36:181–204. (DOI 10.1017/S0140525X12000477.) — The influential statement of predictive processing: perception is the brain testing top-down predictions (priors/schemas) against sensory evidence, mostly processing the mismatch.

  • Friston, K. (2010). "The free-energy principle: a unified brain theory?" Nature Reviews Neuroscience 11:127–138. (DOI 10.1038/nrn2787.) — The mathematical parent framework (active inference): both perception and action work to minimize prediction error — action makes predicted sensory states come true, which is one way to understand automatized, skilled movement.

  • Schemas/scripts (Themes 1–5) and habits/automatization (Themes 6–7) are ways we assess how the brain extracts regularities, caches them as reusable structure, and runs that structure automatically to avoid recomputing from scratch.

Intro Texts From Class and Misc:

Intro to Affordances: https://ixdf.org/literature/book/the-encyclopedia-of-human-computer-interaction-2nd-ed/affordances#423_theoretical_roots-2

AFFORDANCE and Design by Don Norman: https://jnd.org/affordances-and-design/

J.J. Gibson’s The Theory of Affordances (the original text)

Ch 8 of Ecological Approach to Visual Perception https://monoskop.org/images/c/c6/Gibson_James_J_1977_1979_The_Theory_of_Affordances.pdf

Even more clarification if you really want it

https://graphicsinterface.org/wp-content/uploads/gi2000-24.pdf

https://media.pluto.psy.uconn.edu/MC.pdf

A very cool and influential but maybe not so easy paper from some friends in Amsterdam about Landscape and Affordances

Hidden Brain episode about mental mapping (Thanks Amanda)

Thought Process by JR in New YorkerDiaries, Talks, Creations: Here is playlist of messy attempts and conversations and experimenting towards this approach and articulating it across the usual bounds

Here are about 90 more talks on the subjects here form different philosophical POVs.