Here’s the short version, then details and sources:
- What today’s science assumes: A mind-independent world exists (scientific realism), and the mind is what the brain does (physicalism). Direct “naïve” realism is out; perception is a brain-built model constrained by the senses (often framed as predictive processing). Dualism and global idealism have no empirical support.
- Most consistent with current evidence: Scientific realism + physicalism + an indirect/representational (predictive) account of perception. Dualism and (global) idealism are not favored by contemporary neuroscience or cognitive science.
Where each view stands with current science
Direct vs. indirect realism (how we perceive)
- Modern vision and neuroscience largely support indirect/representational realism: the brain actively predicts and models the causes of sensory inputs; perception is a controlled, error-corrected “best guess.” See predictive coding / free-energy work and popular expositions (“perception as a controlled hallucination”).
Links:- Friston, “Predictive coding under the free-energy principle” (2009): https://pmc.ncbi.nlm.nih.gov/articles/PMC2666703/ (PMC)
- Friston, “The free-energy principle: a unified brain theory?” (2010): https://www.nature.com/articles/nrn2787 (Nature)
- Anil Seth interviews/essays on “controlled hallucination” (overview, not primary data): https://lab.cccb.org/en/anil-seth-reality-is-a-controlled-hallucination/ (CCCB LAB)
- There are enactive/embodied proposals emphasizing action and sensorimotor skills (sometimes read as a kind of “directness”), but these are minority within empirical neuroscience and compatible with the idea that internal models guide action.
Links:- Alva Noë, Action in Perception (book overview): https://mitpress.mit.edu/9780262140881/action-in-perception/ (MIT Press)
- Critical/overview pieces: https://ndpr.nd.edu/reviews/action-in-perception/ (ndpr.nd.edu)
Dualism (mind and brain are distinct substances)
- Evidence goes the other way: mental states covary with and are altered by brain states (lesions, neurodegeneration, anesthesia), split-brain data show mind functions track neural integration, and decoding studies now recover semantic content from fMRI. All of this supports mind–brain dependence and weighs against interactionist dualism.
Links:- Split-brain review: https://pmc.ncbi.nlm.nih.gov/articles/PMC7305066/ (PMC)
- 50 years of split-brain research (overview): https://academic.oup.com/brain/article/140/7/2051/3892700(OUP Academic)
- Decoding continuous language from fMRI: https://www.nature.com/articles/s41593-023-01304-9 and NIH summary https://www.nih.gov/news-events/nih-research-matters/brain-decoder-turns-persons-brain-activity-into-words (Nature)
- Background encyclopedia on dualism (for definitions, not empirical support): https://plato.stanford.edu/entries/dualism/ (Stanford Encyclopedia of Philosophy)
Idealism (reality is fundamentally mental)
- As a metaphysical thesis, idealism isn’t testable in a way current neuroscience can adjudicate. Empirically, cognitive science presupposes a mind-independent external world that causes signals brains model. Predictive processing and adversarial tests of consciousness theories engage how the brain generates conscious contents, not support for idealism.
Links:- Scientific realism overview: https://plato.stanford.edu/entries/scientific-realism/ (Stanford Encyclopedia of Philosophy)
- For consciousness-theory testing (GNW vs. IIT) that illustrates current empirical direction: Nature adversarial collaboration: https://www.nature.com/articles/s41586-025-08888-1 (with editorial context: https://www.nature.com/articles/d41586-025-01379-3) (Nature)
Bottom line (which is “most likely,” by today’s evidence)?
- Most favored: Scientific realism + physicalism (mind depends on brain) with indirect/representational perception (predictive processing). This package best explains lesion/anatomy/EEG/fMRI/anesthesia data, illusions, split-brain phenomena, and successful brain-decoding results. (PMC)
- Less favored: Dualism lacks positive empirical support and conflicts with convergent neuroscientific evidence of mind–brain dependence. (PMC)
- Not supported empirically: (Global) idealism isn’t corroborated by the scientific method as practiced; it sits outside current testable frameworks. Contemporary debates in consciousness (e.g., GNW vs. IIT) assume physical brains implementing mechanisms; recent large multi-lab tests compare such physical theories rather than endorsing idealism. (Nature)
If you want, I can map these positions onto specific empirical findings (illusions, predictive feedback loops, anesthesia signatures, lesion syndromes) and note where philosophers still disagree about interpretation even while sharing the same data.
Here’s the idea in plain, senior-year terms—using three pieces that fit together:
1) Scientific realism (what the world is like)
- Claim: There really is a world “out there” whether or not anyone is looking at it—atoms, stars, viruses, rocks, other people.
- Science works because our best theories describe how that real world behaves (imperfectly at first, better over time). Think of science as building maps that get closer and closer to the actual terrain.
2) Physicalism (what minds are made of)
- Claim: Your mind is what your brain and body do. Thoughts, feelings, memories—these are processes happening in physical stuff (neurons, chemicals, signals).
- Evidence vibe: change the brain and the mind changes (injury, drugs, anesthesia, development). So mental life depends on physical mechanisms, not a separate non-physical “mind substance.”
3) Indirect/representational perception, via prediction (how you experience)
- Claim: You don’t experience the world raw. Your brain builds a model of what’s out there and constantly predicts what your senses should be reporting. Then it compares the prediction with the actual signals (eyes, ears, skin) and fixes the errors.
- Everyday analogy:
- VR headset: You “see” a world created by a computer model, updated as you move your head and hands. Your brain is doing a version of this all the time—except the “computer” is your neural circuitry, and the updates come from your senses.
- Auto-correct for perception: The brain guesses the next letter (what’s likely out there), then corrects if the sensory “keystroke” disagrees. Fast, efficient, usually accurate.
- Why this isn’t “it’s all fake”: The model is tied to reality by constant error-checking from the senses and by the fact that acting on the model has to work (catching a ball, driving, reading). It’s indirect (you experience the brain’s model) but realist (the model aims to match a mind-independent world).
How the three fit
- Scientific realism says there’s a real world.
- Physicalism says brains (physical systems in that real world) generate minds.
- Predictive perception says those brains know the world by modeling it and correcting themselves with sensory evidence.
Put simply: There’s a real world. Your brain is a physical device in that world. It lets you experience the world by running a smart, self-correcting simulation that’s constantly checked against your senses.
Why scientists like this package
- It explains illusions (the model can be tricked), expertise (better predictions), and why brain changes change experience.
- It powers tech like brain decoding and good AI perception: modeling + error correction works in practice.
Quick contrasts (to avoid confusion)
- Naïve/direct realism (“we see the world exactly as it is”): too simple—illusions and blind spots show the brain is constructing what you see.
- Dualism (“mind is non-physical”): doesn’t fit well with the tight link between brain changes and experience.
- Global idealism (“it’s all mind”): not how science operates; experiments assume a world that pushes back no matter what we believe.
One-sentence summary
You live in a real world, your mind is your brain at work, and your experience is your brain’s best, constantly corrected prediction of what that real world is doing right now.
Here’s how those views are usually classified—and what their nearest epistemology labels are.
| View | Main branch it belongs to | Epistemology classification/labels | One-line gloss | Typical contrasts |
|---|---|---|---|---|
| Scientific realism | Philosophy of science (with metaphysical implications) | Realism about unobservables, truth-as-correspondence, explanationism / IBE(inference to the best explanation), fallibilism | Our best mature theories are (approximately) true descriptions of a mind-independent world, including unobservables; we’re justified by their explanatory/predictive success. | Scientific anti-realism, instrumentalism, constructive empiricism |
| Physicalism | Metaphysics & philosophy of mind | Not an epistemic thesis by itself, but typically paired with naturalized epistemology and externalist accounts of knowledge such as reliabilism | Everything (including minds) is physical; in epistemology this often pushes toward explaining knowledge in scientific/causal terms rather than a priori. | Dualism, idealism, neutral monism |
| Indirect/representational (predictive) perception | Epistemology & philosophy of perception | Indirect realism (aka representationalism), intentionalism about perceptual content, fallibilist perceptual justification; often externalist (reliabilist) | We experience the world via brain-constructed representations/models constrained by sensory input; perception is trustworthy because the process is reliable, not because it is “raw access.” | Naïve/direct realism, phenomenalism, some forms of disjunctivism |
Notes:
- If you want a single umbrella label for the whole package, philosophers often call it scientific realism plus naturalized, externalist epistemology with a representational theory of perception.
- “Predictive processing” is a theory of perception/cognition that fits representationalism/indirect realism; epistemologically it supports externalist reliabilism and fallibilism (perception is generally reliable but revisable).