Chapter 1. Why We Need Cosmic Cognitive Science

Section 1. The Parochialism Problem

• Psychology studies less than 0.00001% of possible minds
• The WEIRD bias and replication crisis
• Why human-centric research has failed

Section 2. The Alien Cognitive Scientist Test

• Defining the test: would aliens recognize this principle?
• Substrate-independence as the gold standard
• Examples of passing vs. failing the test

Section 3. Foundations of Cosmic Cognition

• Cognition as physical computation under constraints
• Information theory as the parent discipline
• Thermodynamics and embodiment
• Optimization theory and control systems

Section 4. How to Read This Book

• The three-tier hierarchy explained
• Mathematical notation and conventions
• Cross-references and disputed territory flags

Chapter 2. Cosmic Laws of Cognition

Section 1. Shannon's Channel Capacity Theorem

• Statement of the theorem: C = B log₂(1 + S/N)
• Derivation from mutual information maximization
• Physical inevitability: why no system can exceed this
• Evidence from biological systems: retinal ganglion cells
• Evidence from artificial systems: wireless communication
• Violation costs: catastrophic error accumulation

Section 2. Nyquist-Shannon Sampling Theorem

• Statement: sample rate ≥ 2× highest frequency
• Mathematical proof via Fourier analysis
• Aliasing as inevitable consequence of undersampling
• Housefly compound eye implementation
• Human visual system temporal sampling
• Robotic vision failures from undersampling

Section 3. Weber-Fechner and Stevens Power Laws

• Weber's Law: ΔI/I = constant
• Fechner's integration: S = k log(I)
• Stevens' generalization: S = kI^α
• Information-theoretic derivation from efficient coding
• Evidence from bacterial chemotaxis
• Evidence from insect sensory systems
• Evidence from octopus chromatophore control
• Evidence from human psychophysics
• Robotic sensor implementations

Section 4. Fitts's Law

• Statement: MT = a + b log₂(2D/W)
• Information-theoretic derivation
• Why this is inevitable for any aimed movement
• Evidence from human motor control
• Evidence from elephant trunk reaching
• Evidence from octopus arm targeting
• Evidence from industrial robotics
• Zero confirmed violations in 70 years

Section 5. Hick-Hyman Law

• Statement: RT = a + b log₂(n + 1)
• Derivation from entropy of choice set
• Evidence from honeybee foraging decisions
• Evidence from rat maze navigation
• Evidence from human reaction time studies
• Relationship to information transmission limits

Section 6. Predictive Coding and Free-Energy Principle

• Variational free energy definition
• Mathematical proof of optimality for hierarchical inference
• Connection to minimum description length
• Evidence from E. coli chemotaxis
• Evidence from insect olfactory processing
• Evidence from rodent hippocampus
• Evidence from human visual cortex

Section 7. Bayesian Optimal Cue Integration

• Maximum likelihood estimation for Gaussian cues
• Inverse-variance weighting as optimal solution
• Evidence from bacterial multisensory fusion
• Evidence from human vision-touch integration
• Evidence from robotic sensor fusion
• Suboptimality costs: degraded performance metrics

Section 8. Signal Detection Theory

• d-prime and criterion as decision variables
• ROC curves and optimal operating points
• Derivation from Neyman-Pearson lemma
• Evidence across sensory modalities
• Evidence from predator-prey detection systems
• Applications to artificial detection systems

Section 9. Rate-Distortion Theory

• Optimal compression-fidelity tradeoff
• Information bottleneck formulation
• Retinal implementation of rate-distortion curves
• Efficient coding hypothesis (Barlow)
• Sparse coding as compression strategy

Section 10. Marginal Value Theorem

• Charnov's patch-leaving rule
• Mathematical derivation from optimal foraging theory
• Evidence from bumblebee foraging
• Evidence from bird foraging patterns
• Evidence from mammalian resource exploitation
• Evidence from human hunter-gatherer behavior
• Implementation in optimal RL agents

Section 11. Optimal Stopping: The 37% Rule

• Secretary problem formulation
• Mathematical proof of 1/e optimality
• Generalization to unknown horizons
• Evidence from animal mate choice
• Evidence from nest site selection
• Human approximations under time pressure

Chapter 3. Cosmic Heuristics

Section 1. Recognition Heuristic

• Inference rule: recognized > unrecognized
• Ecological rationality conditions
• Evidence from human city-size judgments
• Evidence from artificial agent object search
• When recognition heuristic fails

Section 2. Take-the-Best

• Algorithm: order cues by validity, stop at first discriminator
• Less-is-more effect: mathematical proof
• Comparison to weighted linear models
• Evidence from 20 real-world datasets
• When take-the-best outperforms regression

Section 3. Gaze Heuristic Family

• Constant bearing angle for collision detection
• Constant optical angle for interception
• Geometric proof of optimality
• Evidence from dragonfly prey capture
• Evidence from baseball outfielders
• Evidence from naval collision avoidance

Section 4. Satisficing (Simon)

• Aspiration-level algorithm
• Optimality under search costs and uncertainty
• Evidence from animal habitat selection
• Evidence from human decision making
• Comparison to maximization strategies

Section 5. Equipartition (1/n Rule)

• Equal division when no differentiating information
• Proof of minimax optimality
• Evidence from portfolio allocation
• Evidence from resource distribution
• When 1/n beats optimized strategies

Section 6. Tallying (Unit-Weight Linear Model)

• Count positive cues, ignore weights
• Robustness to estimation error
• Evidence from medical diagnosis
• Evidence from personnel selection
• Comparison to optimally weighted models

Section 7. Tit-for-Tat

• Algorithm: cooperate first, then copy opponent
• Evolutionary stability in iterated prisoner's dilemma
• Evidence from bacterial cooperation
• Evidence from primate reciprocity
• Robustness to noise and defection

Section 8. Lévy Flight Foraging

• Power-law step distribution: P(L) ~ L^(-μ)
• Optimal exponent μ ≈ 2 for sparse resources
• Evidence from jellyfish movement
• Evidence from albatross foraging
• Evidence from mammalian search patterns
• Evidence from human hunter-gatherers
• Controversy: composite Brownian vs. true Lévy

Section 9. Recency Weighting

• Exponential vs. hyperbolic temporal discounting
• Bayes-optimality in non-stationary environments
• Evidence from animal learning curves
• Evidence from human probability estimation
• Adaptive learning rate modulation

Chapter 4. Cosmic Rules

Section 1. Speed-Accuracy Tradeoff

• Optimal decision time balances delay vs. error costs
• Drift-diffusion model derivation
• Race model alternative formulation
• Evidence from insect decision making
• Evidence from human perceptual judgments
• Evidence from artificial sequential sampling

Section 2. Exploration-Exploitation Tradeoff

• Multi-armed bandit problem formulation
• Thompson sampling as Bayes-optimal solution
• UCB1 and other provably optimal algorithms
• Softmax and ε-greedy as biological approximations
• Evidence from animal foraging
• Evidence from human learning tasks
• Implementation in RL agents

Section 3. No-Free-Lunch Theorems

• Wolpert-Macready statement and proof
• Implication: specialization is mandatory
• Why Earth-like priors are not cheating
• Relationship to universal compression limits
• Consequences for cognitive architecture

Section 4. Prediction-Error Priority

• Surprisal as information content
• Only unexpected events teach
• Evidence from attentional orienting
• Evidence from learning-rate modulation
• Implementation in prediction-error neurons

Section 5. Diminishing Marginal Returns

• Value of information falls as 1/n or faster
• Proof from Bayesian updating
• Evidence from sampling behavior
• Evidence from information search tasks
• Connection to satisficing

Section 6. Information Foraging Theory

• Patch-leaving when local rate < global average
• Connection to marginal value theorem
• Evidence from web browsing
• Evidence from library search
• Evidence from scientific literature foraging

Chapter 5. Cosmic Constants and Physical Constraints

Section 1. Landauer's Limit

• Minimum energy per bit erased: kT ln 2
• Thermodynamic proof
• Why biological computation is 10^6× above minimum
• Implications for neural sparsity
• Implications for memory decay

Section 2. Serial Processing Bottleneck

• 1–50 bits/second central limit
• Evidence from human dual-task interference
• Evidence from monkey electrophysiology
• Evidence from artificial global workspace models
• Connection to attentional blink
• Connection to change blindness

Section 3. Power-Law Forgetting

• Retention ∝ t^(-β) with β ≈ 0.3–0.5
• Derivation from superposed exponentials
• Evidence from human memory studies
• Evidence from animal conditioning
• Evidence from artificial neural networks
• Connection to spacing effect

Section 4. Power-Law Practice

• Performance ∝ n^α with α ≈ 0.3–0.6
• Log-log linearity across domains
• Evidence from skill acquisition
• Evidence from perceptual learning
• Neural efficiency explanations

Section 5. Critical Branching (σ ≈ 1)

• Criticality maximizes information capacity
• Neural avalanche distributions
• Evidence from cultured neurons
• Evidence from cortical recordings
• Evidence from reservoir computing
• Subcritical and supercritical failure modes

Section 6. Kleiber's 3/4 Metabolic Scaling

• Metabolic rate ∝ M^0.75
• Geometric derivation from network constraints
• Implications for brain energy costs
• Evidence across species
• Cognitive consequences of scaling

Section 7. Small-World Network Structure

• Path length ≈ ln(N) in efficient networks
• Clustering coefficient requirements
• Evidence from neural connectivity
• Evidence from social networks
• Evidence from artificial architectures

Chapter 6. Biological Instantiations of Cosmic Laws

Section 1. Rescorla-Wagner as Gradient Descent

• ΔV = αβ(λ - ΣV) equation
• Identity with delta rule and backpropagation
• Evidence from honeybee color learning
• Evidence from Aplysia gill withdrawal
• Evidence from rat nucleus accumbens
• Evidence from human fMRI signals
• Rediscovery in deep RL 2022–2025

Section 2. Temporal-Difference Learning and Dopamine

• TD(λ) algorithm formulation
• Dopamine as biological TD-error signal
• 10–100 Hz phasic burst = positive error
• Pause/dip = negative error
• Evidence from zebrafish
• Evidence from songbirds
• Evidence from primates
• Optogenetic replacement experiments

Section 3. Blocking and Overshadowing

• Kamin blocking paradigm (1968)
• Mathematical inevitability from zero prediction error
• Evidence from snails
• Evidence from honeybees
• Evidence from fish
• Evidence from rodents
• Evidence from humans
• Blocking in deep neural networks

Section 4. Successor Representation

• Separation of transition and reward
• SR(λ) formulation
• Hippocampal place cells as successor features
• Orbitofrontal cortex value storage
• Independent discovery in RL (Dayan 1993)
• Evidence from revaluation studies

Section 5. Pavlovian-Instrumental Transfer

• General vs. specific PIT effects
• Nucleus accumbens core/shell distinction
• Evidence from Aplysia
• Evidence from rodents
• Evidence from primates
• Evidence from humans

Section 6. Peak Shift and Super-Stimulus

• Strongest response beyond trained stimulus
• Gradient interaction explanation
• Evidence from pigeon discrimination
• Evidence from insect foraging
• Human aesthetic preferences
• Exploitation in advertising and art

Chapter 7. Biological Heuristics

Section 1. Win-Stay, Lose-Shift

• WSLS algorithm definition
• Optimality in slowly changing environments
• Evidence from bumblebee foraging
• Evidence from rat patch choice
• Evidence from starling food selection
• Human probability matching connection

Section 2. Fluency Heuristic

• Processing ease as validity cue
• Neural efficiency hypothesis
• Evidence from honeybee flower preferences
• Evidence from human truth judgments
• Evidence from aesthetic ratings
• Mere exposure effect connection

Section 3. Affect Heuristic

• Valence-first evaluation architecture
• 100–300 ms advantage over analytic processing
• Amygdala response conservation across vertebrates
• Evidence from fish threat detection
• Evidence from mammalian fear conditioning
• Human risk perception biases

Section 4. Shepard's Universal Law of Generalization

• exp(-cd^p) formulation
• p=1 (city-block) vs. p=2 (Euclidean) metrics
• Evidence from pigeons
• Evidence from rats
• Evidence from humans
• Evidence from neural network models
• Most replicated law in psychology

Section 5. Frequency-Based Probability

• Natural frequency superiority
• Evidence from human Bayesian reasoning
• Evidence from animal probability matching
• Evolutionary argument for frequency formats
• Implications for probability education

Section 6. One-Reason Decision Making

• Single-cue stopping rules
• Evidence from bumblebees choosing flowers
• Evidence from great tits choosing sites
• Time pressure as ecological driver
• Connection to take-the-best

Chapter 8. Biological Rules and Constraints

Section 1. Feature-Positive Effect

• Presence > absence learning speed
• Infinite set problem for absences
• Evidence from pigeons
• Evidence from rats
• Evidence from rabbits
• Evidence from humans

Section 2. Latent Inhibition

• Pre-exposure retards conditioning
• "Already predicted" mechanism
• Evidence from sea slugs
• Evidence from rodents
• Evidence from humans
• Clinical implications (schizophrenia)

Section 3. Pearce-Hall Attention Modulation

• Associability α ∝ |λ - V|
• Surprise-driven attention
• Evidence from rabbit conditioning
• Evidence from rat learning
• Evidence from human attention
• Contrast with Mackintosh model

Section 4. Context-Renewal Effects

• ABA and ABC renewal paradigms
• Extinction as context-gated inhibition
• Evidence from sea slugs
• Evidence from rodent fear conditioning
• Evidence from human anxiety
• Hippocampal and prefrontal mechanisms

Section 5. Rational Volatility Tracking

• Learning rate increases with environmental change
• Optimal inference under non-stationarity
• Evidence from fruit flies
• Evidence from honeybees
• Evidence from fish
• Evidence from rodents
• Evidence from humans

Section 6. Overshadowing and Relative Validity

• Strong cue blocks weak cue
• Compound vs. elemental training
• Evidence from goldfish
• Evidence from pigeons
• Evidence from humans
• Computational accounts

Chapter 9. Biological Constants and Scalings

Section 1. Subitizing Limit 3–4 Items

• Instant enumeration without counting
• Evidence from human adults
• Evidence from human infants
• Evidence from rhesus monkeys
• Evidence from crows
• Evidence from guppies
• Neural capacity explanations

Section 2. Attentional Blink 200–500 ms

• Temporal window for second target detection
• Evidence from human RSVP tasks
• Evidence from macaque electrophysiology
• N2pc and P3 suppression mechanisms
• Individual differences and training

Section 3. Phasic Dopamine Burst Range

• 10–100 Hz firing rate
• Conservation from lamprey to primate
• VTA and SNc characterization
• Tonic vs. phasic modes
• Computational role of burst timing

Section 4. Optimal Learning Rate Window

• 0.01–0.2 sweet spot
• Too low: cannot track change
• Too high: catastrophic forgetting
• Evidence from zebra finch reinforcement
• Evidence from rodent conditioning
• Evidence from human learning
• Evidence from deep RL stability

Section 5. Metabolic Cost of Neural Activity

• ~10^9 ATP per cortical spike
• Energy budget constraints
• Sparsity as energy optimization
• Comparison across nervous systems
• Octopus vertical lobe costs
• Mammalian cortex costs

Section 6. Immediate Serial Recall Span

• 7±2 before chunking (Miller 1956)
• Evidence from verbal span tasks
• Evidence from spatial span tasks
• Evidence across species
• Working memory capacity connection

Chapter 10. Where Humans Deviate from Cosmic Optimality

Section 1. Availability Heuristic as Bug

• Ease of recall vs. true frequency
• Cosmic optimum: full Bayesian integration
• Mass media amplification failures
• Terrorism probability overestimation
• When availability works (small samples)

Section 2. Anchoring and Insufficient Adjustment

• Arbitrary numbers shift estimates 30–50%
• Violation of irrelevance axioms
• Evidence: SSN digit experiments
• Evidence: wheel of fortune studies
• No animal analogs
• Language-dependent mechanism

Section 3. Conjunction Fallacy

• P(A∩B) judged > P(A)
• Linda problem classic demonstration
• Direct violation of probability axioms
• Natural frequency format correction
• Language-dependent origins

Section 4. Base-Rate Neglect

• Representativeness over prior probabilities
• Lawyer/engineer problem
• 95% of humans ignore 90% base rate
• Optimal agents never neglect base rates
• Educational interventions

Section 5. Scope Insensitivity

• Willingness to pay insensitive to magnitude
• 2,000 vs. 200,000 birds equivalence
• Affect-driven vs. scope-sensitive valuation
• Implications for charitable giving
• Policy consequences

Chapter 11. Cultural Amplification of Suboptimal Patterns

Section 1. Framing Effects

• Logically equivalent descriptions, different choices
• 90% fat-free vs. 10% fat
• Gain vs. loss frames (Asian disease problem)
• No animal analogs
• Requires symbolic re-description

Section 2. Narrative Fallacy and Hindsight Bias

• Post-hoc coherent story construction
• "I knew it all along" effect
• Memory distortion mechanisms
• Requires language for counterfactuals
• Financial market illusions

Section 3. Confirmation Bias

• Seeking evidence that supports belief
• Wason selection task failures
• Echo chamber amplification
• When confirmation is rational
• Optimal hypothesis testing contrast

Section 4. Mismatch to Post-Agricultural Environments

• Hyperbolic discounting for 30-year lifespans
• Obesity from sugar/fat super-stimuli
• Debt from temporal myopia
• Climate inaction from distant threats
• Peak-shift exploitation mechanisms

Section 5. Cultural vs. Genetic Evolution Speed

• Norms evolve 10^4–10^6× faster than genes
• Persistent maladaptation consequences
• Prestige bias run amok
• Costly signaling escalation
• Social media novel selection pressures

Chapter 12. Primate-Specific Architectural Kludges

Section 1. Over-Powered Episodic Simulation

• Hippocampal future scenario generation
• Thousands of simulations per day
• Chronic anxiety and rumination costs
• Corvid and ape future planning comparison
• Human dial turned to 11

Section 2. Obligate Theory-of-Mind Module

• 9-month-old false belief tracking
• 65% of conversation is social
• Politics and coalition formation
• Reputation markets and gossip
• Cancel culture mechanisms
• No optimal agent would want this

Section 3. Language-Dependent Cognitive Traps

• Conjunction fallacy disappears in frequencies
• Framing requires symbolic re-description
• Most dual-process biases are linguistic
• Propositional thought side effects
• Non-linguistic reasoning preservation

Section 4. Loss Aversion and Endowment Effect

• Losses loom 2.0–2.5× larger than gains
• Prospect theory S-curve
• Evidence from capuchin monkeys
• Evidence from chimpanzees
• Human ownership narrative amplification
• Status-quo bias connection

Section 5. Recursive Mental States Limit

• "I think that you think that I think..."
• Breaks down at 5–6 levels
• Strategic reasoning consequences
• Literary narrative limits
• Computational complexity explanation

Section 6. Counterfactual Reasoning Scaffolds

• Upward vs. downward counterfactuals
• Regret and relief asymmetries
• Bronze–silver medalist paradox
• Functional vs. normative types
• Language-dependency evidence

Chapter 13. Human-Specific Constants and Scalings

Section 1. Cowan's Magical Number 4±1

• Pure working memory capacity
• Not 7±2 (Miller conflated with rehearsal)
• Evidence from change detection
• Evidence from multiple object tracking
• Crow and parrot comparison (5–6)
• Verbal rehearsal interference

Section 2. Dunbar's Layered Social Channels

• 5 intimate / 15 good friends / 50 / 150 / 500
• Neocortex ratio prediction
• Not a cosmic principle
• Evidence from social network analysis
• Evidence from primate group sizes
• Distributed cognition alternative

Section 3. Phonological Loop Duration

• ~1.8 seconds rehearsal window
• Memory span = words in 1.8s
• Language-specific word length effects
• Articulatory suppression abolishes effect
• Irrelevant for non-speaking minds

Section 4. Semantic Memory Decay Parameters

• Decades-long retention with power-law decay
• Typicality gradient half-lives
• Rosch prototype structure
• 8–12 year decay for unused concepts
• LTP parameter artifact

Section 5. Theory-of-Mind Developmental Timeline

• ~9 months: joint attention
• ~18 months: pretend play
• ~4 years: false-belief understanding
• ~6–7 years: recursive mental states
• Cross-cultural robustness
• Autism as counter-example

Chapter 14. Boundary Conditions: Architecture Space and the Limits of Cosmic Generality

Section 1. The Constraint Set Problem

• Every “cosmic law” is actually “optimal given constraints X, Y, Z”
• Which constraints are truly universal?
• Physical: thermodynamics, speed of light
• Information-theoretic: Shannon limits, compression bounds
• Possibly contingent: embodiment, energy scarcity, sequential processing

Section 2. Sequential vs. Parallel Processing

• The 1–50 bits/sec serial bottleneck revisited
• Is this cosmic or a biological neuron artifact?
• Global workspace theory assumptions
• Thought experiment: perfectly communicating parallel processors
• Arguments for and against an unavoidable bottleneck
• Verdict: unresolved boundary condition

Section 3. Discrete vs. Continuous Time

• Most laws assume discrete sampling (Nyquist, RT models)
• Event-driven architectures as alternative
• Asynchronous biological spiking dynamics
• Spiking vs. rate-coded systems
• Do Hick-Hyman and Fitts's laws apply to continuous dynamics?
• Time discretization: fundamental or implementation detail?

Section 4. Centralized vs. Distributed Control

• Predictive coding assumes hierarchical message passing
• Can flat architectures achieve similar optimality?
• Subsumption architecture (Brooks) as counterexample
• Slime mold computation as biological distributed case
• When centralization is required (credit assignment, coherence)
• Markov blanket perspective on system boundaries

Section 5. Digital vs. Analog Computation

• Landauer’s limit and bit erasure
• Continuous-valued computation and thermodynamic costs
• Precision–energy tradeoffs in analog systems
• Stochastic computing as intermediate regime
• Whether binary representation is universally privileged

Section 6. The Embodiment Question

• How many “cosmic laws” assume embodiment?
• Fitts’s law requires spatial movement
• Weber–Fechner requires physical sensor noise
• Marginal Value Theorem requires spatial foraging
• Disembodied theorem provers as boundary cases
• Pure prediction engines (LLMs) as embodiment-free architectures
• Which laws survive pure information processing?

Section 7. Energy Scarcity as Contingent Factor

• Biological constraints stem from ATP limitations
• Sparsity, forgetting, satisficing as energy-reduction strategies
• Thought experiment: unlimited energy supply
• Would optimal agents still sparsify or forget?
• Finite time vs. infinite energy
• Implications for digital superintelligence

Section 8. The Alien Cognitive Scientist Test Revisited

• Original test: “Would aliens recognize this principle?”
• Refined test: “Would aliens in situation S recognize this?”
• Specifying constraint sets precisely
• Embodied vs. disembodied applicability examples
• Taxonomy of constraint-dependent generalizability

Section 9. What Remains Truly Cosmic

• After removing biology, energy limits, embodiment
• Information theory foundations
• Thermodynamic bounds
• Bayes-optimal inference under uncertainty
• Exploration–exploitation structure
• No-Free-Lunch theorems
• The irreducible core: physics + information

Section 10. Implications for Artificial General Intelligence

• Which principles AI will converge on
• Which constraints AI can escape
• Serial bottleneck: possibly escapable
• Forgetting curves: unnecessary with perfect memory
• Predictive coding: likely convergent for efficiency
• Speed–accuracy tradeoff: persists due to time costs
• Navigating design space across constraint sets

Section 11. Open Questions and Research Directions

• Which laws generalize how far?
• Formal proofs for boundary conditions
• Empirical tests: build systems that violate assumed constraints
• Comparative AI psychology across architectures
• Deriving optimal solutions given arbitrary constraint sets
• Predicting alien cognition from physics