Lecture 15: The Structure of Scientific Revolutions

The Structure of Scientific Revolutions

Thomas Kuhn (1962)

Understanding Scientific Change & Progress

Historical image of Thomas Kuhn

Legacy

• One of the most influential books about science ever written
• Revolutionized how we think about scientific progress
• Key relevance for psychology and cognitive science today

Key Concepts We’ll Cover

• Normal Science & Paradigms
• Scientific Revolutions & Crisis
• Incommensurability
• Case Studies from History of Science
• Applications to Psychology
• Modern Relevance (AI Revolution)

What is Normal Science?

• Scientists working within an accepted framework
• “Puzzle-solving” activities
• Based on shared assumptions and methods
• Example: Current cognitive psychology research paradigm
• Textbooks play crucial role in maintaining normal science

Paradigms: The Core Concept

A paradigm is:

• Shared worldview within scientific community
• Set of exemplary solutions/methods
• Framework for what counts as legitimate problems
• Not just theory - includes practices, tools, values

Paradigms

“Scientists work from models acquired through education & subsequent exposure to the literature often without quite knowing or needing to know what characteristics have given these models the status of community paradigms”

Diagram

Diagram showing paradigm components

Case Study: The Copernican Revolution

Scientific Crisis

• Accumulation of anomalies
• Existing paradigm cannot explain new findings
• Loss of confidence in current methods
• Proliferation of competing theories
• Example in Psychology: Behaviorism crisis

Scientific Revolutions

• Paradigm shift occurs
• “Scientific revolution” = complete perspective change
• Not just new knowledge - new way of seeing
• Cannot be reduced to rule-following
• Involves conversion experience

Incommensurability

• Different paradigms cannot be directly compared
• Use different concepts, see different things
• Communication difficulties between paradigms
• No neutral observation language
• Example: Behaviorist vs Cognitive descriptions

Historical Case Study: Chemistry Revolution

• Phlogiston theory vs Oxygen theory
• Lavoisier’s new paradigm
• Changed basic concepts of element/compound
• Transformed experimental practices
• Parallel to modern psychology shifts?

Psychology’s Paradigm Shifts: A Deeper Look

1. Introspectionism (1879-1920s)

• Wundt’s experimental psychology
• Focus on conscious experience
• Structured self-observation methods
• Titchener’s systematic introspection
• Crisis: Unreliable results, lack of replication

2. Behaviorism (1920s-1960s)

• Watson’s rejection of consciousness
• Observable behavior only
• Stimulus-response mechanisms
• Skinner’s radical behaviorism
• Crisis: Cannot explain language, thought, memory

The Behaviorist Paradigm in Detail

Key Assumptions

• Mind is a “black box”
• Learning through conditioning
• Environmental determinism
• Rejection of mental states

Methods

• Animal studies
• Conditioning experiments
• Controlled environment
• Quantitative measurement

The Cognitive Revolution (1950s-1960s)

Catalysts

• Chomsky’s critique of Skinner
• Information theory development
• Computer science emergence
• Memory research breakthroughs

Key Figures

• Jerome Bruner
• George Miller
• Ulric Neisser
• Donald Broadbent

An exciting amalgam

Early computer and cognitive science

Cognitive Psychology Paradigm

Core Metaphor: Mind as Computer

• Information processing
• Memory storage and retrieval
• Mental representations
• Computational procedures

Methods

• Reaction time studies
• Protocol analysis
• Experimental tasks
• Computer modeling

Neuroscience Revolution (1990s-2000s)

Enabling Technologies

• fMRI imaging
• EEG advances
• Optogenetics
• Neural recording

New Understanding

• Brain networks
• Neural correlates
• Plasticity
• Embodied cognition

Barely imaginable without this

Brain imaging technology

Current AI Revolution

Key Developments

• Deep learning success
• GPT language models
• Computer vision advances
• Reinforcement learning

Challenges to Traditional Views

• Emergence vs representation
• Distributed vs modular processing
• Learning without explicit rules
• Scale of computation

AI Impact on Psychological Theory

Traditional Cognitive Models

• Rule-based processing
• Symbolic representations
• Modular systems
• Sequential processing

AI-Inspired Models

• Parallel distributed processing
• Emergent properties
• Statistical learning
• Dynamic systems

Neural Attention Mechanisms

Neural network architecture

Current Debates in AI-Psychology

1. Consciousness

• Integrated Information Theory
• Global Workspace Theory
• Predictive Processing

2. Intelligence

• Multiple intelligences
• Artificial vs natural
• General vs specific

3. Learning

• Deep learning principles
• Transfer capabilities
• Role of innate structure

Evidence of Paradigm Crisis?

Current Anomalies

• Replication failures
• Limited ecological validity
• Theory-practice gap
• Integration problems

Emerging Questions

• Role of consciousness
• Nature of intelligence
• Boundaries of cognition
• Human-AI interaction

Future Integration?

Potential Synthesis

• Computational neuroscience
• Cognitive architectures
• Hybrid AI systems
• Embodied robotics

Research Directions

• Brain-computer interfaces
• Augmented cognition
• Social-technical systems
• Collective intelligence

Key Critics & Alternative Views

Philosophy of Science after Kuhn

• Kuhn’s work sparked major debates
• Three key alternative perspectives:
  • Karl Popper: Falsificationism
  • Imre Lakatos: Research Programs
  • Paul Feyerabend: Epistemological Anarchism

“Critical!”

Philosophy debate illustration

Karl Popper’s Falsificationism

• Science advances through bold conjectures and refutations
• No amount of evidence can prove a theory true
• But one counterexample can prove it false
• Focus on crucial experiments that could falsify theories
• Criticized Kuhn’s “normal science” as too dogmatic

Extraordinary claims require extraordinary evidence

Popper portrait and falsification diagram

Popper vs Kuhn

Popper

• Science should actively seek falsification
• Clear demarcation between science and non-science
• Progress through rational criticism

Kuhn

• Scientists typically defend paradigms
• No clear demarcation criteria
• Progress through revolution and conversion

Imre Lakatos: Research Programs

• Attempted synthesis between Kuhn and Popper
• Science organized into “Research Programs” with:
  • “Hard core” of fundamental assumptions
  • “Protective belt” of auxiliary hypotheses
  • Positive and negative heuristics

Imre Lakatos

Progressive vs Degenerating Programs

Paul Feyerabend: Against Method

• Rejected both Kuhn and Popper
• “Anything goes” in scientific method
• Science shouldn’t be privileged over other forms of knowledge
• Progress requires breaking all rules
• Advocated theoretical anarchism

Punk Paul

Feyerabend portrait

Feyerabend’s Critique of Kuhn

• Paradigms can and should coexist
• No single scientific method
• Progress through proliferation of theories
• Science needs external criticism
• Democracy over expert authority

Thank You

Questions & Discussion

“The successive transition from one paradigm to another via revolution is the usual developmental pattern of mature science.” - Thomas Kuhn

Research Methods Lecture 15 - Kuhn