Notes on Scientific Method, Theory, and Neuroplasticity Case (London Cab Drivers)
Observations and Reproducibility
Observation is foundational to science: collecting facts about the world with instruments (e.g., microscope) or with our senses. The reliability of science depends on good observation because the enterprise hinges on observing the world accurately.
Reproducibility of observations: any observer should be able to see the same thing. If you point a telescope at a patch of sky and record what you see, others with decent telescopes should be able to aim there and observe the same phenomenon. Similarly, if you observe a slice of rabbit spleen under a microscope, others with similar equipment should observe the same features.
Practical examples discussed: a telescope on the roof of the building; a microscope used in the second lab. The emphasis is that observations must be shareable and repeatable across observers and instruments.
Induction and the History of the Method
The first modern philosopher to propose a rigorous method for doing science emphasized induction: observing specific facts and building from them to general laws of nature.
Induction basics: collect many specific observations; from these, infer general principles or laws.
The instructor’s teaching approach emphasizes focusing on ideas and processes over personalities, noting that science is a historical, messy, and human enterprise.
Note on historical characters: some figures discussed were not perfect; focus on what they did and how their ideas contributed, not on personal virtue.
Francis Bacon anecdote (illustrated with humor): Bacon reportedly tested chilling meat to prevent decay by placing it in snow, but he died from pneumonia. The lesson is not to idolize individuals but to understand the method and its development.
Core quote on induction (paraphrased): induction draws axioms from sense data and particulars by moving through degrees toward statements of higher generality.
Problem with induction: to carry out induction, you must know which facts to observe; selecting relevant observations is itself a challenge. A famous remark from a scientist in the 19th century warned that observations must be for or against some view to be useful in induction.
The Hypothetico-Deductive Method
Hypothesis-first approach: start with a plausible statement about how some aspect of the world might work.
Multiple working hypotheses: generate more than one hypothesis to avoid attachment to a single view and to reduce bias.
From hypothesis to predictions: a good hypothesis should yield precise predictions that can be deduced.
Example of a hypothesis (creativity highlighted): if ethyl alcohol acts on certain brain centers, then going to a party and consuming a lot of alcohol should produce predictable outcomes (e.g., impaired judgment, dancing, etc.).
Creativity in science: formulating elegant hypotheses and testing them with elegant tests is a creative act comparable to artistic creativity.
Real-world creativity reference: James Jamerson, Motown’s legendary bass player, cited as an example of creative problem-solving, analogous to formulating creative hypotheses.
Testing and Validation: Experiments, Observations, and Models
Testing the hypothesis is essential and can take different forms depending on the field:
Formal experiments where feasible.
Carefully controlled observations when experiments are unethical or impractical (e.g., astronomy).
Modeling (physical or computer simulations) when real-world testing is impractical or dangerous (e.g., predicting airplane wing performance, asteroid impact scenarios).
In fields like astronomy, “experiments” may be impossible (we cannot manipulate stars), so careful observation and modeling are key.
A well-designed experiment aims to change and isolate one or a few variables while holding others constant:
This isolates the causal effect of the variable of interest.
In formulas:
ext{Isolate } X ext{ while holding } Y, Z, ext{ etc. constant}
Real-world examples discussed:
Studying a supernova remnant (M2) via observations and measurements of expansion speed and radiation; direct experimentation is not possible.
Bridge design tested through physical wind tunnel models or computer simulations to compare performance.
Outcome of testing:
If observations match the deduced predictions, the hypothesis is provisionally accepted.
If observations do not match, the hypothesis is rejected or revised and retested.
Some hypotheses may be tweaked and retested; others may be discarded entirely.
Absolute proof is not achievable in science; hypotheses and theories remain provisional and subject to revision.
Theory vs Hypothesis
Hypothesis: a plausible proposition about how some aspect of the world might work.
Theory: a hypothesis that has withstood extensive testing across many different contexts and times; it is robust but not absolutely proven.
Everyday misuse: the word theory is often misinterpreted as an educated guess; in science, a theory is a well-supported framework, not a guess.
Examples of strong theories: law of gravity, existence of atoms, germ theory of disease. They are extremely well tested but still open to revision if new contradictory evidence arises.
The COVID-19 pandemic illustrates science in action: recommendations shifted as new data emerged, showing science as a self-correcting process rather than a fixed set of facts.
Practical stance: scientists are human; errors, biases, and fraud can occur, but the self-correcting nature of science tends to weed out bad findings over time.
The strength of science: not absolute truth, but a process that converges toward reliable understanding and is capable of adapting to new evidence.
The Self-Correcting Nature of Science
Scientists are fallible and diverse in capability and belief, but the method itself corrects errors over time.
Fraud, misinterpretation, and bias can exist, but ongoing testing and replication reduce the impact of bad science.
The iterative process of proposing hypotheses, testing them, and revising or discarding them is the core strength of science.
The COVID example underscores how science evolves with new data rather than clinging to initial claims.
Case Study: London Cab Drivers and Brain Plasticity
The Knowledge: becoming a licensed London taxi driver requires a detailed, mental map of the entire city (roughly 36 square miles in central London with about 25,000 streets) and the ability to find the shortest routes between any two points.
Training duration: typically 3 to 4 years to build and maintain this mental map.
Brain imaging findings:
The hippocampus, a brain region important for memory formation, was measured using scans in cab trainees.
Trainees who completed the program showed a significant increase in hippocampal volume (about a one-third increase) from the start to the end of training.
Dropouts did not show this growth; controls (people who never pursued cab driving) showed no change.
This provides evidence that the brain can structurally adapt (neuroplasticity) in response to demanding cognitive tasks requiring extensive memory formation.
Implications for neuroscience:
The brain can remodel itself and even generate new neurons in some regions (neurogenesis), particularly in the hippocampus.
The old belief that brain cells stop being produced after infancy has been challenged by these and other studies.
Brain plasticity is evidence for how training and experience can alter brain structure and function over time.
Key Takeaways and Synthesis
Observations must be reproducible to be scientifically valid; instruments and shared protocols help ensure this.
Induction builds general laws from specific observations but faces the problem of selecting the right observations and the risk of incorrect generalization.
The hypothetico-deductive method provides a robust framework: formulate hypotheses, derive testable predictions, test via experiments/observations/models, and revise or discard based on results.
The term theory denotes well-supported, broad, and repeatedly tested explanations, not an unsupported guess. Absolute proof is unattainable in science; reliability and robustness come from extensive testing and ongoing scrutiny.
Science is a human, fallible, self-correcting enterprise. Its strength lies in its willingness to revise ideas when new evidence challenges established views.
Real-world examples—such as the London cab driver hippocampus study—illustrate how learning and experience can physically reshape the brain, emphasizing neural plasticity as a fundamental aspect of biological science.
Definitions and Quick References
Observation: systematic recording of phenomena, using instruments or senses.
Reproducibility: others can observe the same results under the same conditions.
Induction: deriving general laws from specific observations.
Hypothesis: a testable explanation or educated guess about how something works.
Deduction: deriving specific predictions from a general rule.
Hypothetico-Deductive Method: hypothesis, deduced predictions, testing, revision.
Model: a simplified representation (physical or computational) used to test hypotheses when direct experimentation is impractical.
Theory: a broad, well-supported framework that explains a wide range of phenomena; not absolute proof.
Neuroplasticity: the brain's ability to reorganize itself by forming new neural connections.
The Knowledge (London cab drivers): extensive mental map of London required for navigation; training yields measurable hippocampal growth in successful trainees.
ext{Hypothesis-based predictions: } H
ightarrow (P1 \land P2 \land \dots \land Pn) ext{If any } \neg Pi \text{ is observed, } H \text{ is falsified or revised.}
ext{Isolate } X \text{ while holding other variables constant in an experiment.}
ext{Traction example: } \Delta H \approx +33\% \text{ hippocampal volume increase in trainees.}