Summary, Terminology and Practice (DP IB Theory of Knowledge): Revision Note

Scope

The natural sciences aim to explain and predict patterns in the natural world using observation and testing. 

Scientific knowledge includes laws (descriptions of consistent relationships), theories (explanations that unify evidence and generate predictions) and models (simplified representations used when systems are too complex to test directly). 

Scope is limited to questions that can be answered using evidence from the natural world, so science cannot settle questions of value, aesthetics, or the supernatural. 

Practical constraints (cost, technology and ethics) also limit what can be tested and what can be known scientifically at a given time.

A lab team can test the claim “this fertiliser increases plant growth” by setting up controlled trials, measuring biomass and repeating the experiment to check reliability. But the question “is it morally acceptable to prioritise crop yield over biodiversity?” is not something the natural sciences can answer, because it requires value judgements rather than testable relationships in nature.

What makes a question “scientific” rather than outside the scope of the natural sciences? 

How do practical constraints (ethics, cost, technology) shape the boundaries of what science can know at a given time? 

When models rely on simplifying assumptions, what justifies trusting their predictions as scientific knowledge?

Perspectives

Scientists work within shared paradigms, so evidence is often interpreted through an existing framework rather than from scratch; this can create resistance to abandoning well-established theories. 

Power and culture also play a role in shaping scientific perspectives because they influence which questions get asked/which hypotheses are formed, what and how much funding will be provided, what data is selected for the study, and how scientific knowledge is disseminated and viewed. 

Outside the scientific community, people’s values and identities shape how they interpret scientific knowledge, and whether they accept it or reject it, e.g. knowledge about medical treatments or environmental behaviour.

1) When Uranus’s orbit did not match Newtonian predictions, many scientists assumed Newton was still right and looked for an unseen planet; this perspective led to Neptune being discovered rather than the theory being rejected. 

2) Two people read the same guidance about vaccine risk. One trusts medical institutions and focuses on the overall risk reduction; the other distrusts government and industry, and focuses on rare side effects, deciding to avoid the vaccine despite the broader evidence.

How do paradigms shape what scientists treat as a worthwhile question and how they interpret unexpected results? 

How do funding priorities shape the boundaries of what the natural sciences investigate, and what gets left unknown? 

How do pre-held trust and values influence which parts of scientific messages people accept, doubt or ignore when deciding whether to act on scientific advice?

Methods and tools

In the natural sciences, methods include careful observation, controlled experiments (changing one factor while keeping others constant), accurate measurement, repeating studies to test reliability, and peer review. 

Scientists also use models and simulations when systems are too large or complex to test directly. 

Tools include instruments that extend the senses (e.g. microscopes and electronic sensors), standard units and calibration systems that make measurements comparable, agreed protocols for consistent procedures, statistical tests for analysing data, and publishing systems that allow findings to be scrutinised and replicated.

1) A team repeats another lab’s experiment on enzyme activity using the same protocol and measurement tools. When results match across labs, confidence increases. When they do not, the finding is treated as uncertain: scientists repeat the work, check for differences in protocol, and may redesign the experiment or controls to test which factor explains the mismatch before drawing a conclusion.

2) Researchers use a computer model to forecast how a disease might spread under different assumptions (e.g. contact rates or vaccination levels). The model helps test “what if” scenarios, but its conclusions depend on the quality of the data and the assumptions built into the model.

How do standard units, calibration, and agreed protocols support reliable knowledge across different scientists and labs? 

What role does peer review play in checking scientific knowledge, and what are its limits? 

When models are used instead of direct experiments, what counts as good justification for trusting their predictions?

Ethics

Research ethics protect humans, animals, and the environment, but they also limit what evidence can be collected, so scientists sometimes must rely on observational or indirect evidence that carries more uncertainty than controlled experiments. 

Because scientific knowledge enables technology, scientists share responsibility for anticipating foreseeable harms and building safeguards. 

In public-interest decisions, action may be needed before certainty is possible, so responsible practice means weighing risks against benefits and being transparent about what is still unknown. 

Ethical responsibility also includes fairness in what evidence is produced: if certain groups are underrepresented in research, claims may be overgeneralised and risks or benefits may fall unevenly.

1) A medicine is tested mainly on adult men, then prescribed widely. Later, it turns out the optimal dose and side-effect profile are different for women and for older adults. Poor representation in the research made the evidence less fair and less reliable for everyone, and the harms fall disproportionately on groups who were not properly included.

2) A research team receives industry funding for a study; they publish their results but also disclose the funding and share the data and methods so others can check for bias and attempt replication, supporting public trust.

How do ethical constraints on experiments shape the certainty and scope of scientific knowledge? 

When evidence is incomplete, what makes it ethically responsible to act in the public interest anyway? 

To what extent should scientists be responsible for foreseeable harms from technologies built on their research?

Paradigm

A pattern, model or example that provides a framework of understanding 

Paradigm shift

When new knowledge is uncovered to make previous knowledge or models change 

Pseudoscience

Scientific “knowledge” that can not be verified has not been produced with the rigour of the scientific method, but uses the language and style of scientific knowledge.

Hypothesis

A proposed explanation or starting point, based on limited evidence that can be tested in an investigation

Controlled experiments

Experiments that are performed with carefully regulated variables to provide a standard of comparison for similar experiments with just one differing variable 

The observer effect

The principle that the act of observing a phenomenon changes the phenomenon being observed 

Peer review

The evaluation of academic or scientific work by experts working in the same field 

Scientism

An exaggerated trust in the efficacy of the methods of the natural sciences