Appearance
1.1 — What Psychology Actually Is
Two statements. Read them and decide which one is true.
Opposites attract. Birds of a feather flock together.
Both feel obviously right, which is the first thing worth noticing about your own mind: it will confidently endorse a claim and its exact opposite, depending on which one you happen to be looking at. The second thing worth noticing is that this question has an answer. Decades of research on who actually ends up with whom, in which countries, at what rates, gives a clear result. Similarity wins, and it is not close — people pair up with others close to them in age, education, background, values, political outlook and even height, at rates far beyond chance. The "opposites attract" idea survives because a single vivid couple who bicker charmingly is more memorable than ten thousand quietly compatible ones.
That gap — between what feels obviously true about people and what is actually true about people — is the entire reason this field exists.
The definition, and why each word is in it
Psychology is the study of behaviour and mental processes.
Take it in two halves, because the split is a real argument that took the field about a century to settle.
Behaviour is anything a person or animal does that could in principle be recorded by a camera or a stopwatch: what they say, how long they take, what they choose, where they look, whether they help the stranger who drops the folder. It has one enormous advantage, which is that two observers can agree on it.
Mental processes are the things going on inside: thinking, remembering, feeling, deciding, wanting, imagining. These have the opposite property. They are what you actually care about, and nobody outside your skull can see them directly.
For about forty years in the middle of the twentieth century, one whole school of psychology tried to solve this problem by declaring the second half off-limits — study only what you can see, treat the mind as a sealed box. That project failed, and 1.2 covers how and why. What replaced it is the working compromise the field still runs on: you cannot observe a mental process, but you can make it produce a measurable consequence, and then measure that.
Here is what that looks like in practice. Suppose you want to know whether people process a written word for meaning even when they are trying not to. You cannot open the skull and watch. What you can do is print the word RED in blue ink and ask someone to name the ink colour as fast as they can. They will be reliably slower and make more errors than if the word had been a meaningless string of Xs in blue ink. That delay — a few hundred milliseconds, measurable with an ordinary computer — is your evidence. The meaning of the word got read automatically and then had to be pushed out of the way, and the pushing took time. This is the Stroop effect, first published by John Ridley Stroop in 1935, and it is one of the most reliably reproduced results in the whole of psychology.
That is the whole trick of the field, repeated ten thousand times in ten thousand forms. Something unobservable becomes something countable.
The four problems psychology has that chemistry does not
It is worth being honest early about why this subject is harder than it looks, because those difficulties explain most of the bad psychology you will meet.
The thing being studied knows it is being studied. A hydrogen atom does not behave differently because it has guessed what the experimenter is hoping for. A person does. Participants read the room, work out the expected answer, and often supply it — a pattern researchers call demand characteristics. The fix is to hide the real question, use a cover story, or arrange things so participants cannot tell which group they are in.
People differ enormously. Drop any two stones and they fall at the same rate. Put any two people in the same argument and one walks away and the other throws a plate. So psychological findings are almost always statements about averages and tendencies, not about individuals, and reading them as personal predictions is the single most common way people misuse this subject. A finding that extraverts are happier on average tells you almost nothing about whether the specific quiet person in front of you is happy.
Most of what you would like to test, you are not allowed to test. The clean way to find out whether childhood neglect causes adult anxiety would be to take a thousand infants, randomly assign half of them to be neglected, and check back in thirty years. That is monstrous and illegal, so it is never done. Instead the field has to work with people who were already neglected and compare them to people who were not, which immediately raises the possibility that whatever caused the neglect — poverty, a parent's illness, a violent household — is the real cause of the anxiety. This is the correlation is not causation problem, and in psychology it is not a pedantic footnote. It is the central difficulty of half the field.
The samples have historically been strange. For decades the typical participant was an undergraduate student at a Western university taking part for course credit. In 2010 a team of researchers pointed out that these people are WEIRD — from Western, Educated, Industrialised, Rich and Democratic societies — that they are a few per cent of humanity, and that on a whole range of measures they are the unusual ones rather than the representative ones. On some visual illusions, on how much people defer to a group, on what counts as fair in a bargaining game, the answers differ substantially between an American student and a farmer in rural Karnataka. Where this volume states a finding that may not travel, it says so.
The five levels, and why they are not competitors
Ask why someone lashed out at their partner and you can answer at five different heights. Every one of them can be simultaneously true.
Biological. Sleep debt, alcohol, a stress hormone response that has been running high for weeks, a genetic tendency towards a short fuse.
Cognitive. They interpreted a neutral remark as an accusation, because they were already primed to expect one.
Behavioural. Shouting has worked before. It ends the argument quickly. A behaviour that gets rewarded gets repeated.
Social. The argument happened in front of a relative, so backing down carried a cost it would not have carried in private.
Cultural and developmental. They grew up in a house where this was how conflict was conducted, and nobody ever showed them another way.
Beginners treat these as rival explanations and argue about which one is the cause. Professionals treat them as five instruments pointed at the same event. This matters practically: the level you explain a problem at determines the level you try to fix it at, and most stuck problems are stuck because someone is working at the wrong level. Endless conversation about a partner's childhood will not fix a temper that is mostly caused by four hours of sleep a night.
Who is who, and who you would actually pay
This confuses almost everyone, and since one aim of this volume is that you understand your own mind well enough to look after most of it yourself, it should be clear from the start what the professions actually are.
| Title | Training | Can prescribe | Mostly does |
|---|---|---|---|
| Psychiatrist | Medical degree, then psychiatry | Yes | Diagnosis, medication, severe illness |
| Clinical psychologist | Psychology doctorate or MPhil | No (nearly everywhere) | Assessment, structured therapy |
| Counselling psychologist | Postgraduate in counselling psychology | No | Talking therapy for life difficulties |
| Psychotherapist / counsellor | Varies widely by country | No | Talking therapy; the least regulated title |
| Psychoanalyst | Long training in one specific tradition | No | Intensive long-term depth work |
| Research psychologist | PhD, no clinical licence | No | Studies, teaching, no patients |
Two practical points come out of that table.
A psychiatrist is a doctor and a psychologist is not. If medication is the question — an antidepressant, a mood stabiliser, an antipsychotic — a psychiatrist is who decides that, and no amount of reading substitutes for it.
"Therapist" and "counsellor" are the loosest words in the list. In many countries, including India, anybody can print those words on a card. When Part 15 gets to choosing someone, it gives you the specific questions to ask about qualification and registration, because the title alone tells you nothing.
What this field can honestly tell you, and what it cannot
It can tell you what tends to happen. Under these conditions, most people do this, at roughly this rate. That is genuinely valuable — it is how you predict that a public commitment will make you more likely to follow through, or that an unpredictable reward schedule will make a behaviour extremely hard to stop.
It can tell you what does not work, which is often more useful. Venting anger by hitting something does not reduce anger and reliably increases it. Teaching to a student's supposed "learning style" does not improve their learning. Most one-session workshops change nothing at all a month later.
It cannot tell you what you specifically will do, and it cannot tell you what you should want. A finding is a probability distribution, and you are one sample from it. Whenever you read "studies show that people who do X are happier", the honest translation is "in the groups studied, the average score of the X group was somewhat higher, with the two distributions overlapping heavily." That sentence is far less exciting and far more accurate.
What to do with this
Four questions, and they will carry you through the rest of this volume and through every psychology claim you meet outside it. Get into the habit of running them automatically.
One: what was actually measured? Not the headline word — the number. "Happiness" in a study is usually a self-rating from one to seven, given at a single moment, by someone who knows what the researcher wants. "Aggression" is often how much hot sauce a participant chooses to put in a stranger's cup. These are reasonable stand-ins, and they are not the thing itself. Knowing what was really counted deflates most overclaiming instantly.
Two: compared with what? A claim with no comparison group is not a finding. "Ninety per cent of people who did the programme improved" is meaningless until you know how many improved without it, because a great many problems improve on their own with time.
Three: how big is the effect? Not whether it exists — how much of it there is. A difference can be real, statistically respectable, and far too small to matter to your life. 1.3 gives you the specific numbers to look for.
Four: who was in the sample, and would I have been? Fifty psychology undergraduates in Ohio in 1998 is a real result about fifty psychology undergraduates in Ohio in 1998. It might generalise to you. It might not, and the page should tell you which.
Apply those four to the claim that opened this page and you will see why it is settled. The measure was who actually married whom — recorded, not self-reported. The comparison was against chance pairing. The effect is large and consistent. And the samples span dozens of countries. That is what a solid finding looks like, and by the end of Part 1 you will be able to spot one at a glance.
Next: 1.2 walks the history — a German laboratory in 1879, a Viennese consulting room, forty years of rats in boxes, and the moment in the 1950s when the mind was allowed back into the science.