PSYCHOLOGY has failed to produce a coherent or consistent explanation for many of the most fundamental aspects of human behaviour.

Let’s start with intelligence; psychologists are proud of being able to measure this. However, IQ tests are still extremely controversial, and, despite the fact that they are 100 years old, we seem to be no nearer a definition of what intelligence is. No one can agree whether it is one thing or several, whether it is predominantly inherited or learned, or whether the tests discriminate in favour of middle-class children.

Personality is in a worse state. At least IQ tests can predict how well people will succeed in certain walks of life. But, according to Oliver James, one of our leading psychologists: “Contrary to the misleading claims in the glossy brochures of the companies that tout personality tests, they have not been proven to predict winners any better than chance. When followed up years later to see if tests successfully anticipated which candidates would succeed in a job, it turned out that you might as well have put the names in a hat.”

Today, the so-called “primary personality types”, developed as a result of paper-and-pencil tests, are as follows: extraversion/introversion; conscientiousness/disregard for civility; agreeableness/irritability; neuroticism/emotional stability; openness to experience/lack of intellectual curiosity. However, by no means all of these dimensions immediately make sense in evolutionary terms (why should extraversion be a better adaptive strategy than introversion?), — nor do they hang together conceptually. Again, what exactly is personality? We have next to no idea how intelligence and personality are manifested in the neurons of the brain.

Schizophrenia and dreams are two other areas that are a mess conceptually. The public understands the meaning of schizophrenia as a “split personality”. At one stage, the favoured theory was that the “schizophrenogenic” family caused this illness, in particular families where the mother put the child in a “double bind” situation, sending conflicting messages — a bit like a severe form of Miss Jean Brodie’s disease: “Do as I say and not as I do!”

Since then, the genetic evidence for schizophrenia has accumulated to the point where, in May 2000, scientists in Edinburgh announced that they had identified sites on two genes that were partially responsible. No less important is the accumulating evidence that schizophrenics have fewer cells in certain parts of the brain (the medial dorsal nucleus), which impairs the coordination of speech, hearing and the ability to plan. The crucial point is that schizophrenia is emerging as an illness of thought and judgment and has nothing to do with the personality, “split” or otherwise. It has a biological aetiology not a psychological one.

The situation regarding dreams is no better. They have no “deep” meaning, as Freud tried to claim. The latest research shows that the biochemistry of the brain changes during sleep and that the “zaniness” of dreams is a by-product of this activity. Some dreams may reflect our current concerns, but the idea that we re-enact our childhood problems in our dreams is absurd.

The biggest philosophical failure is in our understanding of the mind. In a university bookshop recently, I counted 156 titles in the “Philosophy of Mind” section. What was interesting about these books was that they were in the philosophy section but almost all were about psychology — the brain, neuroscience, cognitive studies, consciousness, artificial intelligence. Reading through at least some of these books it is clear that the mind is a much-reduced concept. As the brain has given up more and more of its secrets, mind is used to refer to less and less.

In the latest twist, Colin McGinn, the British psychologist-turned-philosopher, has made a name for himself by arguing that mind — consciousness — will never be explained because we are not equipped by evolution to understand it. Is this the greatest admission of defeat yet?

It is worrying that so many of the fundamental building blocks of psychology remain so vague and unresolved 100 years on.

Last May, two Danish researchers reviewed 114 published studies involving 7,500 patients and 40 conditions, which showed that there was no such thing as “the placebo effect” — the notion that, whatever their illness, a third of patients will improve if they are given a dummy pill and told it is real. Again, a well-known psychological “fact” turns out to be no such thing.

Wherever you turn in psychology, the same — unreliable — picture shows itself. In the 1980s, James Flynn, from the University of Otago in New Zealand, was studying the history of intelligence testing in the military when he found that each new generation of recruits performed better on the same test than previous generations. In the 1990s, he found that this was also true for other types of tests in 20 countries. So far no one has an explanation, but Flynn’s observations imply either that, judged by the standards of today, people in the recent past were morons, or that, judged by the standards of the past, we are geniuses. A particular measurement at one time means something quite different at another. This flatly contradicts common sense. It’s like saying that 10 inches in 1953 is 11 inches now.

If there is one psychological “fact” that is not in dispute, you might think, it is that someone’s relationship with his or her parents in the early years is all-important for later intellectual and emotional development. But the picture is nowhere near so clear-cut.