Philip Johnson-Laird’s contributions to cognitive psychology have focused on the psychology of reasoning, language, music and emotion. His early work showed that people make systematic errors in reasoning and that the word order and content of the premises has a large effect on the form of their conclusions — a result in conflict with the view that they employ formal rules of inference.
He developed the idea that the lexical meanings of words could be treated as computational procedures, executed in the process of understanding discourse — an idea that led to series of experiments on how people make inferences. The results of these experiments strongly support the conjecture that linguistic comprehension consists in the construction of mental models, and reasoning in the manipulation of these models.
Philip’s studies of music have led to computational models of musical improvisation, of what causes chords to be consonant or dissonant, and of how music communicates emotions. His current investigations focus on how individuals infer the probabilities of events, and how to integrate probabilistic reasoning with deduction.