Philip Nicholas Johnson-Laird
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Philip Nicholas Johnson-Laird has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Marco Ragni
Technische Universität Chemnitz
7 shared publications Paolo Legrenzi
University of Ca’ Foscari, Venice, Italy
6 shared publications- Sangeet Khemlani
US Naval Research Laboratory Washington 20175 DC USA
6 shared publications - Ruth M. J. Byrne
University of Dublin
5 shared publications Vittorio Girotto
4 shared publications- Monica Bucciarelli
University of Turin
3 shared publications
Latest Publications
- Open Access
- 10 August 2026
- Published by Center for Open Science in PsyArXiv
Reasoning Beyond Reality: A Mental Model Theory of Counterfactuals
The sentence: If he had been the thief then he would have fled, is interpretable as a counterfactual conditional: he is neither the thief nor did he flee. Theorists often presuppose that counterfactuals are subjunctive conditionals of this sort. We present a new theory of counterfactuals based on mental models. Four experiments corroborated it, and its computational implementation simulates its dual systems of intuition and deliberation. As it assumes, no linguistic cue in English signals that a sentence is counterfactual. Depending on its context, any assertion can describe a factual or a counterfactual domain, whether it is indicative or subjunctive (Experiment 1). Assertions can have mixed moods with one indicative clause and one subjunctive clause, e.g.: She paid her mortgage or she would have moved. People reason that this disjunction implies that she did pay her mortgage and did not move (Experiments 2 and 3). Any sort of compound assertion based, say, on and, if, and or, can have a counterfactual interpretation. It yields inferences from models in both factual and counterfactual domains, and from a model in one domain to its denial in another domain. Their difficulty depends on the number of intuitive models of their premise (Experiment 4). We discuss how counterfactuals can elicit certain correct inferences more readily than factual assertions do, how verifications of counterfactuals can depend on mental simulations using kinematic models, and whether alternative theories of our results are feasible. We conclude that factuals and counterfactuals can have the same language, models, and inferences.
- 10 August 2026
- Published by Center for Open Science in PsyArXiv
Open AccessReasoning Beyond Reality: A Mental Model Theory of Counterfactuals
The sentence: If he had been the thief then he would have fled, is interpretable as a counterfactual conditional: he is neither the thief nor did he flee. Theorists often presuppose that counterfactuals are subjunctive conditionals of this sort. We present a new theory of counterfactuals based on mental models. Four experiments corroborated it, and its computational implementation simulates its dual systems of intuition and deliberation. As it assumes, no linguistic cue in English signals that a sentence is counterfactual. Depending on its context, any assertion can describe a factual or a counterfactual domain, whether it is indicative or subjunctive (Experiment 1). Assertions can have mixed moods with one indicative clause and one subjunctive clause, e.g.: She paid her mortgage or she would have moved. People reason that this disjunction implies that she did pay her mortgage and did not move (Experiments 2 and 3). Any sort of compound assertion based, say, on and, if, and or, can have a counterfactual interpretation. It yields inferences from models in both factual and counterfactual domains, and from a model in one domain to its denial in another domain. Their difficulty depends on the number of intuitive models of their premise (Experiment 4). We discuss how counterfactuals can elicit certain correct inferences more readily than factual assertions do, how verifications of counterfactuals can depend on mental simulations using kinematic models, and whether alternative theories of our results are feasible. We conclude that factuals and counterfactuals can have the same language, models, and inferences.
- Open Access
- 4 October 2024
- Published by Society for Neuroscience in eneuro
- Vol. 11 (10)
Mistakes in Thinking about Cognitive Science and How to Reduce Them
This article allows readers to assess their ability to detect errors in thinking in seven case histories of psychologists’ thoughts about cognitive science. It explains the nature of the errors and shows that some of them reflect faulty reasoning. It presents a “model method” to improve reasoning. It is based on the theory of mental models, which gives a general account of how individuals think, both deductively and indicatively, and which postulates that individuals construct mental models of possibilities in the world. The model method enhances both the accuracy and speed of reasoning. The article concludes with some general reflections on the role of knowledge of meanings, the world, and context in thinking.
- 4 October 2024
- Published by Society for Neuroscience in eneuro
- Vol. 11 (10)
Open AccessMistakes in Thinking about Cognitive Science and How to Reduce Them
This article allows readers to assess their ability to detect errors in thinking in seven case histories of psychologists’ thoughts about cognitive science. It explains the nature of the errors and shows that some of them reflect faulty reasoning. It presents a “model method” to improve reasoning. It is based on the theory of mental models, which gives a general account of how individuals think, both deductively and indicatively, and which postulates that individuals construct mental models of possibilities in the world. The model method enhances both the accuracy and speed of reasoning. The article concludes with some general reflections on the role of knowledge of meanings, the world, and context in thinking.
- 4 June 2024
- Published by Springer Nature in Minds and Machines
- Vol. 34 (3) , 1-22
Models of Possibilities Instead of Logic as the Basis of Human Reasoning
- 4 June 2024
- Published by Springer Nature in Minds and Machines
- Vol. 34 (3) , 1-22
Models of Possibilities Instead of Logic as the Basis of Human Reasoning
- 14 February 2024
- Published by Taylor & Francis in Journal of Cognitive Psychology
- p. 1-22
Is creativity computable?
- 14 February 2024
- Published by Taylor & Francis in Journal of Cognitive Psychology
- p. 1-22
Is creativity computable?
- Open Access
- 25 September 2023
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 120 (40)
Human verifications: Computable with truth values outside logic
Cognitive scientists treat verification as a computation in which descriptions that match the relevant situation are true, but otherwise false. The claim is controversial: The logician Gödel and the physicist Penrose have argued that human verifications are not computable. In contrast, the theory of mental models treats verification as computable, but the two truth values of standard logics, true and false , as insufficient. Three online experiments (n = 208) examined participants’ verifications of disjunctive assertions about a location of an individual or a journey, such as: ‘You arrived at Exeter or Perth’. The results showed that their verifications depended on observation of a match with one of the locations but also on the status of other locations (Experiment 1). Likewise, when they reached one destination and the alternative one was impossible, their use of the truth value: could be true and could be false increased (Experiment 2). And, when they reached one destination and the only alternative one was possible, they used the truth value, true and it couldn’t have been false , and when the alternative one was impossible, they used the truth value: true but it could have been false (Experiment 3). These truth values and those for falsity embody counterfactuals. We implemented a computer program that constructs models of disjunctions, represents possible destinations, and verifies the disjunctions using the truth values in our experiments. Whether an awareness of a verification’s outcome is computable remains an open question.
- 25 September 2023
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 120 (40)
Open AccessHuman verifications: Computable with truth values outside logic
Cognitive scientists treat verification as a computation in which descriptions that match the relevant situation are true, but otherwise false. The claim is controversial: The logician Gödel and the physicist Penrose have argued that human verifications are not computable. In contrast, the theory of mental models treats verification as computable, but the two truth values of standard logics, true and false , as insufficient. Three online experiments (n = 208) examined participants’ verifications of disjunctive assertions about a location of an individual or a journey, such as: ‘You arrived at Exeter or Perth’. The results showed that their verifications depended on observation of a match with one of the locations but also on the status of other locations (Experiment 1). Likewise, when they reached one destination and the alternative one was impossible, their use of the truth value: could be true and could be false increased (Experiment 2). And, when they reached one destination and the only alternative one was possible, they used the truth value, true and it couldn’t have been false , and when the alternative one was impossible, they used the truth value: true but it could have been false (Experiment 3). These truth values and those for falsity embody counterfactuals. We implemented a computer program that constructs models of disjunctions, represents possible destinations, and verifies the disjunctions using the truth values in our experiments. Whether an awareness of a verification’s outcome is computable remains an open question.
- Open Access
- 25 January 2023
- Published by Center for Open Science in PsyArXiv
Disjunctions, Metadisjunctions, and Mental Models
A disjunction of two assertions, Either p or else q, but not both, where p and q are everyday assertions, is equivalent to a metadisjunction, such as:Witness 1 asserts p.Witness 2 asserts q.One witness is telling the truth, and one witness is lying.In normal logics, the disjunction and metadisjunction have the same ‘truth-functional’ meanings: they are true provided that one of their two clauses, p and q, is true and one of them is false. The theory of mental models differs: the disjunction refers to a conjunction of two possibilities, p and q, that hold in default of knowledge to the contrary. Intuitive models represent the two possibilities as p and as q; deliberative models represent them as p and not-q and as not-p and q. Four experiments showed that people tend to rely on intuitive models, which for certain predictable inferences lead them into systematic errors. They also showed that participants infer that a metadisjunction of assertions of explicit possibilities: possibly p or possibly q implies possibly p. This inference violates all normal logics, which rely on the truth-functional meaning of disjunctions. A similar well-known ‘paradox’ concerns permission, e.g.: You can have a court-appointed lawyer or a private lawyer; therefore, you can have a private lawyer. The multitude of theories of these paradoxes either maintain the truth-functions semantics and posit that disjunctions yield pragmatic inferences, or else, like the model theory, posit an alternative semantics. Our results are contrary to the pragmatic accounts, because a metadisjunction depends, not on a single disjunction, but on several assertions.
- 25 January 2023
- Published by Center for Open Science in PsyArXiv
Open AccessDisjunctions, Metadisjunctions, and Mental Models
A disjunction of two assertions, Either p or else q, but not both, where p and q are everyday assertions, is equivalent to a metadisjunction, such as:Witness 1 asserts p.Witness 2 asserts q.One witness is telling the truth, and one witness is lying.In normal logics, the disjunction and metadisjunction have the same ‘truth-functional’ meanings: they are true provided that one of their two clauses, p and q, is true and one of them is false. The theory of mental models differs: the disjunction refers to a conjunction of two possibilities, p and q, that hold in default of knowledge to the contrary. Intuitive models represent the two possibilities as p and as q; deliberative models represent them as p and not-q and as not-p and q. Four experiments showed that people tend to rely on intuitive models, which for certain predictable inferences lead them into systematic errors. They also showed that participants infer that a metadisjunction of assertions of explicit possibilities: possibly p or possibly q implies possibly p. This inference violates all normal logics, which rely on the truth-functional meaning of disjunctions. A similar well-known ‘paradox’ concerns permission, e.g.: You can have a court-appointed lawyer or a private lawyer; therefore, you can have a private lawyer. The multitude of theories of these paradoxes either maintain the truth-functions semantics and posit that disjunctions yield pragmatic inferences, or else, like the model theory, posit an alternative semantics. Our results are contrary to the pragmatic accounts, because a metadisjunction depends, not on a single disjunction, but on several assertions.
- 1 January 2023
- Published by American Association for the Advancement of Science (AAAS) in Intelligent Computing
- Vol. 2
What Should Replace the Turing Test?
Today, chatbots and other artificial intelligence tools pass the Turing test, which was Turing’s alternative to trying to answer the question: can a machine think? Despite their success in passing the Turing test, these machines do not think. We therefore propose a test of a more focused question: does a program reason in the way that humans reason? This test treats an “intelligent” program as though it were a participant in a psychological study and has 3 steps: (a) test the program in a set of experiments examining its inferences, (b) test its understanding of its own way of reasoning, and (c) examine, if possible, the cognitive adequacy of the source code for the program.
- 1 January 2023
- Published by American Association for the Advancement of Science (AAAS) in Intelligent Computing
- Vol. 2
What Should Replace the Turing Test?
Today, chatbots and other artificial intelligence tools pass the Turing test, which was Turing’s alternative to trying to answer the question: can a machine think? Despite their success in passing the Turing test, these machines do not think. We therefore propose a test of a more focused question: does a program reason in the way that humans reason? This test treats an “intelligent” program as though it were a participant in a psychological study and has 3 steps: (a) test the program in a set of experiments examining its inferences, (b) test its understanding of its own way of reasoning, and (c) examine, if possible, the cognitive adequacy of the source code for the program.
- Open Access
- 22 November 2022
- Published by Center for Open Science in PsyArXiv
Reasoning with Quantified Modal Premises: Some Everyday Inferences May Be Logically Invalid
We present a theory of quantified modal reasoning, and four studies of individuals’ spontaneous conclusions that test its predictions. From premises, such as: All of those artists are businessmen, Paulo is possibly one of those artists, participants tended to conclude: Paulo is possibly a businessman (Experiment 1). This inference follows as necessary from an intuitive mental model that represents, say, three artists who are each businessmen, and Paulo as one of the artists. Deliberation can lead to a model in which Paulo is not an artist, which is consistent with the presupposition of the premise about him, but a possibility follows provided it holds in at least one correct model. Individuals tend to infer the model theory’s predicted conclusions about possibilities even when standard modal logics evaluate them as invalid (Experiment 2). The theory also predicts common errors, such as inferences of categorical conclusions from premises about possibilities (Experiment 3) and even from premises for modal syllogisms (Experiment 4). The paper concludes with a general account of quantified modal reasoning.
- 22 November 2022
- Published by Center for Open Science in PsyArXiv
Open AccessReasoning with Quantified Modal Premises: Some Everyday Inferences May Be Logically Invalid
We present a theory of quantified modal reasoning, and four studies of individuals’ spontaneous conclusions that test its predictions. From premises, such as: All of those artists are businessmen, Paulo is possibly one of those artists, participants tended to conclude: Paulo is possibly a businessman (Experiment 1). This inference follows as necessary from an intuitive mental model that represents, say, three artists who are each businessmen, and Paulo as one of the artists. Deliberation can lead to a model in which Paulo is not an artist, which is consistent with the presupposition of the premise about him, but a possibility follows provided it holds in at least one correct model. Individuals tend to infer the model theory’s predicted conclusions about possibilities even when standard modal logics evaluate them as invalid (Experiment 2). The theory also predicts common errors, such as inferences of categorical conclusions from premises about possibilities (Experiment 3) and even from premises for modal syllogisms (Experiment 4). The paper concludes with a general account of quantified modal reasoning.
- Open Access
- 21 November 2022
- Published by Center for Open Science in PsyArXiv
Mental models and the algorithms of deduction
Pose the following problem to a smart eight-year-old: "All machines can break down. Alexa is a machine. What follows?" and the child is likely to reply: "Alexa can break down." So, as experiments confirm, human beings unschooled in logic are able to make deductions. Yet, this easy deduction defeats Alexa, Siri, and other virtual assistants. To build machines that reason, students of reasoning need to know the answers to three questions: (1) Which deductions do human reasoners make? (2) How do they make them? And (3) How can computers simulate them? The goal of this chapter is to describe the main efforts to simulate human deduction. It aims to provide its own intellectual life-support system so readers can understand it without having to consult anything else.
- 21 November 2022
- Published by Center for Open Science in PsyArXiv
Open AccessMental models and the algorithms of deduction
Pose the following problem to a smart eight-year-old: "All machines can break down. Alexa is a machine. What follows?" and the child is likely to reply: "Alexa can break down." So, as experiments confirm, human beings unschooled in logic are able to make deductions. Yet, this easy deduction defeats Alexa, Siri, and other virtual assistants. To build machines that reason, students of reasoning need to know the answers to three questions: (1) Which deductions do human reasoners make? (2) How do they make them? And (3) How can computers simulate them? The goal of this chapter is to describe the main efforts to simulate human deduction. It aims to provide its own intellectual life-support system so readers can understand it without having to consult anything else.
- Open Access
- 26 April 2022
- Published by Center for Open Science in PsyArXiv
Human verifications: Computable with truth values outside logic
The theory of mental models allows modal truth values (‘possibly true and possibly false’) and counterfactual truth values (‘true but it could have been false’), which are outside standard logics. Naive individuals have the intuition that the disjunction, ‘You arrived at Preston or at Crewe,’ is true if you arrived at Preston, but their deliberations consider the status of the alternative destination. If this counterfactual possibility is false, e.g., the road to Crewe is blocked, they are likely to select a modal truth value (in Experiment 1), or a counterfactual truth value (in Experiment 2). The results are pertinent to claims that cognition cannot be algorithmic (as Gödel argued on the grounds that consistent formal systems cannot prove truths that mathematicians grasp about the arithmetic of natural numbers). Human thinking, however, differs from formal logics: it relies on semantic algorithms that can evaluate truth and falsity.
- 26 April 2022
- Published by Center for Open Science in PsyArXiv
Open AccessHuman verifications: Computable with truth values outside logic
The theory of mental models allows modal truth values (‘possibly true and possibly false’) and counterfactual truth values (‘true but it could have been false’), which are outside standard logics. Naive individuals have the intuition that the disjunction, ‘You arrived at Preston or at Crewe,’ is true if you arrived at Preston, but their deliberations consider the status of the alternative destination. If this counterfactual possibility is false, e.g., the road to Crewe is blocked, they are likely to select a modal truth value (in Experiment 1), or a counterfactual truth value (in Experiment 2). The results are pertinent to claims that cognition cannot be algorithmic (as Gödel argued on the grounds that consistent formal systems cannot prove truths that mathematicians grasp about the arithmetic of natural numbers). Human thinking, however, differs from formal logics: it relies on semantic algorithms that can evaluate truth and falsity.