Inventiveness and the progress of product innovation. Quantitative models of inventiveness will be used to forecast the potential rate of improvement of a technology and to re-design products to improve more rapidly and steadily. By focusing on innovation in products and technologies in energy conversion, this research can guide development funding for low-carbon energy generation.
How is information organised in the mind? Learning structured mental representations from data. One of the biggest questions in psychology is to understand the principles that the mind uses to organise information. This project is both a search for these underlying psychological laws, and an attempt to develop new statistical technologies and mathematical tools that can be used to organise information in applied settings.
Computational models of synergies in human language acquisition. How do children learn language? Do they first learn to recognise words and then associate words with meanings, or do they use the meanings to figure out what the words are, or do they do both at the same time, and if so, how? This project will investigate questions like these using advanced computational models of the way children learn from their environment.
Incremental syntactic parsing and coreference resolution. As computers become smaller, keyboards and screens become increasingly impractical. We'd like to be able to talk to our computers, but they'd have to understand what we say. This project will develop a computational model that tracks which things are talked about and identifies 'who did what to whom' in text or speech.
Learning and planning with qualitative models. This project will give a robot the ability to learn how to interact with its environment, using common sense reasoning to guide trial-and-error learning. The outcome will be a robot that is able to quickly adapt to new and changing environments, such as those which might be encountered in applications like robots for urban search and rescue.
Tracking the Flow of Perceptual Information Through Decision Networks. The choices we make define our lives. Despite exciting progress in neuroscience, we still don’t know how the inner workings of the brain give rise to simple decisions. This project brings together experts from diverse domains of computational neuroscience to investigate how our brains turn perceptual information into action. Together, we will develop new methods to track information flow through the brain during the decision ....Tracking the Flow of Perceptual Information Through Decision Networks. The choices we make define our lives. Despite exciting progress in neuroscience, we still don’t know how the inner workings of the brain give rise to simple decisions. This project brings together experts from diverse domains of computational neuroscience to investigate how our brains turn perceptual information into action. Together, we will develop new methods to track information flow through the brain during the decision making process. By doing so, we will develop a world-leading model of how the brain makes decisions, and also provide the broader scientific community with a set of exciting new tools for studying information processing in the brain.Read moreRead less
The dog that didn't bark: a Bayesian account of reasoning from censored data. This project aims to develop and test a new computational theory of inductive reasoning. Inductive reasoning involves extending knowledge from known to novel instances, and is a central component of intelligent behaviour. This project will address the cognitive mechanisms that allow people to draw inferences based on both observed and censored evidence. The project intends to test the model through an extensive program ....The dog that didn't bark: a Bayesian account of reasoning from censored data. This project aims to develop and test a new computational theory of inductive reasoning. Inductive reasoning involves extending knowledge from known to novel instances, and is a central component of intelligent behaviour. This project will address the cognitive mechanisms that allow people to draw inferences based on both observed and censored evidence. The project intends to test the model through an extensive program of experimental investigation and computational modelling. The anticipated benefits include an enhanced understanding of human inference, especially in domains such as the evaluation of forensic or financial evidence, where data censoring is common.Read moreRead less
Uncovering the processes underlying human reasoning: A state-trace approach. This project aims to answer the most important unresolved question in the psychology of reasoning; how many distinct cognitive processes underlie human reasoning? To answer this question, this project aims to conduct an extensive experimental investigation of the factors that selectively impact inductive and deductive inferences and the application of high-dimensional state-trace analysis; a powerful new method for diag ....Uncovering the processes underlying human reasoning: A state-trace approach. This project aims to answer the most important unresolved question in the psychology of reasoning; how many distinct cognitive processes underlie human reasoning? To answer this question, this project aims to conduct an extensive experimental investigation of the factors that selectively impact inductive and deductive inferences and the application of high-dimensional state-trace analysis; a powerful new method for diagnosing underlying processes from behavioural data. The project is expected also to develop a new computational model that accounts for both inductive and deductive forms of reasoning.Read moreRead less
The role of inductive reasoning in generalization of associative learning. This project seeks to develop a better understanding of how learning is generalised to novel stimuli. Learning about associations around us helps us to obtain resources and minimise threat. A critical task for the learner is how far to extrapolate this knowledge: too little generalisation reduces the benefits of learning and too much risks distraction and maladaptive responding. Recent evidence has shown an important role ....The role of inductive reasoning in generalization of associative learning. This project seeks to develop a better understanding of how learning is generalised to novel stimuli. Learning about associations around us helps us to obtain resources and minimise threat. A critical task for the learner is how far to extrapolate this knowledge: too little generalisation reduces the benefits of learning and too much risks distraction and maladaptive responding. Recent evidence has shown an important role for reasoning processes in human associative learning. This project aims to apply insights from the inductive reasoning literature to study the role of hypothesis and category induction in generalisation of associative learning. The results are expected to have important implications for our understanding of associative learning and generalisation which may inform techniques to promote adaptive generalisation in fields such as education, training and clinical practice.Read moreRead less
Improving the diagnosticity of eyewitness memory choices. Eyewitness identification error is common and costly. This project aims to improve the quality of information provided by eyewitnesses, and the ability of police officers and triers of fact (e.g., juries, judges) to evaluate this information. Laboratory investigations will determine how best to test memory and confidence to achieve this aim. A new class of cognitive models will provide a unified account of response accuracy, response time ....Improving the diagnosticity of eyewitness memory choices. Eyewitness identification error is common and costly. This project aims to improve the quality of information provided by eyewitnesses, and the ability of police officers and triers of fact (e.g., juries, judges) to evaluate this information. Laboratory investigations will determine how best to test memory and confidence to achieve this aim. A new class of cognitive models will provide a unified account of response accuracy, response time, and confidence, suitable for application to computerized testing scenarios. The models and testing methods validated in the laboratory will be refined for application in eyewitness memory settings, facilitating better evaluation of identification evidence, and potentially reducing wrongful convictions.Read moreRead less