The neural basis of response time variability

Response times are highly variable. For example, when driving the onset of a red traffic light can lead to large variability in the time to press the brake even when all other factors such as alertness, vigilance, visibility etc. are constant. Such response variability is still present in highly controlled laboratory conditions. The origins of this variability remains a mystery, but understanding it is a central question for scientists with an interest in understanding the relation between brain and behaviour, and - as the example above illustrates - is relevant in a number of applied contexts. A comprehensive account of response time variability should explain at a functional level where the variability originates and determine the neural processes and anatomical structures involved. In my research group we focus on the study of the eye movement response. Eye movements are interesting for a number of reasons. First, they are important for visual perception: we only see fine detailed information when the eyes point directly at a location. Second, eye movements are ubiquitous: we make more eye movements in a day than heart beats. Third, we have a detailed knowledge of the neurophysiology of eye movements control in non-human primates which allows us to link functional and neural explanations. For the last ten years the major focus of my research has been to identify functional explanations for eye movement responses using mathematical models and behavioural experiments. The next major step for this research is to investigate the brain processes that account for this variability. Functional Magnetic Radiation Imaging (fMRI) and Magnetoencephalography (MEG) are both methods for imaging the human brain while participants carry out a task. This allows the brain areas involved in the task to be identified. This Fellowship will allow me to introduce these methods to my research to study the neural basis of response time variability.

The project will use the human saccadic eye movement response to investigate the more general problem of response time variability. Our previous work has established the importance of the visual stimulus and the nature of the response required in determining response time. Alongside this we have used mathematical models to explain changes in response time in terms of changes in threshold and accumulation rate of the decision variable to respond (c.f. Carpenter & Williams, 1995). The next step in this research is to use brain imaging (MEG and fMRI in this case) to investigate in more depth the neural basis of response time variability. A central first question for the development of models of choice and reaction time variability is understanding the origins of changes in response time when the stimulus and the response are held constant. The variability, which must be internally generated, cannot be investigated used standard behavioural methods which infer properties of the systems by changing either the stimulus or response. Equally the computational models in this area do not specify directly where this variability comes from and understanding the neural substrate of this variability will provide a significant constraint on these models. Pilot work, using MEG, collected in preparation for this application showed that the state of visual cortex in the pre-target period plays a major role in determining response latency. This leads to a set of questions that will be addressed in this application using MEG and fMRI: Why is there decreased Gamma activity for fast saccades in the pre target interval? Do the small changes in the visual response to target really reflect the differences between fast and slow saccades? Do the differences in visual response for fast saccades match responses to more visually salient targets?

The work outlined in this application addresses a fundamental question: what is the basis of response time variability. In humans in particular, answering this question has extremely wide ranging implications within and outside the academic community. My research group has always engaged a range of non-academic stake holders. Three currently active research projects that apply the research developed in this fellowship illustrate this. Our work recording and modelling eye movement behaviour of CCTV operators has Manchester City Council as research partner. Our project on natural dynamic scenes and human vision is jointly funded by EPSRC and The Defence Science and Technology Laboratory (Dstl). The relationship with Dstl and the Vision Group at the University of Bristol is a long established one and provides a mechanism by which our research has a direct effect on government. We are developing a working model of human vision which includes movements of the eyes. The project is funded by QinetiQ Group PLC which ensures that our work will have an impact within the commercial sector. My group will continue to work in this manner with local government, central government and the commercial sector to ensure the impact of this new research. The new Clinical Research and Imaging Centre (CRIC) project in Bristol will bring a wide range of clinical staff from the NHS into contact with basic researchers to develop new solutions for the care of patients. My ongoing involvement in CRIC will provide a structured opportunity for this work to have a potential impact on patient care within the NHS. I am a member of the Applied Vision Association (AVA) and Bristol Vision Institute (BVI) which bring together researchers from Universities, Industry and Government for regular conferences to exchange ideas and new research. Alongside a wide range of public engagement work on a more ad-hoc basis, for a number of years I have been working with an artist and staging art exhibitions in which the art is inspired and often produced by eye movement data. These exhibitions are always accompanied by a number of joint talks in which I discuss the science in detail to members of the public. I have a strong track record and continuing commitment to ensuring that my work has an impact outside the academic community. The structures and relationships for this to continue are in place and I will, of course, add to this portfolio of contacts as this project progresses.