The Placement Project
Goal-directed behaviours are often performed as part of an on-going chain of visual processes that are fluently converted into action1. The visual information we gather is determined by the position of our eyes and this in turn determines the next eye position. A great deal of work examining the processing of visual information underlying the control of saccades has been restricted to the events which take place prior to and during a single movement being executed2. In contrast to this, saccades do not usually occur in isolation but as part of a more complex set of eye movements. There is a growing body of evidence showing that when considered in this way, visual information used for action can be encoded continuously, suggesting that our vision and saccade processing may be more labile and flexible than originally conceived3,4,5,6,7. One of the major drawbacks with existing eye movement paradigms examining sequencing of behaviour is that they often produce a small number of trials on which sequences are generated and only consider a limited sequence length, i.e., two or three consecutive saccades.
In this project we developed and examined a novel paradigm in which these problems are addressed. In order to examine the limits of visual encoding of information for action we manipulated the amount of prior information given about the saccade sequence, e.g., in a sequence of 7 saccades only the location of the next 3 locations was indicated and this was updated as saccade responses were made, thereby maintaining prior information about the next 3 locations throughout the sequence. This allows saccades to be partially programmed prior to their execution. We predicted that increasing the amount of prior information would confer latency and accuracy benefits for subsequent saccades. The point at which further prior information confers no further benefit in the latency and accuracy of saccade targeting indicates the limits of the on-going parallel encoding of visual information.
In a second set of experiments, we investigated the spatial aspects of target visual information that are encoded prior to its being fixated. This is important in consideration of how the spatial layout of a visual environment will impact an individuals’ ability to encode it and act effectively within it. Here the spatial properties of visual information were manipulated in order to explore the impact on its encoding. The positions of subsequent targets at different points further along the sequence was shifted by a small amount (in distance, direction or both) causing a mismatch between their previously encoded and their new physical location. The extent to which targets can be shifted while still inducing a benefit in saccade latency and accuracy is indicative of how spatially specific the encoding is and accuracy errors show the extent to which distance and direction are encoded.
Skills developed during the placement
This placement helped me develop essential research skills by allowing me to work in a hands-on, research-focused environment. Over the course of the placement I was charged with a variety of tasks and given intellectual independence while still being offered the support necessary to fully benefit from the learning opportunities that those tasks provided.
As part of the placement I had the opportunity to experience nearly all aspects of academic research. I took part in designing the study, which helped me understand the perspective required to integrate all the different components of a study in a coherent way in order to productively address the research question in hand. I was trained on programming eye-tracking experiments that will carry out the experimental design, create and display the visual stimuli, and record peoples eye movements (using SR Research Experiment Builder). I was also trained in the use of the eye tracker, e.g., how to calibrate and properly use the eyetracker (Eyelink II) and analyzing the datafrom raw eye movement output to statistical analysis. Furthermore, I was responsible for most of the administrative aspects of the project, e.g., handling data for subsequent analysis, recruiting participants, scheduling them for testing, and making sure they receive remuneration for their participation.
The project was extended to explore the limits of parallel encoding of visual information in the elderly population and is being finalized with a view to publication.
(1)Land,M.F.,Tatler,B.W.(2009).Looking and Acting.OUP
(2)Findlay,J.M.,Gilchrist,I.D.(2003).Active Vision. OUP
(7)Godijn,R.Theeuwes,J.(2002).J Exp Psychol,28,1039-1054