Attention and camouflage in three dimensions

Supervisors: Julie M. Harris, Søren K. Andersen

Project Description:
Binocular vision is thought to have evolved primarily for the perception of depth and shape. Yet our ability to use two eyes to see depth has also been suggested as a way for predators to ‘break camouflage’, and beat the elegant camouflage that many prey animals display in terms of their patterning and shading. Surprisingly, the extent to which binocular vision does allow camouflage to be broken, and the brain mechanisms used to do so, have been little studied.  There are at least 3 stages involved in breaking camouflage: attend to a location, detect an object is present, use shape and depth to identify the object.  In this project we will explore the first two of these using behavioural and human neuroscience techniques to, for the first time, characterise the importance of depth for the selective attention process.

The student will work with two research groups: Harris (St. Andrews: expert in binocular vision and animal patterning) and Andersen (Aberdeen: expert in selective attention) to study fundamental questions around how binocular vision allows us to attend to objects defined by binocular depth.  Harris is currently working on how animal patterning hides (camouflage) and/or draws attention (warning signals) to the animal (e.g Penacchio et al, 2015). Andersen is currently working on a quantitative model of attention, understanding how attention from different visual features affects stimulus processing (Andersen et al., 2015).  We will combine these interests to study how depth from binocular vision fits into the attention processing hierarchy.

The key research question is whether depth can be integrated into attentional frameworks (e.g. Reynolds & Heeger, 2009) in an analogous way to other visual features. Depth might be special because it is also a spatial dimension and because depth perception in itself may be more complex than simpler features, such as colour or orientation. This project will address this question by applying carefully quantified binocular depth stimuli to classic behavioural attention paradigms and human EEG (in particular steady-state visual evoked potentials).

The project is interdisciplinary, bringing together evolutionary biology, experimental psychology and human neuroscience. The student will be trained in a broad range of techniques including Matlab programming, behavioural experiment design, binocular vision hardware, eye tracking, and steady-state visual evoked potentials (EEG). Many of these techniques require some mathematical knowledge, thus the project would suit a graduate in neuroscience, psychology or physical science/engineering, who would relish the opportunity to learn and apply quantitative techniques to biological problems.


Penacchio, O., Cuthill, I. C., Lovell, P. G., Ruxton, G. D. and Harris, J. M. (2015), Orientation to the sun by animals and its interaction with crypsis. Funct Ecol, 29: 1165–1177. doi:10.1111/1365-2435.12481

Andersen, S.K., Müller, M.M. and Hillyard, S.A. (2015). Attentional selection of feature conjunctions is accomplished by parallel and independent selection of single features. Journal of Neuroscience, 35(27): 9912-9919.

Reynolds, J.H., Heeger, D.J. (2009). The normalization model of attention. Neuron 61:168–185.