Skip to content

Visual neuroscience and perception

Research by this group centres on the study of the fundamental mechanisms of vision and perception. In pursuit of this basic scientific knowledge the group employs a multi-methodological, cross-disciplinary approach using behavioural, neuro-imaging, neuro-stimulation and mathematical modelling techniques to study a variety of different aspects of visual perception and cognition including: colour, motion, as well as how the brain integrates different types of sensory information.

Multisensory perception

A sound can influence when (or at what location) an observer sees a moving dot reverse its direction. Real-world events involve both sound and vision, but how do we integrate the information from our eyes and ears into a single unified percept? We are interested in the effects of sound on visual perception, and vice-versa.

We also study somatosensory (tactile) stimuli as a third modality, with particular interest in the relative perceived timing of events across the modalities. It emerges that perceived time is far more flexible than previously thought, being particularly susceptible to recent experience.

An additional area of interest is the way in which our motor actions relate to our sensory perception. In a particularly fascinating illusion, we are able to trick the brain into perceiving a sensory event prior to the motor action producing the event!

Time perception

How does the brain tell time?

  • Why is it that ‘time flies when you are having fun’ and ‘a watched pot never boils’?
  • Are their specialist areas of the brain that handle time perception?
  • If so, why do they get it so wrong so often?

Perceived time is often unreliable and inaccurate and the aim of this research is to use psychophysical techniques to investigate the mechanisms by which our brains signal the passage of time.

Colour vision and graphics

Colour is a fundamental aspect of our visual experience that helps us identify objects and interact with the world. The colour lab undertakes research in colour constancy and colour appearance. We are also interested in various aspects of normal and abnormal colour perception, in how and where colour is analysed by the visual system and in optimising colour display and conversion systems.

Colour memory and colour constancy

Thanks to colour constancy your favourite t-shirt appears to be the same colour inside, under artificial light, as it would outside, under natural light even though the spectral distribution of the light that reaches your eyes has changed.

To achieve good colour constancy it is fundamental to have good colour memory. In our lab we use real object and lights as well as computer displays to try to understand how colour constancy happens and how it is related to other phenomena such as colour memory, adaptation and contrast.

Colour in computer graphics

High Dynamic Range Imaging & Compression

High Dynamic Range (HDR) Imaging allows the capture of the entire luminance range of a natural scene. HDR displays can have a luminance output of up to 3000 cd/m2 and thus a contrast ratio of 1:30,000. They create a very vivid viewing experience for the observer.

Colour to greyscale conversion

Tone-mapping operators (TMO) compress the high dynamic range to fit the range of a standard monitor. Our aim is to identify the best possible compression.

Motion tracking

How good are we at attending to several items at the same time?

In this research project, our 'items' are dots moving in straight lines through a gap. One of the dots changes direction. How easy is it to detect this, and does the task get harder with more dots? These experiments help us understand how attentional and memory resources are distributed when monitoring several items.

Computational vision

Research on computational vision focusses on the intersection between machine and human perception. Research concentrates on edge detection in human and machine, leading to a new theoretical account of optimal edge detection, which explains human blur detection better than all other competing models. Members of the group have also developed models of visual search which explain the variation in search speed, and the correlation between reaction times in search experiments.

ERG electrophysiology

This research project involve the study of the human Electroretinogram (ERG) which is an electrical signal that originates in the retina and can be measured non-invasively using electrodes placed on or close to the cornea.  The ERG signal recorded at the cornea is initiated by light absorption which leads to electrical activity in the photoreceptors (of which there are four main types L-, M- and S-cones and rods) and in their post-receptoral pathways. Light absorption in the distinct photoreceptor types may lead to different ERG responses caused either by differences between the photoreceptors themselves or between their post-receptoral pathways. A complete characterisation of contributions of the different photoreceptor types to the ERG may therefore give more detailed insight into the functional integrity of the human retina. Such a description can be obtained by isolating the responses of a single photoreceptor type.

Current state-of-the-art LED stimulators now allow precise control of differently coloured light stimuli and have an advantage over more traditional CRT based stimulation systems in that they can better isolate the activity of rod and L-, M- and S-cone receptor populations. This precision, coupled with our knowledge of cone and rod spectral characteristics, enables a precise description and control of photoreceptor excitation. Theoretically, any desired combination of photoreceptor excitation modulation can be achieved, including conditions in which the activity in only one photoreceptor type is modulated (i.e. silent substitution). In this manner the response of one photoreceptor type can be isolated without changing the state of adaptation. This technique can and has been used to study the contribution of signals originating in the different photoreceptor types to the human ERG.