Visuelle Wahrnehmung  Leiter:  Prof. Michael Morgan

(Visual Perception)

The visual system receives its input in the form of an optical image. The interpretation of this image allows us to perceive our environment and carry out visually guided actions. It is believed that these operations involve both serial processing in steps of varying complexity and parallel processing of different information, such as color and movement in different brain areas. An active process referred to as Selective Attention determines the allocation of resources to different tasks.

Our goal is to understand the processing strategies of the different brain areas and the way in which resources are allocated.

Measurement Methods

a) Psychophysics

High resolution visual displays and signal detection methods allow us to measure human observers’ ability to distinguish differences between patterns, such as different textures.

b) Functional Magnetic Resonance Imaging (fMRI)

We are studying how and why attention affects the initial magnitude and adaptation of the BOLD response to visual stimuli. In future work we would like to measure the neural correlates of the lapses in consciousness (‘Mackworth’s Blocks’) that are known to occur during tasks requiring prolonged  vigilance.

c) Eye movement and pupillary responses

A sampling rate of 2000 Hz allows us to measure eye movements with high precision, and to present saccade-triggered images in real time.

d) Comparative measurements

We measure the changes in visual perception in people with dyslexia or synaesthesia or following brain lesions.

Eyetracker Setup

Examples of ongoing projects

Basic Visual Geometry and the ‘Primal Sketch’

Poggendorf Illusion: the lower of the two lines on the left is actually collinear with the line on the right, but appears displaced downwards.

The visual brain is normally highly accurate at basic geometrical operations, such as deciding whether two lines are collinear. We have been carrying out functional imaging experiments to determine which brain areas are involved in these operations (Ref. 12) and psychophysical experiments to determine the computational rules used by the visual system in basic geometry (Ref. 6). An important clue comes from cases where the observer makes errors, such as the classic Poggendorff illusion of alignment  (see right illustration).

Nach oben

Vision, action and eye movements

We have also used the Poggendorff illusion for studying the relation between visual perception and action. We find that when subjects make hand-pointing or eye movements to imaginary points of intersection on oblique lines, they make the same errors as they do when making a perceptual alignment. Another illusion, in which moving objects are located ahead of their instantaneous position, is also shared by the rapid eye movement system.

Nach oben

Global Saccadic Adaptation

If the target for a rapid eye movement (saccade) shifts its position during the eye movement, the shift is usually invisible. If the shift is consistent,rapid adaptation occurs so that the saccade end point is shifted in the direction of the new position. This is true even if the target disappears during the saccade. We have developed a novel global version of this adaptation in which the target moves 5 deg clockwise no matter where it appears around a circle centered on the starting eye position. After adaptation, the saccade end point is shifted clockwise, even if the initial target position is in a direction not previously tested. We find no evidence that the perceptual position of the target was shifted.

Schema for training global saccadic adaptation. During training, the subject fixates the center position and moves the eyes as rapidly as possible when a target (red crosses) is presented on the outer circle. During the eye movement, the target moves 5 deg clockwise. After training, the target appears in untrained test positions.

Nach oben

Perception of Regularity in Textures

It is a mystery why the world looks as regular and stable as it does, given the constantly changing nature of the retinal image as we move our eyes around the world, and the prevalence of ‘noise’ in the nervous system, which is revealed by sensory detection experiments. We have used experiments with not-quite-regular textures to test the hypothesis that the visual system has an underlying template for regularity, which is tested against incoming sensory evidence (Refs. 4,11).

The dots on the left are regularly spaced. In the middle the position of each dot is randomly perturbed by a probability function that is below the sensory threshold. On the right, the shifts are large enough to be noticeable. If you stare for some time at the center dot in the right-hand figure, the texture will begin to look more and more regular.

Nach oben

Attention and visual sensitivity

Adaptation refers to the loss of sensitivity following prolonged exposure to a sensory stimulus. An example from everyday life is the underestimation of our car’s speed when we leave an Autobahn. Experiments to measure the changes in perception following adaptation often confound real changes in sensitivity with changes in the observer’s guessing strategy (Ref. 2). We have developed novel psychophysical methods for making unbiased measurements of adaptation, and have used these to show that the amount of adaptation is not affected by attending to the stimulus (Refs. 1,3,7).

Nach oben

How can we count large numbers?

There would be obvious advantages to a foraging animal in perceiving at a glance which tree had the most fruits. Not surprisingly, then, there are many demonstrations of relative numerosity discrimination in animals and humans. It has been suggested that numerosity is an elementary quality of perception, similar to colour, but if so, the mechanism is unknown at present. Our experiments have supported the idea that relative numerosity is a type of texture discrimination, and that a simple model computing the energy in just two spatial frequency-tuned pass bands can do better than human observers (Ref. 5).

Nach oben

Mechanisms of ‘crowding’ in visual stimuli

We have shown in psychophysical investigations that observers can report the average (mean) tilt of the patches, and that the centre patch contributes to this perceived average even though its individual orientation is perceived at chance level. More generally, we can extract summary statistics from images without being able to report the properties of individual elements. We have been measuring the sampling efficiency of observers as they extract these statistics (Refs. 9,10,11,15).

Nach oben

Signal Detection Theory and Psychophysics

How do you compare apples and oranges?  This is the task faced by a human observer looking at two stimuli and trying to decide whether they are different in, for example, blur or in contrast. Classical Signal Detection Theory (SDT) assumes that both blur and contrast are noisy signals and that the observer has to ‘bet’ which of them is more different. But different in what units? We have found that the metric used by observers to make their choice is the dimensionless z-score of the two signals, which implies that they have some knowledge of their own, internal noise (Ref. 14).

Nach oben

Job offer

Interested in joining this research group?

We are currently offering 1 postoctoral position.

Selected Publications

list as PDF

1    Morgan, M. J. Sustained attention is not necessary for velocity adaptation. J Vision 13(8):26, 1–11 (2013).

2    Morgan, M., Dillenburger, B., Raphael, S. & Solomon, J. A. Observers can voluntarily shift their psychometric functions without losing sensitivity. Atten Percept Psychophys 74, 185-193, doi:10.3758/s13414-011-0222-7 (2012).

3    Morgan, M. J. Motion adaptation does not depend on attention to the adaptor. Vision Res 55, 47-51, doi:10.1016/j.visres.2011.12.009 (2012).

4    Morgan, M. J., Mareschal, I., Chubb, C. & Solomon, J. A. Perceived pattern regularity computed as a summary statistic: implications for camouflage. Proc Biol Sci 279, 2754-2760, doi:10.1098/rspb.2011.2645 (2012).

5    Dakin, S. C., Tibber, M. S., Greenwood, J. A., Kingdom, F. A. & Morgan, M. J. A common visual metric for approximate number and density. Proc Natl Acad Sci U S A 108, 19552-19557, doi:10.1073/pnas.1113195108 (2011).

6    Morgan, M. J. Features and the 'primal sketch'. Vision Res 51, 738-753, doi:10.1016/j.visres.2010.08.002 (2011a).

7    Morgan, M. J. Wohlgemuth was right: distracting attention from the adapting stimulus does not decrease the motion after-effect. Vision Res 51, 2169-2175, doi:10.1016/j.visres.2011.07.018 (2011b).

8    Morgan, M. J., Chubb, C. & Solomon, J. A. Evidence for a subtractive component in motion adaptation. Vision Res 51, 2312-2316, doi:10.1016/j.visres.2011.09.002 (2011c).

9    Mareschal, I., Morgan, M. J. & Solomon, J. A. Cortical distance determines whether flankers cause crowding or the tilt illusion. J Vis 10, 13, doi:10.8.13 [pii]10.1167/10.8.13 (2010).

10    Mareschal, I., Morgan, M. J. & Solomon, J. A. Attentional modulation of crowding. Vision Res 50, 805-809, doi:10.1016/j.visres.2010.01.022 (2010a).

11    Morgan, M., Mareschal, I. & Solomon, J. Sampling Efficiencies for Spatial Regularity. J Vis 10, 1362, doi:10.1167/10.7.1362 (2010b).

12    Tibber, M. et al. The neural correlates of visuospatial perceptual and oculomotor extrapolation. PLoS One 5, e9664, doi:10.1371/journal.pone.0009664 (2010).

13    Tomassini, A., Morgan, M. J. & Solomon, J. A. Orientation uncertainty reduces perceived obliquity. Vision Res 50, 541-547, doi:10.1016/j.visres.2009.12.005 (2010).

14    Raphael, S., Dillenburger, B. & Morgan, M. Computation of relative numerosity of circular dot textures. J Vis 13, 17, doi:10.1167/13.2.17 (2013).

15    Solomon, J. A., Morgan, M. & Chubb, C. Efficiencies for the statistics of size discrimination. J Vis 11, 13, doi:10.1167/11.12.13 (2011).