Search

We carried out six experiments to find out whether simple manual reaction time (RT) to flux-equated visual stimuli of different size is modulated by size constancy or by the retinal angle subtended by the stimuli. We found that RT decreased with the increase in perceived stimulus size rather than retinal angle and that this relationship depended on the use of familiar 3-D-like stimuli and on the availability of other size-constancy cues. Thus, a stereotyped speeded motor response, such as that employed in a simple RT paradigm, is modulated by size constancy, as is the case with perceptual judgments. The present results provide original evidence on the relationship between simple RT and perception.

This is an informal account of Sir Frederic Bartlett's department at the time I was a student and started research in the late 1940s and 1950s, developing some themes of his influence and how his ideas, with Craik's, have had a major effect on experimental psychology ever since. As one of the diminishing tribe of those who knew “Sir Fred”, it seems appropriate to make this quite a personal statement, representative of his influence on us, rather than a formal paper. His interests were very great, including not only memory, but thinking and social issues largely based on anthropology, as well as practical problems of selection for skill, and design of machines to be appropriate for humans. A “subliminal” theme is errors of perception. It is this that will be developed here as, following his influence as my teacher, this has been the central interest of my work, which took off from the Cambridge Psychology Department of that time.

Visual illusions provide evidence of perceptual processes, especially the use of knowledge about objects and the working rules for seeing. The classification of phenomena is necessary for any science because it facilitates induction and deduction. The phenomena of illusions are no exception. It is hoped that the tentative classification presented here may stimulate further research and raise new questions by revealing gaps in the conceptual structure of how perception proceeds. Finally, a speculation is offered for the functional use of sensations, the qualia of consciousness.

In the triangle-bisection illusion, a dot is inscribed exactly halfway up the height of an equilateral triangle, but it looks apparently far more than halfway up. The illusion is found for second-order triangles defined by stereo depth and by equiluminous texture. It is strongest for equilateral triangles, and even stronger for concave triangles with curved sides. We propose that the observers are probably responding to the centre of area or centre of gravity of the triangle rather than to its half-height.

A cognitive account is offered of puzzling, though well known phenomena, including increased size of afterimages with greater distance (Emmert's Law) and increased size of the moon near the horizon (the Moon Illusion). Various classical distortion illusions are explained by Size Scaling when inappropriate to distance, 'flipping' depth ambiguity being used to separate botton-up and top-down visual scaling. Helmholtz's general Principle is discussed with simpler wording - that retinal images are attributed to objects - for object recognition and spatial vision.