Purchase this article with an account.
Ashli F. Milling, Paul C. Knox; Perception And Oculomotor Actions: No Evidence Of A Dissociation With The Titchner Circles Illusion. Invest. Ophthalmol. Vis. Sci. 2011;52(14):4695. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Debate continues as to whether perception/action dissociations can be demonstrated in healthy individuals using visual illusions. With the Muller-Lyer and Kanizsa compression illusions, oculomotor actions (saccades) are affected in the same way and to the same extent as perception. A recent metanalysis of published Muller-Lyer studies (Bruno et al, Vis. Res. 50:2671) suggested that once saccade-specific factors were taken into account, no dissociation between perception and saccades was present. Perhaps a different illusion might induce misperceptions while not affecting saccades.
Ten subjects were exposed to the Titchener circles illusion (targets placed on the horizontal meridian of a single circle, CON, are perceived as being closer than those placed with the same separation on two separate circles, EXP). Iin runs of interleaved trials, after a randomised fixation time leftward and rightward contracting and expanding stimuli of two amplitudes (separations: 2° and 4°) and 4° control stimuli (two point targets) were presented. One target was always presented at fixation and stimulus display time was 200ms. In the perceptual experiment subjects verbally reported the distance from fixation to the eccentric target relative to standard line (length "100") presented after the stimulus disappeared. In the saccade experiment, eye movements (saccades from fixation to the eccentric target) were recorded using infra-red oculography. Latency and amplitude were measured offline.
We observed the expected illusion effect on perception (F1,10=50; p<0.001) and in the saccade experiment a statistically similar result (F1,10=21; p=0.001). The amplitude of the illusion effect size on saccades at 4° was slightly reduced compared to perception, but not statistically different (Sac: 7±6% vs Percept:12±8%; paired t test = 1.1, p=0.14).
These data suggest that, as with other illusions we have tested, the Titchener circles illusion modulates both perception and saccade amplitude, providing little evidence of a perception/action dissociation.
This PDF is available to Subscribers Only