Ebbinghaus figures that deceive the eye do not necessarily deceive the hand

Abstract

In support of the visual stream dissociation hypothesis, which states that distinct visual streams serve vision-for-perception and vision-for-action, visual size illusions were reported over 20 years ago to ‘deceive the eye but not the hand’. Ever since, inconclusive results and contradictory interpretations have accumulated. Therefore, we investigated the effects of the Ebbinghaus figure on repetitive aiming movements with distinct dynamics. Participants performed a Fitts’ task in which Ebbinghaus figures served as targets. We systematically varied the three parameters which have been shown to influence the perceived size of the Ebbinghaus figure’s target circle, namely the size of the target, its distance to the context circles and the size of the context circles. This paper shows that movement is significantly affected by the context size, but, in contrast to perception, not by the other two parameters. This is especially prominent in the approach phase of the movement towards the target, regardless of the dynamics. To reconcile the findings, we argue that different informational variables are used for size perception and the visual control of movements irrespective of whether certain variables induce (perceptual) illusions.

Grasping of Real-World Objects Is Not Biased by Ensemble Perception

“…In contrast to the perceptual estimates, the univariate analysis of grasping did not detect any bias of ensemble size and orientation on maximum grip aperture and grip orientation, respectively. The failure of ensemble perception to influence grip aperture is consistent with studies which report that this measure resists the perceptual bias induced by pictorial illusions on targets embedded in them (Brenner and Smeets, 1996;Haffenden and Goodale, 1998;Jackson and Shaw, 2000;Servos et al, 2000;Danckert et al, 2002;Chen et al, 2015;Knol et al, 2017). Note, however, an important distinction between our paradigm and those used when studying the effects of visual illusions on grasping.…”

Dissociating the Influence of Perceptual Biases and Contextual Artifacts Within Target Configurations During the Planning and Control of Visually Guided Action

“…To-date, the findings have been rather mixed with aiming movements appearing to be consistent with the Ebbinghaus/Tichener circles illusion when they are executed in the absence of online visual feedback (open-loop) (van Donkelaar, 1999) as opposed to it being present throughout the movement (closed-loop) (Fischer, 2001) (for similar effects within different tasks and illusions, see Elliott & Lee, 1995;Heath, Rival, & Neely, 2006;Meegan et al, 2004;Westwood & Goodale, 2003). Likewise, there is an increased perceptual bias induced by this same illusion when the aiming movements are discrete as opposed to continuous (Alphonsa, Dai, Benham-Deal, & Zhu, 2016;2017; for alternative findings, see Knol, Huys, Sarrazin, Spiegler, & Jirsa, 2017;Skewes, Roepstorff, & Frith, 2011). These discrepancies may be explained by the ventral pathway more greatly contributing toward preresponse, memory-guided aiming movements, while the dorsal pathway primarily contributes to much smoother, visually-regulated aiming movements.…”

“…Thus, it stands to reason that any perceptual bias within the planning and/or control of movement should manifest in the time to and after peak velocity, respectively. To-date, the findings have shown a minimal perceptual bias associated with the Ebbinghaus/Titchener circles illusion in the time to peak velocity, although it may alternatively unfold in the time after peak velocity (Handlovsky, Hansen, Lee, & Elliott, 2004;Knol et al, 2017). Along these lines, researchers have additionally investigated the role of planning and control by introducing a sudden perturbation to the Ebbinghaus/Titchener circles illusion following movement onset.…”

“…For example, the introduction of same-sized annuli relative to the targetthus negating any size-contrast illusion-can cause an increase in the overall movement time of aiming movements (Roberts, Welsh, & Wakefield, 2019; see also, . In addition, a decrease in the gap size between the target and surrounding annuli without necessarily altering the perceived target size has been known to increase the acceleration time of aiming movements (Knol et al, 2017; see also, Haffenden, Schiff, & Goodale, 2001). Thus, it appears the presence of surrounding annuli and their proximity to the target may create some form of obstacle, which must be avoided by the movement in order to reach the central target (for further examples of context effects, see Coren, 1986;Glazebrook et al, 2005;Searleman, Porac, Dafoe, & Hetzel, 2005).…”

Examining the equivalence between imagery and execution – Do imagined and executed movements code relative environmental features?

“…That is, visual illusions influence perceptual judgements because the relative features are coded in the ventral stream, while the online control of movement fails to reflect the nature of these illusions because their absolute features are coded in the dorsal stream. However, subsequent research findings have heavily attributed at least some influence of perceptual illusions within movement (Franz, Bülthoff, & Fahle, 2002;Glover & Dixon, 2001;Grierson & Elliott, 2009;Handlovsky, Hansen, Lee, & Elliott, 2004;Knol, Huys, Sarrazin, Spiegler, & Jirsa, 2017;Mendoza, Elliott, Meegan, Lyons, & Welsh, 2006;Roberts et al, 2013;van Donkelaar, 1999;Westwood & Goodale, 2003; see Goodale (2011), for an extensive review). For example, when adapting the Ebbinghaus (or Titchener-circles) illusion featuring large or small annuli surrounding the target, aiming movements were found to be longer or shorter in time when facing perceptually small (large annuli) or perceptually large (small annuli) targets, respectively (Handlovsky et al, 2004;Knol et al, 2017;van Donkelaar, 1999).…”