Investigating the contribution of metacontrast to the Fröhlich effect for size Elena Carbone a, * , Ulrich Ansorge b a Department of Psychology, Bielefeld University, P.O. Box 100131, D-33501 Bielefeld, Germany b Institute of Cognitive Science, University of Osnabrück, Germany article info Article history: Received 31 October 2007 Received in revised form 19 March 2008 Accepted 20 March 2008 Available online 15 May 2008 PsycINFO classification: 2323 Keywords: Fröhlich effect Metacontrast Size perception Fröhlich effect for size abstract According to the Fröhlich effect, observers perceive the initial position of a fast moving stimulus dis- placed in the direction of motion. On the basis of Kirschfeld and Kammer’s as well as Fröhlich’s original assumption that metacontrast plays an important role in the emergence of the phenomenon, we pre- dicted different amounts of misperception for stimulus enlargement compared to stimulus reduction. These basic predictions were confirmed in Experiment 1. In Experiment 2, we investigated whether an overestimation-bias might account for these results. But the overestimation of non-changing stimuli was too small to adequately explain the dissociation. In Experiment 3, we predicted and found different effects of the factor stimulus lightness on misperception in the enlargement and reduction condition. In Experiment 4, we showed that misperception in the enlargement condition is reduced when frames are used instead of filled stimuli, as in the earlier experiments. Results are discussed with respect to the ori- ginal Fröhlich effect. Ó 2008 Elsevier B.V. All rights reserved. 1. Introduction When indicating the initial position of a fast moving visual stimulus appearing at an unpredictable location, observers make systematic errors: Compared to the actual starting location the perceived starting location is shifted into the motion direction. This perceptual illusion is known as the Fröhlich effect, after the Ger- man physiologist Friedrich W. Fröhlich who discovered it in the early 1920s (Fröhlich, 1921, 1929). Recently, a phenomenon has been discovered that seems to be the mere opposite of the Fröhlich effect: When asked to indicate the first position of the moving stimulus, the observers make errors into the opposite direction. They indicate earlier positions that were not included in the trajec- tory. However, compared to the Fröhlich effect this error occurs with lower velocities and seems to be caused by judgment biases (Kerzel, 2002; Thornton, 2002). Commonly, the Fröhlich effect is explained in terms of visuo- spatial attention (e.g., Carbone & Pomplun, 2007; Kerzel & Gegenfurtner, 2004; Müsseler & Aschersleben, 1998; Müsseler & Neumann, 1992; Müsseler, Stork, & Kerzel, 2002; Neumann & Müsseler, 1990): According to this account, the onset of the moving stimulus at an unpredictable location triggers an atten- tion shift towards its location. During the shifting of attention, the stimulus moves further, and because the stimulus can only be consciously perceived after the attention shift has been com- pleted, a shifted location is erroneously perceived as being the first one. Kirschfeld and Kammer (1999, 2000) proposed a related and yet more sophisticated account that closely resembled Fröhlich’s origi- nal explanation. Like the proponents of the attentional account, they assume that attentional processes play a major role in the emergence of the Fröhlich effect, but attribute the phenomenon to an interplay of visual focal attention and metacontrast masking. The authors begin with the fact that for a moving stimulus each stimulus location is preceded as well as followed by a stimulus at an adjacent location. The only exceptions are the first and the last stimulus locations. For the first stimulus location no preceding adjacent stimulus exists, whereas for the last stimulus location no following adjacent stimulus exists. The temporally preceding adja- cent stimulus on the motion trajectory, S (tÀ1) , is supposed to serve as a cue, which shortens the perceptual latency, intensifies the per- ception and prolongs the perceptual persistence of the current stimulus, S (t) , whereas the temporally subsequent adjacent stimu- lus, S (t+1) , prolongs the latency, diminishes the perceived intensity and shortens the persistence of S (t) by backward metacontrast masking (cf. Breitmeyer & Ogmen, 2006; Kahneman, 1968). According to Kirschfeld and Kammer, the interplay between visual focal attention and metacontrast leads to a shorter latency, a high- er perceived intensity and a shorter persistence for each stimulus location on the motion trajectory except for the first one. For the very first stimulus location at stimulus and motion onset, no 0001-6918/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.actpsy.2008.03.008 * Corresponding author. Tel.: +49 521 106 4516; fax: +49 521 106 8075. E-mail address: elena.carbone@uni-bielefeld.de (E. Carbone). Acta Psychologica 128 (2008) 361–367 Contents lists available at ScienceDirect Acta Psychologica journal homepage: www.elsevier.com/locate/actpsy