Role of Inducer Length in the Formation of the Ehrenstein Illusion

We wanted to study the role of the length of the inducers in the Ehrenstein illusion and, particularly, the possible interdependence of inducer length and gap size in sustaining the illusion.

We used a subthreshold summation paradigm. As described previously (Tassi et al, 1995 Perception 24 127), the presence of an illusory contour reduces the threshold for detection of a real line superimposed on it. In agreement with the model of Grossberg and Mingolla (Grossberg and Mingolla, 1985 Psychological Review 92 173 – 211; Grossberg, 1994 Perception & Psychophysics 55 48 – 120), these authors suggested that the Ehrenstein illusion was generated by a bipole operator joining two line ends. In an Ehrenstein figure induced by four white lines, we varied the size of the central gap from 1.6 to 4 deg. The threshold of a white line joining the ends of two inducers was measured with a spatial alternative (left - right). The Ehrenstein figures involved either long (2.5 deg), or short (0.58 deg) inducers.

The result shows that whatever the length of the inducers, the threshold for detecting the real line increases with increasing gap size. The results obtained with short and long inducers were very similar. Moreover, in this experiment, the spatial limit of integration of the illusion occurs for a gap size of 2.4 deg: the detection threshold for the real line was significantly higher from this value. This result, consistent with previous research (Salvano-Pardieu and Bonnet, 1994 Perception 23 104), suggests that the length of the inducer is a factor less important than the gap size in the formation of the illusion.