When children see the world differently: Examining children’s experience of visual illusions

• In Event: Poster Session 12
  In Poster Session: PS 12 Topic - Perceptual, Sensory, Motor

Abstract

Context differentially affects memory across childhood and adolescence (Edgin et al., 2014), which may be due to ongoing hippocampal development (Gogtay et al., 2006; Jabès et al., 2015; Schapiro et al., 2017). The hippocampus and associated regions are also integral to the integration of perceptual information (Barense et al., 2010). The present experiment explores how perceptual information is integrated in visual illusions when hippocampal development is still underway. Before the age of 7, children are less susceptible to the Ebbinghaus illusion (Doherty et al., 2010). In this illusion, two target circles are flanked by surrounding circles that make the inner circles appear misleadingly larger or smaller (Figure 1). One interpretation of this finding is that children are less likely to process the contextual information that flanks the circles and thus do not see the inner circles as larger or smaller than they actually are. Thus, experiencing this illusion may require perceptual integration via the hippocampus, because the target circles are visually disconnected from their surroundings. We therefore contrasted the Ebbinghaus illusion with two other illusions in which the contextual information is directly connected to the target information, and thus additional integration via the hippocampus is unnecessary (Figure 1). In the Sander’s illusion, the target lines are connected with the surrounding lines. In the Vertical-Horizontal illusion, the target lines are connected to each other. In the current experiment, 4- to 10-year-old children (N = 93) and adults (N = 30) viewed versions of the Ebbinghaus, Sander’s, and Vertical-Horizontal illusions in which the helpfulness of the surrounding contextual information varied. On no context trials, target items were presented alone, and thus, there was no surrounding information to provide participants with context. On misleading trials, the surrounding information facilitated the appearance of the illusion, making the target item look misleadingly larger or longer. On helpful trials, the surrounding information highlighted the actual size difference between the targets. Four- to six-year-olds did not perceive the Ebbinghaus illusion and correctly identified the larger circle more often in misleading contexts than 7- to 10-year-olds and adults. Notably, young children also did not show a strong benefit when the context was helpful, providing more evidence that they do not automatically integrate spatially disconnected information when viewing these illusions (Figure 2a). However, participants of all ages were susceptible to the Sander’s and Vertical-Horizontal illusions, indicating that even the youngest participants experience illusions when surround and targets are already integrated (Figure 2b and 2c). Thus, when the target items are visually disconnected from surrounding perceptual information, as in the Ebbinghaus illusion, younger children do not experience illusions. These findings suggest that children younger than 7 years do not automatically integrate spatially disconnected visual information, possibly due to ongoing maturation of neural structures central to these processes, such as the medial temporal lobe and hippocampus.

Authors

• Samantha Gualtieri, University of Toronto
  Presenting Author

• Kay Otsubo, University of Toronto
  Non-Presenting Author

• Asaf Gilboa, University of Toronto
  Non-Presenting Author

• Morgan Barense, University of Toronto
  Non-Presenting Author

• Amy S Finn, University of Toronto
  Non-Presenting Author