知覺與錯覺

During my time at UC San Diego, I was fortunate to have taken many classes taught by Stuart Anstis, VS Ramachandran, and Diana Deutsch. Their approach in using illusions as a tool to reverse-engineer the perceptual system left a deep impression in me. I remember being bewildered by all the visual and auditory illusions that were shown in class, and decided that psychology would be my choice of career. I am now also teaching the Sensation & Perception class at TMU. 

Here at TMU I also lead a group of undergraduate students to investigate the different conditions that are necessary for various visual/auditory illusions to occur. Some examples can be found on Yi-Tsen Kuo's website here. For example, our team created a one-of-its-kind visual illusion called the Sequin Illusion (youtube demo), which made it into the top 10 finalists in the 2018 Illusion of the Year Contest. We later published our full-length investigation of it, hypothesizing that although it looks like Hermann grid in appearance, the two illusions actually do not have much in common and mostly likely orginate from different neural mechanisms (with the Sequin Illusion occuring earlier in the feedforward stream).

My former graduate student Amit Rawal and I also investigated one variation of the fat face illusion, where when two identical pictures of a face are stacked vertically, the upper face appears thinner. Using a series of rotated pictures, we demosntrate that such directionality is not always true as it is the result of face contour/geometry. This suggests that maybe the fat face illusion is really face-specific at all, and indeed, we demonstrate that the same thinning illusion can appear in stacked houses as well. My current graduate student Yun is also working on a series of studies on using Diana Deutsch's tritone paradox. 

Lastly, using a bistable silhouette spinner animation and eye tracker we have also found that most participants feel that they can use their willpower to change the spinner's 'switch' rate, and they are indeed correct, but what they don't know is that they are doing so by selectively looking at the body or the feet of the spinner, which decreases and increases the switch rate, respectively, for some reason (Liu et al., 2012).

People (including researchers) often ask me what we can actually get out of illusion research besides a good laugh or "wow" moment. Contrary to their belief, the world of illusion is not only fun, but also useful. I think of this in two ways: microscopically, understanding of how illusions occur can help us design better user interfaces, either by taking advantage of illusions or by preventing it from happening. But macroscopically, and perhaps more importantly, by learning about the "mistakes" that our visual system makes, we learn about our brain's "prioritities" and "preferences". And, by learning about these priorities and preferences, we learn about its evolutionary past, and how such past gave rise to what we have (as well as the mistakes that we make) today. Therefore, illusions present an unique opportunity for us to get a snapshot of our history. 

For some still-image demos, see https://yi-tsen.wixsite.com/illusions 

Kuo, Y-T. & Tseng, P. (2019). The Sequin Illusion. i-Perception, 10(6), 1-7.. [PDF]

Rawal, A. & Tseng, P. (2020). A geometrical account to explain the fat-face illusion. i-Perception, 11(6), 1-11. [PDF]

Liu, C.H., Tzeng, O.J.L., Hung, D.L., Tseng, P., Juan, C.H. (2012) Investigation of bistable perception with the "silhouette spinner": Sit still, spin the dancer with your will. Vision Research, 60, 34-39. [PDF]

Bridgeman, B., Winter, D., Tseng, P. (2010) Phenomenology of grapheme-color synesthesia. Perception, 39, 671-676. [PDF]