The Ebbinghaus illusion in a fish (Xenotoca eiseni)

Abstract

The tendency of fish to perceive the Ebbinghaus illusion was investigated. Redtail splitfins (Xenotoca eiseni, family Goodeidae) were trained to discriminate between two disks of different sizes. Then, fish were presented with two disks of the same size surrounded by disks of large or small size (inducers) arranged to produce the impression (to a human observer) of two disks of different sizes (in the Ebbinghaus illusion, a central disk surrounded by small inducers appears bigger than an identical one surrounded by large inducers). Fish chose the stimulus that, on the basis of a perception of the Ebbinghaus illusion, appeared deceptively larger or smaller, consistent with the condition of training. These results demonstrate that redtail splitfins tend to perceive this particular illusion. The results are discussed with reference to other related illusions that have been recently observed to be experienced by fish (such as the Navon effect), and with regard to their possible evolutionary implications.

Why do animals differ in their susceptibility to geometrical illusions?

“…Size-classification task Reversed susceptibility Sovrano, Albertazzi, and Rosa Salva (2015) 8 Xenotoca eiseni (redtail splitfin fish) Two-choice discrimination task Yes The arrows on the ends of the lines are similar to the corners on the inside and outside of a building or room, such that the two target lines are perceived as being at different distances, and thus different sizes Pressey (1972) Assimilation theory The '> <' arrows at the ends of one line make it appear longer than the line with arrows facing outward '< >' since the overall figure is longer Howe and Purves (2005) Probabilistic theory In natural scenes, '> <' arrows are more likely to indicate longer lines, suggesting that the Müller-Lyer illusion could be due to a probabilistic strategy of visual processing Ginsburg (1984) Selective filtering theory We judge visual information with lower spatial frequencies as being further away, causing the inner line of the '> <' arrow to be perceptually rescaled as longer inward pointing arrows '> <' were attached and shorter when outward pointing arrows '< >' were attached (Table 2). This potentially argues against Gregory's (1963) inappropriate constancy-scaling theory, since few of the species tested are likely to have been exposed to modern architecture.…”

To Err Is (Not Only) Human: Fallibility as a Window Into Primate Cognition

“…This might suggest human-uniqueness in seeing this illusion, as would the reported reversed Ebbinghaus-Titchener illusions in pigeons (Nakamura, Watanabe, & Fujita, 2008) and bantam chickens (Nakamura, Watanabe, & Fujita, 2014). However, the story again is not that simple, as other nonprimate species have shown evidence of perceiving this illusion in the direction shown by humans (dolphin: Murayama, Usui, Takeda, Kato, & Maejima, 2012;domestic chicks: Rosa Salva, Rugani, Cavazzana, Regolin, & Vallortigara, 2013; redtail splitfin fish: Sovrano, Albertazzi, & Rosa Salva, 2015). And, as outlined in more detail next, our work suggests that monkeys show evidence of a related illusion called the Delboeuf (1865) illusion (Parrish, Brosnan, & Beran, 2015).…”

Do Dogs See the Ponzo illusion?

“…Variation in illusion susceptibility is not unique to dogs. A variety of studies have found humanlike susceptibility, no susceptibility, or reversed susceptibility in animals (e.g., Agrillo, Parrish, & Beran, 2014;Fujita, 1996Fujita, , 1997Murayama, Usui, Takeda, Kato, & Maejima, 2012;Nakamura, Watanabe, & Fujita, 2008;Sovrano, Albertazzi, & Salva, 2014;Watanabe, Nakamura, & Fujita, 2011, 2013. Additionally, even within species mixed evidence has been observed in regards to susceptibility to the same illusion (e.g., Nakamura et al, 2008;Nakamura, Watanabe, & Fujita, 2014;Salva, Rugani, Cavazzana, Regolin, & Vallortigara, 2013).…”