Behavioral The official journal of the Ecology ISBE International Society for Behavioral Ecology Behavioral Ecology (2014), 25(3), 464–469. Invited Commentaries Confusion and illusion: understanding Merilaita 2009). If nothing has been detected because the visual visual traits and behavior. A comment on system fails to differentiate object versus background, then presum- Downloaded from https://academic.oup.com/beheco/article/25/3/464/510898 by guest on 28 September 2021 Kelley and Kelley ably there is no perception of the shape of the object, because the viewer is unaware of that object’s existence. A further example Martin Stevens is distractive camouflage. Kelley and Kelley outline work testing Centre for Ecology and Conservation, University of Exeter, Penryn whether distractive markings draw the viewer’s eye to high-contrast Campus, Penryn TR10 9FE, UK nonmarginal markings, thus preventing the observer from detect- ing the body outline. However, as normally described, this does not Most of us have been fooled by visual illusions. Thousands exist to seem to constitute an illusion because it refocuses gaze or attention fool human perception, creating the impression of objects that do toward one object feature at the expense of another by manipulat- not exist, are distorted in size, shape, or color, or suggest movement ing attention, although Kelley and Kelley’s work may challenge us where none exists. Whether animals have specifically evolved illusion- to think more carefully about how distractive features may work. ary traits to exploit others is not a new suggestion, but Kelley and It is also not clear whether traits that manipulate receiver behav- Kelley’s (2014) paper is probably the first to systematically discuss ior or overstimulate/inactivate receiver responses should be exam- such a concept and to present a clear review of where illusions may ples of illusions. Clearly, there is likely to be a continuum and so exist in nature and how they might work. Their paper brings to the pigeon-holing can be artificial, but it is important to be clear in forefront this exciting and significantly understudied area of biology. order to carefully guide future work. For example, if the circular Here, I focus mostly on visual illusions and communication. eyespots found on many animals that Kelley and Kelley discuss Gregory (1998) discusses illusions in depth and describes them as do mimic eyes (although supporting evidence is currently lacking; occurring when perception departs from the reality of the external Stevens and Ruxton 2014), should they and other types of mim- world. Akin to Gregory, Kelley and Kelley define illusions as traits icry be considered as illusions? I argue that mimicry would usually that “…act to distort the perception of the viewer such as the size, not constitute an illusion because it involves a direct resemblance to shape, coloration or movement…” of individuals or parts of individ- another object, and this is not the same as distorting perspective or uals. How we define illusions is a fundamental factor in how future creating the impression of something that is not really there. work may proceed and how we may interpret many animal traits. The above debates matter because if we are to understand com- Most of the potential examples of illusions that Kelley and munication and signals we need to comprehend the mechanisms Kelley (2014) present, I am in agreement with. For example, they by which they work. The issue of what is an illusion is not simple discuss potential cases involving the perceived size or color of sig- because perception is something quite specific to the sensory sys- nals and individuals, including where well known illusions exist (e.g., tems and brains of each animal. At the moment, more behavioral Ebbinghaus illusion). Likewise, illusions of perspective, such as in experiments are needed to demonstrate that other species really do recent bowerbird courtship studies (Endler et al. 2010), or contour perceive illusions, especially of more naturalistic stimuli. Kelley and completion in animals, seem well explained as illusions. Kelley and Kelley’s (2014) definition seems sensible and a good start in deter- Kelley also discuss motion dazzle, whereby the presence of stripes or mining which traits are illusions. zigzag markings found on fish, snakes, and some other animals could Finally, as Kelley and Kelley (2014) point out, arrangements of distort the observer’s ability to judge the speed and direction of object color and brightness can affect the perception of color patches movement (e.g., Stevens et al. 2008). Along with deterring biting flies, adjacent to one another. We need to pay more attention to the motion dazzle is a leading contender to explain the illusive function of composite arrangement of animal and plant markings, rather zebra stripes, and recent modeling of motion vision shows that zebra than focusing on individual elements in isolation. Studies of com- stripes may interfere with motion assessment (How and Zanker 2013). munication and signaling commonly analyze the color of a single Despite Kelley and Kelley’s (2014) valuable attempts to define patch and correlate this with some aspect of behavior and then and conceptualize illusions and relate them to animal traits and infer what information the receiver may have extracted. Yet, if sig- behavior, conceptual refinement is still needed in terms of what nal color patches operate synergistically, then this approach could animal traits we consider as illusions. This is no bad thing because misinterpret the information available and underestimate the effi- it can lead to future discussion and study. For example, Kelley and cacy of the signal. In short, we need to consider the 2D and 3D Kelley suggest that disruptive coloration, which works by destroying properties of animal signals, and studies of the strategic aspects the outline of a body or body parts, could be viewed as an illusion of signaling need to more commonly consider issues of signal effi- of shape. Although I can see how this may be a possibility were dis- cacy and multicomponent communication. In conspicuous signals, ruptive coloration to work by preventing recognition, it is normally for example, we would expect adjacent color patches that stimu- regarded as a type of crypsis that prevents detection (Stevens and late different opponent color channels or sets of receptors (such as © The Author 2014. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: [email protected] Stevens • Confusion and illusion 465 ultraviolet-blue patches found next to fluorescent yellow patches on receivers. Are they really illusions, effects that act to distort the perception some parrots; Pearn et al. 2001). In short, understanding the colors of the viewer as defined by Kelley and Kelley (2014), or more simply present in signals requires considering the coloration of the animal optimization of signal efficacy? as a whole (and the background). Clearly, there is much left to do to One may wonder if the concept of illusion is different from the understand visual traits. exploitation of perceptual biases developed by Schaefer and Ruxton (2009), which also considers sensory, neuronal, and higher cogni- tive processes involved in the evolution of traits of the sender that FUNDING exploit perceptual biases of the receiver? Maybe not, even if this M.S. was supported by a Biotechnology and Biological Sciences model has mainly been used to explain deception by mimicry, it Research Council (BBSRC) David Phillips Research Fellowship can also be applied to the examples of illusions advanced by Kelley (BB/G022887/1). and Kelley (2014). The concept of animal illusions is within the range of sensory exploitation, bias, and traps. However, those Downloaded from https://academic.oup.com/beheco/article/25/3/464/510898 by guest on 28 September 2021 I thank G. Ruxton for helpful comments on a version of this paper. have mainly been used in the study of female preferences and mate choice, whereas illusions in their present broad sense also Address correspondence to M. Stevens. E-mail: [email protected]. encompass nonsexually selected traits like disruptive coloration, countershading, distractive markings, or motion dazzle. This con- Received 13 January 2014; revised 15 January 2014; accepted 15 January 2014; Advance Access publication 17 February 2014. cept, to my opinion, could even be extended to several other cases and sensory modalities, for example, from aggressive mimicry of doi:10.1093/beheco/aru013 bolas spiders which create the illusion of presence of female moths Forum editor: Sue Healy by mimicking their pheromones to prey on males (Stowe et al. 1987), to aggressive mimicry of katydids which create the illusion REFERENCES of receptive female presence by mimicking their wing sounds to capture males (Marshall and Hill 2009) or even to orchids which Endler JA, Endler LC, Doerr NR. 2010. Great bowerbirds create the- attract pollinators by visual and olfactory illusions of food, mates, aters with forced perspective when seen by their audience. Curr Biol. and oviposition sites (Jersakova et al. 2006; Gaskett 2011). 20:1679–1684. Gregory RL. 1998. Eye and brain: the psychology of seeing. Oxford: Using the term illusion brings a focus on the neuronal, psycho- Oxford University Press. logical, and cognitive processes implicated in visual perception of How MJ, Zanker JM. 2013. Motion camouflage induced by zebra stripes. naturally and sexually selected traits, which are rarely considered Zoology. doi: 10.1016/j.zool.2013.10.004. in nonhuman communication. This approach will certainly con- Kelley LA, Kelley JL. 2014. Animal visual illusion and confusion: the importance of a perceptual perspective. Behav Ecol. 25:450–463. tribute to significant advances in the field, like the recent work of Pearn SM, Bennett ATD, Cuthill IC. 2001. Ultraviolet vision, fluorescence motion dazzle studied on locust neurons by Santer (2013).