Does Octopus vulgaris Have Preferred Arms? Ruth A. Byrne, Michael J. Kuba, and Daniela V. Meisel Konrad Lorenz Institute for Evolution and Cognition Research Ulrike Griebel University of Memphis Jennifer A. Mather University of Lethbridge Previous behavioral studies in Octopus vulgaris revealed lateralization of eye use. In this study, the authors expanded the scope to investigate arm preferences. The octopus’s generalist hunting lifestyle and the structure of their arms suggest that these animals have no need to designate specific arms for specific tasks. However, octopuses also show behaviors, like exploration, in which only single or small groups of arms are involved. Here the authors show that octopuses had a strong preference for anterior arm use to reach for and explore objects, which points toward a task division between anterior and posterior arms. Four out of 8 subjects also showed a lateral bias. In addition, octopuses had a preference for a specific arm to reach into a T maze to retrieve a food reward. These findings give evidence for limb-specialization in an animal whose 8 arms were believed to be equipotential. Keywords: cephalopod, octopus, behavior, arm preference With their obvious lack of skeletal structure (Kier, 1988) and seeming lack of organization, octopus arms are fascinating. Mather (1998) provided a catalogue of possible movements of octopus arms. In addition, octopus arms have even been used as models for robotics (Sumbre, Gutfreund, Fiorito, Flash, & Hochner, 2001; Yekutieli et al., 2005). The combined capability of flexibility and precision is of special interest to neurobiologists and robotic en- gineers (Gutfreund et al., 1996), but so far no one has looked for coordination of arm use or arm preferences in octopuses. Because all eight arms of an octopus are potentially capable of fulfilling the same tasks (apart from mating), one would not necessarily expect a division of labor. However, there could be behavioral asymme- tries along left and right as well as anterior and posterior dimen- sions or a preference for specific arms. In the last 35 years, the phenomenon of limb specialization, especially lateralization, has been extensively studied in verte- brates (for reviews see Bisazza, Rogers, & Vallortigara, 1998; Bradshaw & Rogers, 1993; Hellige, 1993; Rogers & Andrew, 2002; Vallortigara & Rogers, 2005). Studies in primates have included those by McGrew and Marchant (1999) for chimpanzees and Rigamonti, Prato Previde, Poli, Marchant, and McGrew (1998) for macaques. For reports on other mammals, see the studies by Iwaniuk and Whishaw (1999) on raccoons and Waters and Denenberg (1994) on rodents. Birds also exhibit lateral limb preferences, referred to as “footedness” (Rogers & Workman, 1993). Bisazza, Cantalupo, Robins, Rogers, and Vallortigara (1996) showed a behavioral asymmetry in forelimb usage in toads. In addition, Robins, Lippolis, Bisazza, Vallortigara, and Rogers (1998) demonstrated that toads show footedness in turning over after being placed on their backs. Recently, Bisazza, Lippolis, and Vallortigara (2001) also claimed lateralization of ventral fin use in fish for exploration. Evidence for handedness or limb preference in invertebrates is scarce. However, Ades and Novaes Ramires (2002) recently found that spiders collected in the field had legs missing pre- dominantly on their left side and argued that this happens because of defensive and predatory episodes. Additional obser- vations in the lab confirmed that this species of spiders used their left frontal legs more frequently in handling their prey than their right frontal ones. Crustaceans also show anatomical and functional specialization of their legs (Ruppert, Fox, & Barns, 2004). For example, male fiddler crabs exhibit a claw dimor- phism, with one cheliped enlarged, which they use to fight other males and as an ornament to court females (Oliveira & Custo- Ruth A. Byrne, Michael J. Kuba, and Daniela V. Meisel, Konrad Lorenz Institute for Evolution and Cognition Research, Altenberg, Austria; Ulrike Griebel, Department of Biology, University of Memphis; Jennifer A. Mather, Department of Psychology, University of Lethbridge, Lethbridge, Alberta, Canada. Michael J. Kuba is now at the Interdisciplinary Center for Neuronal Computation, Hebrew University of Jerusalem, Jerusalem, Israel. This work was carried out in and financed by the Konrad Lorenz Institute for Evolution and Cognition Research as well as supported by the science funds of the Land Niedero¨sterreich. Octopuses were provided by the Statione Zoologica di Napoli with the help of Flegra Bentivegna and Gianfranco Mazza. We thank Janja Che, Eva Walcher, and Alice Ruf- fingshofer for their help in the lab; Irene Zweimu¨ller and Astrid Ju¨tte for assistance with statistical questions; and Werner Callebaut, Gerd B. Mu¨ller, and Wayne Christiansen for their comments on the article. Many thanks also to Brian M. Byrne for proofreading the article. Special thanks to Ludwig Huber, University of Vienna, for co-supervising this project. Correspondence concerning this article should be addressed to Ruth A. Byrne, who is now at the Department of Biology, Millersville University, Millersville, PA 17551. E-mail: ruth@byrne.at Journal of Comparative Psychology Copyright 2006 by the American Psychological Association 2006, Vol. 120, No. 3, 198 –204 0735-7036/06/$12.00 DOI: 10.1037/0735-7036.120.3.198 198