Anim. Behav., 1998, 55, 787–797 Path integration in dogs VALERIE SE u GUINOT*, JENNIFER CATTET* & SIMON BENHAMOU² *L aboratoire d’Ethologie, Universite ´ de Gene `ve ² CRNC, CNRS, Marseille ( R eceived 16 A pril 1997; initial acceptance 16 M ay 1997; final acceptance 15 July 1997; M S. number: 5523) Abstract. We studied path integration in dogs, Canis familiaris, required to return to the starting point of L-shaped outward paths in the absence of landmark information. The analysis of the distributions of errors in the return direction and distance showed that the dogs made both systematic and random errors. They tended to overestimate by about 6 the amount of turning required at the end of the outward path to take the correct return direction. Random angular errors were characterized by a standard deviation of about 9. The dogs also tended to underestimate the correct return distance by about 6%, and random distance errors were characterized by a standard deviation of about 13% of the correct return distance. While random errors are assumed to stem simply from inaccuracies in kinaesthetic and vestibular systems, systematic errors are likely to be generated because path integration is implemented at the neural level through approximate rather than mathematically exact solutions. A path integration model in which the moving animal is assumed to overestimate slightly the inferred motion parallax of the starting point can roughly account for the dogs’ directional bias, but not for systematic errors in distance estimation.  1998 The Association for the Study of Animal Behaviour Path integration refers to the process by which an animal updates the direction and distance of the starting point of its outward journey relative to its own current position by processing route-based information, that is, information collected en route about the running changes of its position (rotations and translations). This ability has been convincingly demonstrated in numerous taxa (reviews in Benhamou & Poucet 1995; Maurer & Se ´guinot 1995; see also Benhamou 1997a for theoretical considerations). Path integration in mammals has been mainly studied in rodents such as gerbils, M eriones unguiculatus (Mittelstaedt & Mittelstaedt 1980, 1982), hamsters, M esocricetus auratus (Etienne et al. 1985, 1986, 1988, 1991; Se ´guinot et al. 1993), mice, M us musculus (Alyan & Jander 1994; Alyan 1996) and rats, Rattus norvegicus (Potegal 1982; Benhamou 1997b ). Very few empirical data, however, are available for other mammalian species. The ability to rely on path integration can be easily shown in an animal required to return along a direct path to the starting point of its outward path in the absence of landmark information. Hymenoptera such as desert ants, Cataglyphis fortis, which are able to rely on a sun (or polariz- ation skylight pattern) compass, are very efficient in performing this return task (e.g. see Wehner & Wehner 1990). Indeed, using allothetic (compass- based) information to estimate the changes of direction during the outward path prevents the updating mechanism from accumulating random errors (Benhamou et al. 1990). Furthermore, the sun compass also enables ants to keep a suitable heading during the return path, as shown by Santschi’s (1911) famous mirror experiment in which ants were induced to reverse their homing direction when the apparent sun position was shifted by 180. When desert ants are forced to walk an outward path in which right and left turns are not balanced, however, they tend to make a systematic error in determining the return direc- tion (Mu ¨ller & Wehner 1988). Such an angular error appears clearly after an L-shaped outward path: the ants tend to turn too much at the end of the outward path (just before returning to the Correspondence: S. Benhamou, Centre de Recherche en Neurosciences Cognitives, Centre National de la Recherche Scientifique, Marseille, F -13402 Marseille cedex 20, France (email: simon@lnf.cnrs-mrs.fr). V. Se ´guinot and J. Cattet are at the Laboratoire d’Ethologie, Universite ´ de Gene `ve, 54 rue des Acacias, CH-1227 Carouge, Switzerland. 0003–3472/98/040787+ 11 $25.00/0/ar970662  1998 The Association for the Study of Animal Behaviour 787