Answers are grounded in this paper and cite their sources.
Two Grey parrots (Psittacus erithacus) were tested on various types of mirror use: mirror image stimulation, mirror-mediated object discrimination, and a simple form of mirror-mediated spatial locating. During exposure to a mirror, neither bird clearly demonstrated self-exploratory behavior but responded instead in ways similar to those of marmosets, monkeys, dolphins, extremely young children(<18 months), and to the initial responses of orangutans and young chimpanzees. The parrots' behavior was not a consequence of an inability to process mirrored information, because in subsequent tasks they used mirrors to discriminate among exemplars and to locate hidden objects; these birds are the first nonmammalian subjects to exhibit all these behavior patterns. Their behavior on all the tasks can be compared to that of humans, great apes, dolphins, monkeys, and Asian elephants. Mirror studies generally have two goals. One goal is to determine whether subjects demonstrate selfrecognition (e.g. Anderson. 1984; Gallup. 1970; Povinelli. Rulf, Landau, & Bierschwale, 1993; Swartz & Evans, 1991) and, by extension, self-awareness and advanced social cognition (e.g., Gallup. 1991; Mitchell, 1993). A second goal is to learn whether subjects that fail standard tests of self-recognition do so because they lack capacities for processing mirror information (e.g. Menzel, Savage-Rumbaugh, & Lawson, 1985; Povinelli, 1989). Studies usually include mirror-image stimulation, mirror-triggered search, mirror-mediated object discrimination, mirror-mediated spatial locating, and mirror-guided reaching. After describing what each of these tasks requires of a subject, we suggest a third goal for mirror studies: to provide data for cross-species comparisons of cognitive and perceptual abilities. During mirror-image stimulation, subjects generally view a reflection of their entire body (Amsterdam, 1972; Gallup. 1970). The length and number of mirror exposures varies among studies; data consist of the type, number, and sometimes length of the observed mirror interactions. Mirror-image stimulation is most often used to test self-recognition, but it also provides data on the developmental stage of children (Amsterdam, 1972) and. for animals, on how extensively visual (rather than acoustic or olfactory) cues mediate interactions in a given species (Anderson, 1984). Mirror-triggered search is the most basic task in which mirrors are used to find hidden objects (Povinelli. 1989). Subjects learn that desirable items are likely found in a few fixed locations in the presence of a mirror. The subject need not correlate the object's location and information in the mirror; a mirror is merely a cue to begin a search (see Anderson. 1986; Menzel et al., 1985). Subjects who engage in mirrortriggered searches are not necessarily able to use mirrors to locate, without trial and error, rewards that are hidden in novel sites. In mirror-mediated object discrimination, unlike mirror-triggered search, a subject uses a mirror to choose between objects (e.g., hidden rewards and aversive stimuli). A subject must recognize some correlation between an object and its reflection (Menzel et al.. 1985), but the subject need not understand that the image is a representation of the actual object nor use the mirror to monitor its actions. Subsequent appropriate responses to novel positive and negative objects in familiar locations rule out the possibility of mirror-triggered search. Mirror-mediated spatial locating and mirror-guided reaching require more advanced capacities. In the former task, subjects use mirrors to locate, without trial and error, items hidden in novel locations. Here a subject shows that it understands the correspondence between the location of the object in real space and reflected information (Povinelli, 1989). Only in the latter task, however, must a subject also relate its own movements to those depicted in the mirror (e.g., a sequence of fine hand movements; Itakura, 1987; Menzel et al., 1985; Povinelli, 1989). Mirror use thus examines not only self-recognition but also how animals respond to reflections and process information about spatially displaced objects and actions. Specifically, because various mirror tasks require different levels of information processing, we suggest that mirror studies can provide a hierarchical scale for assessing and comparing perceptual and cognitive abilities of diverse species. Animals that, for example, succeed on mirror-mediated discriminations but fail tests of mirror-mediated locating demonstrate specific cognitive deficits compared with those that pass both tests. Of particular interest are comparisons between mammals and nonmammals because of their neurological and anatomical differences. Several researchers have examined mirror use in nonhuman mammals. Mirror-image stimulation has been studied in pygmy marmosets (Cebuella pygma; Eglash & Snowdon, 1983); elephants (Eleplws maximus; Povinelli, 1989); dolphins (Tursiops truncatus; Marino, Reiss, & Gallup, 1994); chimpanzees (Pan troglodytes; Gallup, 1970); orangutans (Pongo pygmaeus; Robert, 1986; Suarez & Gallup, 1981); and various monkey species (see reviews in Anderson, 1984, and Gallup, 1991). Although only some of the great apes unequivocally demonstrate self-recognition (see review in Anderson, 1993), other mammals can use mirrored information in various ways. Elephants (Povinelli, 1989) and monkeys (e.g., Cebus capucinus, Marchal & Anderson, 1993; Macaca mulatta, Anderson) 1986; Menzel et al., 1985; Macaca fuscata fuscata, Itakura, 1987) were tested on mirror-triggered search and spatial locating, and monkeys can perform mirror-mediated object discriminations. Monkeys (Itakura, 1987; Menzel et al., 1985) and chimpanzees (Menzel et al., 1985) have been tested on mirror-guided reaching. The question now is whether such capacities are limited to mammals. With the exception of mirror-image stimulation studies that failed to show self-recognition (budgerigars [Melopsittacus undulatus] and house sparrows [Passer domesticus domesticus], Gallup & Capper, 1970; a kea [Nestor notabilis], Diamond & Bond, 1989; chickadees [Parus atricapillus], Censky & Ficken, 1982; zebra finches [Poephila guttata], Ryan, 1978; glaucous-winged gulls [Larus glaucescens], Stout, Wilcox, & Creitz, 1969; blue grouse [Dendragapus obscurus], Stirling, 1968; cedar waxwings [Bombycilla cedrorum] and juncos [junco hyemalis], Andrews, 1966; and lovebirds [Agapornis roseicollis], Delsaut & Roy, 1980), no comparable work has been performed on birds. Specifically, can subjects whose brains are organized so differently from those of mammals (Striedter, 1994) and who lack hands and comparable feet (and thus certain fine motor skills) use mirrored information in a manner similar to that of mammals? Given that a Grey parrot (Psittacus erithacus) has demonstrated cognitive capacities similar to those of nonhuman primates and marine mammals (e.g., comprehension of concepts of category, number, same-different, absence of information, relative size; Pepperberg, 1990a, 1990c, 1992, 1994a; Pepperberg & Brezinsky, 1991), this species would be a good subject for studies of avian mirror use. We therefore tested two Grey parrots on mirror-image stimulation, mirror-mediated object discrimination, and mirror-mediated spatial locating. Each study was designed both to test for capacities comparable to those of mammals and to accommodate avian anatomy. Experiments on mirror-guided reaching form the basis of a separate study (see General Discussion). Experiment 1. Mirror-Image Stimulation Our first goal was to examine our subjects' reactions to their mirror images. Even if they failed to exhibit those self-exploratory reactions that provide the rationale for conducting a test for self-recognition (Gallup & Povinelli, 1993), we could compare our birds to mammals with respect to social, aggressive, and search behavior. Too, because mirror use may develop with experience (Amsterdam, 1972; Anderson, 1984; Field & Hogg, 1992; Gallup, 1987), our subjects required exposure to mirrors before we tested skills involving discrimination and locating.
No takes yet. Share an insight, caveat, or question.
Pepperberg et al. (1995) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: