“Self-awareness” in the pigeon: A replication
Any questions may be directed to Robert W. Allan, email@example.com. Thanks for browsing.
Gallup (1970; 1984) and others have argued that behavior directed towards bodily marks visible only in a mirror may serve as an operational definition of “self-awareness.” According to this definition, Gallup would claim that only humans, chimpanzees, orangutans, and perhaps gorillas may be labeled “self-aware.”
1. Gallup’s procedure:
While the animal is anesthetized, the upper brow and ear are smeared with rouge. Upon awakening, the animal is exposed to a mirror. Experimenters observe whether the mirror occasions species-specific social behavior. If the animal instead investigates its body and touches the rouge-marked location, such behavior is interpreted as evidence of so-called “self-awareness.”
In 1981, Epstein, Lanza and Skinner published a Science paper in which they showed the spontaneous appearance of mirror-mediated, self-directed behavior in pigeons.
2. Epstein, Lanza, & Skinner’s (1981) procedure:
In successive phases, two repertoires were trained: (a) localizing the origin of blue stimuli briefly reflected in a mirror, and (b) in the absence of a mirror, pecking at stick-on dots pasted to the feathers. The first repertoire familiarized the animals with the properties of a reflective surface, while the second established a history of self-directed behavior under “no mirror” conditions.
In subsequent tests, the birds met Gallup’s “self-awareness” criterion. A blue dot was placed on the pigeons’ feathers and covered with a cotton bib. As a control for tactile stimulation from the bib or the dot, movements toward the bib were counted while the mirror was covered. No such movements were observed. Once the mirror was uncovered, however, the dot under the bib was visible in the mirror as long as the birds stood upright and their beaks pointed toward the mirror. Spontaneously and without food present, pigeons emitted the same mirror-mediated, self-directed behavior Gallup had defined as evidence of “self-awareness.”
Subsequent replications of Epstein, Lanza, and Skinner (1981) remained either unpublished (e. g., Cheney’s unpublished successful replication cited in Epstein ), or unsuccessful (Thompson & Contie, 1986). Continuing discussions of “awareness” (Dymond & Barnes, 1997) misinterpret the procedures of Epstein, Lanza, and Skinner.
Cardinal (aka, Drossel), Allan, Barnes, and DeLabar (1998) reported a successful replication of the original Epstein, Lanza, and Skinner study and provided a clear demonstration of spontaneous, mirror-mediated, self-directed behavior in the pigeon. A new chamber was constructed that allowed controlled analyses of the mirror-mediated behavior. Photographs of the chamber as it is presently configured are offered below.
A further replication, together with additional data derived from a series of control conditions, waspresented at the Association for Behavior Analysis Annual Conference, Chicago, Illinois (1999), in an attempt to clarify the contributions of the several trained repertoires to successful mirror-mediated, self-directed behavior in the pigeon.
Figure 1. View of the chamber from the hopper side. The extensive wiring allows for computer control of keylights, food hopper, and houselight. Additional wiring allows for computer-monitoring of a floorboard panel closure, infrared hopper entry, and key pecking. The chamber is 56 cm long x 29 cm wide.
Figure 2. Details of wiring, including 4 electromechanical relays controlling keylight presentations. Also depicted (in the midst of all the wires) is a standard BRS/LVE food hopper. In the background the optically-isolated input and output relays may be seen just in front of the computer keyboard.
Figure 3. Looking from above the chamber enclosure the response keys may be seen hanging on either Plexiglas wall. The keys have been designed to be moved gradually closer to the mirror and are angled so that the key surfaces may be seen from a distance only by looking into the mirror. The reflections of those key panels can be seen in the mirror positioned opposite the hopper wall.
Figure 4. With the vinyl floor mat removed, details of the spring-loaded floorboard panel may be seen clearly. The microswitches are designed to detect the position of the pigeon in the chamber.
Figure 5. A less oblique view of the mirror surface (see Figure 3) showing reflections of both key panels as well as the hopper opening.
Figure 6. After more than 1.5 months training, a partition was inserted between the left and right banks of keys to prevent the pigeons from seeing both key banks simultaneously. The number of correct entries (left or right) was used to judge accurate mirror use; i.e., if the pigeons entered the left side when an upper or lower-left stimulus was present, then a correct response was counted. If the pigeons entered the side without a stimulus before entering the correct side, an error response was counted.
Figure 7. Top view clearly depicting the partition separating the left and right stimulus-key banks. The mirrored wall is at the top of the photograph with key banks placed close to the mirror surface. The partition prevents simultaneous viewing of left and right key banks, thereby forcing the pigeons behavior toward the side with an illuminated key.
Figure 8. Pigeon #3 wearing a vinyl bib that will partially cover a dot that, from the pigeon’s view, may only be seen when looking into the mirror while standing erect. This photo was taken from an habituation session.
Figure 9. Pigeon #3 wearing a new cotton bib. The new covering was created because the pigeons’ responses did not habituate well to the vinyl version. After 3 habituation sessions, all four birds appear to direct very little behavior at the bib.
Figure 10. Pigeon #3 in the mirror test – the bib and dot can be plainly seen.
Figure 11. Pigeon #3 looking into the mirror, after which he immediately begins bobbing and pecking at the dot he cannot see directly (see Figure 12, below).
Figure 12. Pigeon #3 soon after looking into the mirror!
Movie clip. Pigeon #3 at the start of the mirror test (plus bib, plus dot). Initially, you will see some “confused” behavior, followed by bobbing and pecking at the dot which he can only see in the mirror! And psychologists have been telling us for years that only humans and some primates can pass this test…hmmm…a successful replication of Epstein, Lanza and Skinner, 1981.
Cardinal, C. D., Allan, R. W., Barnes, D., & DeLabar, J. S. (1998). “To hold as’t were, the mirror up to nature”: Are pigeons really “self-aware?” Paper presented at the Association for Behavior Analysis Annual Conference, Orlando, Florida.
Epstein, R. (1996). Cognition, creativity, and behavior. Westport, CT: Praeger.
Epstein, R., Lanza, R. P., & Skinner, B. F. (1981). “Self-awareness” in the pigeon. Science, 212, 695-696.
Gallup, G. G. (1970). Chimpanzees: Self-recognition. Science, 167, 86-87.
Gallup, G. G. (1984). Will reinforcement subsume cognition? Contemporary Psychology, 29, 593-594.
Thompson, R. K. R., & Contie, C. (1986). Further reflections on mirror usage by pigeons. Paper presented at the annual meeting of Psychonomic Society, New Orleans.
Thompson, R. K. R., & Contie, C. (1994). Further reflections on mirror-usage by pigeons: Lessons learned from Winnie-the-Pooh and Pinnochio too. In S.T. Parker, R.W. Mitchell & M.L. Boccia (Eds.), Self-awareness in animals and humans: Developmental perspectives. Cambridge University Press: Cambridge, England.
Presentations of the Research
Cardinal, C.D., Allan, R.W., Barnes, D., & DeLabar, J.S. (1998). The reflected “self”: Are pigeons really “self-aware?” The International Society for Behaviorology Annual Convention, Tampa, Florida.
Cardinal, C.D., Allan, R.W., Barnes, D., & DeLabar, J.S. (1998). “To hold as ‘t were, the mirror up to nature”: Are pigeons really “self-aware?” Association for Behavior Analysis Abstracts, Orlando, Florida.
Cardinal, C.D., Allan, R.W., & DeLabar, J.S. (1999). “Self-awareness in the pigeon: A replication and controls. Association for Behavior Analysis Abstracts, Chicago, Illinois.