haut de page

mise à jour du
20 janvier 2002
Yawning as a stereotyped action pattern
and releasing stimulus
Robert R Provine
Department of Psychology University of Maryland Baltimore
Autres articles de R. Provine et R. Baenninger
[...]Discussion : Yawns as Stereotyped Action Patterns

Yawning is a behavior of the type called "fixed," "modal," or "stereoryped" by ethologists (BARLOW 1977; HOYLF 1984; LORENZ 1932; SCHLEIDT 1974). (For simplicicy, the term "stereotyped action pattern" is used here to refer to such behavior; doing so does not deny the legitimate concerris about behavioral nomenclature raised in the analyses cited above.) Consider the following properties of yawning that are characteristic of commonly described stereotyped motor patterns. Yawning is consistent in duration, it occurs periodically, and withinsubject stability in yawn duration and frequency is maintained over intervals of several weeks. Yawning is under a high degree of genetic control because it is performed by normal (GESELL 1928; PREYER 1923; TAYLOR-JONES 1927) and anencephalic human newborns (GAMPER, CATEL, both cited in HEUSNER 1946); these populations are presumed to have little opportunity to learn how to yawn. Yawns are unitary, being performed at so-called "typical intensity" fractional yawns detectable as atypically short yawns in the analyses of yawn duration were seldom if ever produced. Yawns are "released" by witnessing yawns or yawn-related stimuli. (The internal physiological stimulus for "Spontaneous" yawning is unknown.) The amplitude and duration of yawns seem to be independent of the amplitude of the releasing stimulus; yawns seem to be "all-ornone" actions. Once initiated, yawns go to completion with minimal influence of sensory feedback; everyone is familiar with the difficulty of trying to stifle a yawn. Yawns are complex in spatio-temporal organization and have facial, respiratory, and other components; yawns are not simple reflexes of short duration. Although not established empirically, the components of a yawn seem to occur in only one order and the timing of components is consistent from yawn to yawn. This stability of sequence contributes to the yawn's unmistakable appearance. The finding that yawns are prominent behavior of people who are "waiting" (KATAOKA 1975) or performing monotonous work (KISHIDA 1973), of fish that are changing from one activity to another (MYRBERG 1972), and of dogs on the threshold of aggression or being forced to participate in an aversive activity (P. BORCHELT, pers. comm.) is consistent with the performance of yawns as "displacement acts" and "vacuum activity" Stereotyped action patterns often occur as displacement and vacuum activity. Given the above properties and the frequency and prominence of yawning in everyday life, it is curlous that yawning has been almost totally ignored by human ethologists (EiBL-EIBESFELDT 1975; MORRIS 1977). Indeed, yawning may be the best example of a stereotyped action pattern and releaser in humans. Specific properties of the yawn as a motor act and stimulus are explored below.

Once a yawn is initiated, it seems to go to completion with the inevitability of a sneeze. The procedure of yawning with clenched teeth was instituted to test for effects of eliminating or modifying movement-produced feedback associated with the gaping component of the yawn. Normal respiration was permitted through the clenched teeth. The frequency and duration of normal and clenched teeth yawns were similar, if not identical. Thus, the respiratory component of yawning, that which the subjects recorded, could be performed in the grossly abnormal sensory environment of the clenched-teeth condition. Normal sensory feedback from the gaping component of the yawn was not necessary for the initiation, maintenance, or completion of a yawn; the motor process responsible for yawning operates with relative independence from sensory input. However, subjective reports by subjects that clenched-teeth yawning was unpleasant, did not satisfy the urge to yawn and gave the impression of being "stuck" in midyawn suggests that yawn-produced sensory feedback plays some role in normal yawning.

Yawn Function

Clenched-teeth yawning is useful in evaluating accounts of yawn function that stress a role for respiration or for "stretching" the face. The facial stretch explanation assumes that the contraction of the facial muscles during a yawn forces blood through cerebral blood vessels to the brain, presumably to increase alertness (BARBIZET 1958; HEUSNER 1946). A yawn also opens the eustachian tubes, balancing atmospheric pressure between the air-filled middle car and the environment (LASKIEWICZ 1953). The respiratory explanations emphasize the increased blood 02 levels that may occur during the deeper-than-usual inspirations of the yawn (BARNETT et al. 1971). The clenching of teeth during yawns reduced or prevented stretching movements of the law that might normally have increased cerebral blood flow or opened the eustachian tubes; only isometric contractions of the jaw and facial muscles were possible. Although the clenching of teeth did not block yawns, the lack of satisfaction and unpleasant sensation produced by clenched-teeth yawning suggest that some function served by normal yawning went unfulfilled in this experimental condition. The respiration that was possible during the clenched-teeth condition, by itself, was insufficient to satisfy the urge to yawn in most subjects. The inadequacy of the respiratory hypothesis is also suggested by the absence of a significant correlation between spontaneous yawn duration and inter-yawn interval (i.e., infrequent yawners did not compensate by producing longer than normal yawns). By elimination, the jaw stretching movements of the facial muscles must play an important, if not principal, physiological role in normal yawning. A role for stretching is also indicated by the involuntary stretching movements of the paralyzed limbs of hemiplegics during yawns (WALSHE 1923).

Yawns as Releasing Stimuli

The present study confirms the common knowledge that yawns are "infectious" and provides details about the nature of the effective stimulus and the probability and latency of the yawn response. Visually observed yawns were potent yawn-producing stimufi. Most of the subjects viewing a series of videotaped yawns yawned within 5 mn of viewing the first yawn. Non-visual stimuli were also effective in evoking yawns. Reading about yawning produced actual yawns in over 1/4 of subjects, and approximately 3/4 of the subjects either yawned or thought about yawning within 5 min of starting to read a passage about yawning. Readers of this paper have probably yawned. Simply instructing subjects to "think about yawning" was effective in evoking yawns in the majority of cases; this was the technique used to evoke yawns in the descriptive studies of the yawning act. Thus, direct visual stimulation was not a necessary characteristic of an effective yawn stimulus. This conclusion is supported further by the interesting although largely anecdotal report by MOORE (1942) that auditory recordings of yawns evoked yawning in blind subjects. The wide variety of yawnproducing stimuli contributes to the "Infectiousness" of yawns.

Yawns are releasers of the stereotypic action pattern that is the yawn. Releasers studied by ethologists in non-humans range from the simplicity of a red spot on the mandible of the female herring gull to complex dances performed by courting male avians (EIBL-EIBESFELD 1975). Although we yawn in response to observed yawns, yawn-induced yawning is not the result of a conscious desire to imitate the yawner. An observed yawn initiates a complex series of neurobehavioral events that culminates in a yawn in the observer.

Yawn-induced yawning may provide insights into the recently reported ability of human neonates to "imitate" facial expressions (Field et al. 1982; MELTZOFF & MOORE 1977). Since the newborns have not had an opportunity to learn how to imitate facial expressions, a releasing mechanism such as here proposed for yawning may be involved. Indeed, yawn-induced yawning may be an example of such "imitative" behavior that has survived into adulthood. Aithough yawning is performed by human newborns, the presence of yawninduced yawning has not been tested. Its detection would add to the repertoire of newborn "imitative" responses and provide further evidence of the innate nature of the yawning act.


The present rescarch establishes the yawn as a stereoryped action pattern and releasing stimulus, both of which have been rarely identified in human behavior. Yawn-evoked yawning provides an excellent and perhaps unique opportunity to observe in ourselves the sensation of having a.stereotyped action pattern released. Much additional work is needed to define the possible physiological, social and developmental roles of yawning in health and disease. At present, yawning may have the dubious distinction of being the least understood, common human behavior. Yawning may prove to be elther an unimportant or vestigial act, or a maneuver with profound physiological consequences that may provide important clues to the mechanisms and evolution of human behavior. The prescrit, incomplete, evidende favors the later alternatives. The prominence of spontaneous and yawn-induced yawning suggests that yawning has, or once had, a major function; yawning dots not deserve its current status as a minor behavioral curiosity.