The process through which early behavior pattern becomes restricted to a particular type of object as a result of a relatively brief exposure of that object at a particular early life stage
After some animals are born or hatched, they have the instinct to acquire or learn the behavioral characteristics from the object exposed to them. This kind of phenomenon can be easily observed in avian such as domestic chickens and ducklings, and it was described as “imprinting” by Konrad Lorenz in 1935 based on his observations of goslings. The word “imprinting” indicates that the learning process of the characteristics of certain objects in young animals is like an inborn and fixed mechanism. Some advanced behavioral and psychological development, such as social bonding and language skills, are believed to have a close relationship with imprinting process.
Imprinting as an Ethology Topic
The concept of imprinting was developed from the observation with animals. Young birds such as ducklings imprint on the first moving object they observe in life; in most cases the object is the duck’s mother. After imprinting, the baby duck walks just after the imprinting object. This remarkable learning process was first documented by Douglas Spalding, an English biologist, and the process was described as “stamping in” (Spalding 1873). The similar phenomenon was also described by a German biologist Oskar Heinroth in 1911 (mentioned in Bateson 2000), one of the founders of ethology, who used the German term Prägung (means “imprinting”) to describe this phenomenon. However, it was Konrad Lorenz (1935), a student of Heinroth, who made the phenomenon of imprinting well-known through a considerable amount of research work. The most famous experiment was that Lorenz showed his own leg to goslings during a particular period after hatching, and the goslings would follow Lorenz rather than their own mother. Lorenz also found that goslings would imprint on inanimate objects (e.g., a box placed on a running model train) if exposed to it during a particular period. This simple but special act of following was thought to be a determination of social bond between the newborn and their parents (Lorenz 1937).
It was further discovered by Lorenz that the age of the birds influenced imprinting (Lorenz 1937). Imprinting would develop naturally and steadily if the object was shown during the first two days after birth. Otherwise, when the object was shown before or after that period, the phenomenon of imprinting would hardly occur. This indicates imprinting only happens during specific periods of development, and the specific period is called critical period or sensitive period. Learning a particular behavior would be easiest during the corresponding critical period.
The finding of imprinting is regarded as one of the most famous studies of ethology (e.g., Krebs and Sjölander 1992). Unlike behaviorists who estimate the responses of animals to specific stimuli which are usually manipulated in laboratory context, ethologists prefer putting much emphasis on animal’s evolutionary adaptivity. They explore the common behavior patterns under natural conditions among different species and believe that existing animal behavioral patterns are generally resulted from evolution. Imprinting is obviously thought of as an adaptive behavior. The function of imprinting is to enable the young animal to recognize and follow the parent instinctively shortly after birth, which will increase the offspring’s survival probability greatly. Through imprinting the newborn offspring could recognize its parent for protection from the dangerous situation, such as being hunted by enemies or being isolated by other members of the same species. For the parent, imprinting could help them identify their offspring so they would not waste time and energy caring for those who are not their offspring. Thus, the probability of species’ successful genetic transmission could be improved greatly through imprinting.
Furthermore, the imprinting developed at the early age also shows long-lasting influence on the offspring’s mating behaviors, which is called sexual imprinting. For instance, male zebra finches preferred to mate with those who looked like the female bird that reared them, rather than the birth parent, in the case when they are born and reared separately by different parents (Immelmann 1972). Though sexual preferences may also be influenced by other predictors throughout later developmental experience, sexual imprinting still has a long-lasting and unavoidable impact on sexual choices across species.
Imprinting and Learning
Some researchers pointed out that imprinting was a special type of learning in nature. In order to emphasize that imprinting is distinct from learning, Lorenz (1970) proposed the process of imprinting with four characteristics: (1) it happened during a restricted time interval which was called sensitive period or critical period; (2) the process of imprinting was irreversible; (3) this type of learning may occur long before it could be behaviorally demonstrated; and (4) it consisted of supra-individual and species-specific characteristics. According to Lorenz the two most important characteristics of imprinting are irreversibility (the property that after imprinted by a specific object, the organism would always prefer the object to other stimuli) and the existence of critical period.
Although the claim by Lorenz of the irreversibility and critical period of imprinting has been refuted with experimental evidence in birds (e.g., Hoffman et al. 1972; Salzen and Meyer 1968), the characteristics of imprinting concluded by Lorenz still have important significance for differentiating imprinting from other types of learning such as associative learning and perceptual learning.
From the perspective of perceptual learning theory, imprinting is the consequence of being exposed to the particular stimulus at early time. In studies of perceptual learning, animals present behavior pattern similar to imprinting. For example, rats performed better in discrimination tasks if exposed to the similar stimulus from birth (Gibson and Walk 1956), which indicated that similar with imprinting, perceptual learning was also related to the familiarity with objects.
Yet, perceptual learning is more like the formation of internal representation of a specific stimulus, rather than a more complex process in which organisms form a behavioral (van Kampen 1996).
Another stream of research has been focused on the similarities and differences between imprinting and associative learning. Associative learning includes classical conditioning and operant conditioning. Some researchers pointed out that some features of an imprinting stimulus acted as unconditioned stimulus (US), which were motivationally significant, and other neutral features acted as conditioned stimulus (CS). This approach hypothesized that birds got attached to shape and color (CSs) of specific object because they detect the relationship between shape or color information with the movement or sound (USs) of the object (e.g., Hoffman and Ratner 1973).
However, with the emergence of more evidence, the claim that imprinting is similar to associative learning was questioned (Bateson 1966; Bolhuis 1991). For example, even if birds’ behavior can be reinforced by presentation of the stimulus on which they are imprinted, evidence also showed that birds, which were not able to imprint, could still learn operant tasks (Bateson and Reese 1969). Besides, both classical conditioning and operant learning refer to the formation of associations between neutral and unconditioned stimulus, which can be demonstrated simultaneously or separately, but imprinting seems more like a process through which an animal can detect an object naturally and form a representation of it, rather than predict a specific event. Thus, imprinting helps the young offspring learn to recognize the characteristics of the parent, while associative learning helps a subject to predict the environment.
In addition, the mechanism of imprinting differs from that of associative learning. In associative learning, predicting the environment involves detecting the order of separate events. But in imprinting, learning to discriminate one object (or characteristic) from another may be the most important task, and the order in which the features occur is not that important (Bateson 1990). In one of Bateson’s studies, birds performed better in discrimination tasks if the two objects were demonstrated 5 min or more separately than those in the control group which had not been exposed to the objects. Interestingly, when the two objects were demonstrated 30 s or less separately, the imprinted birds performed poorer than the control group in discriminating the objects. This result can be explained that if presented separately within a short time, different stimuli can be perceived as the properties of the same event, no matter the order of stimuli. In general, there are some similarities between imprinting and associative learning, yet differences also exist in the functions and the mechanisms of them.
Imprinting and the Development of Attachment
Imprinting also provides an alternative explanation of how the attachment pattern is developed.
Based on the characteristics of imprinting, imprinting might be a way to understand the formation of the deep and long-lasting ties between the young and their caregivers. It was concluded by Lorenz (1935) that imprinting served as a specific way for offspring to learn their species’ identity, and it lead offspring to form two behavioral patterns: filial bonding and sexual preferences. Attachment, as a motivational and behavioral system (Bowlby 1982), directs the young child to seek proximity with a familiar caregiver and thus is thought to be highly related to the process of imprinting. For example, both of imprinting and attachment take place on the basis of the relationship between the offspring and the primary caregiver, and both of them focus on the offspring’s early experience. Although Bowlby did not include imprinting theory in attachment theory, he also considered that attachment behavior would be best explained as instinctive. These similarities between imprinting and attachment led much literature (e.g., Reed and Leiderman 1983) to discuss whether the development of attachment is a result of imprinting.
Although observational studies of young children in natural settings provided behavioral evidence that might indicate attachment was similar to imprinting, there remained distinct differences between them. For instance, Bowlby’s attachment theory included the idea that attachment involved learning from experience during a limited age period, especially influenced by the early social interaction between infants and the primary caregiver after birth, which is different from imprinting (Bowlby 1982). Another example was when staying within a certain distance of the mother without effort on her part and picking up small objects, young children would bring them to the mother but not to others (Anderson 1972). Such kind of behavior may be evidence of attachment, but it may lack the evidence of the essential sensitive period if explained by imprinting. In addition, by studying behaviors of young children, attachment can be described as a developmental process, related to an infant’s entire caregiving history, yet imprinting tends to be a learning process that happens in a comparatively shorter period (Reed and Leiderman 1983).
Overall, the concept of imprinting has been developed from the observations of young animals by ethologists. This phenomenon is regarded as an instinct model of early learning which happens within the critical period and is irreversible. Though similar with perceptual learning and associate learning, they are actually different internal mechanisms. Both of imprinting and attachment behaviors emphasize the young organism’s instinctive behavioral pattern; however, they are distinct with each other in nature.
- Anderson, J. W. (1972). Attachment behaviour out of doors. In N. Blurton-Jones (Ed.), Ethological studies of child behavior (pp. 199–216). Cambridge: Cambridge University Press.Google Scholar
- Bateson, P. (2000). What must be known in order to understand imprinting? In C. Heyes & L. Huber (Eds.), Vienna series in theoretical biology. The evolution of cognition (pp. 85–102). Cambridge, MA: The MIT Press.Google Scholar
- Lorenz, K. (1970). Studies in animal and human behavior. Cambridge, MA: Harvard University Press.Google Scholar
- Spalding, D. A. (1873). Instinct with original observations on young animals. Macmillan’s Magazine, 27, 282–293.Google Scholar