The present model accounts for the characteristic phenomena of imprinting in terms of familiar behavioral processes by postulating that certain aspects of imprinting stimuli are primary reinforcers that innately elicit filial behavior. In doing so, these aspects serve as unconditioned stimuli, enabling the development of familiarity with the other characteristics of a given imprinting stimulus through classical conditioning. Familiarity serves to prevent novelty-induced fear reactions which would otherwise compete with the filial response at later stages of ontogenetic development. We suggest that these behavioral processes are the common denominators for socialization in several types of animals.According to many current views, imprinting and the behavioral effects it entails cannot be interpreted using the theoretical constructs that are familiar to psychologists. As a consequence, the now extensive literature in this area is often considered to be irrelevant to the broader issue of behavioral control in a social context, and imprinting is treated as a somewhat specialized phenomenon that is characteristic of only a limited range of animals such as precocial birds.Harlow, Gluck, and Suomi (1972) gave expression to this attitude when they noted: "We do not believe that the imprinting phenomenon aids in the analysis of infantmother ties in monkeys and men [p. 709]." We suggest here that the attitude characterized by such statements is based on a restricted and, in many respects, incorrect view of imprinting. More specifically, it is argued that there are a number of investigations in this area that imply that imprinting can be interpreted in terms of familiar behavioral processes and that this in turn implies that 1 Preparation of this manuscript as well as the research on which it is based was supported by Grant MH 19715 from the National Institute of Mental Health. We wish to thank Leonard Eiserer who made many important contributions to the research and to the conceptions that are central to this model. We also thank Richard Gonzalez for his valuable suggestions.2 Requests for reprints should be sent to Howard

Suppression of newly hatched ducklings' distress vocalization by two distinctive stimuli was assessed before and after subjects were imprinted to one of them. Initially, both stimuli strongly suppressed distress calls. Later, only the stimulus involved in imprinting did so. The results suggested that the loss of control by the nonimprinted stimulus was age related and represented the development of fear to insufficiently familiar stimuli. In a second experiment, older ducklings were permitted lengthy exposure to a novel imprinting-type stimulus. It was found that under these circumstances distress vocalization eventually came under stimulus control. The overall pattern of results suggests that imprinting renders stimuli that innately evoke filial behavior familiar and in doing so prevents their later evoking noveltyinduced fear.Among the most salient characteristics of imprinting in ducklings is the range of objects that can serve as the focus of filial attachment. Fabricius (1951aFabricius ( , 1951bFabricius ( , 1955 and Fabricius and Boyd (1954), e.g., have successfully imprinted ducklings to humans, live ducks, stuffed animals, and a variety of other objects such as balloons, boxes, and model railway cars. Data such as these point to the conclusion that for ducklings, the shape, size, and color of the imprinting stimulus, within wide limits, makes little difference in its capacity to elicit filial responses.When, however, imprinting to one stimulus is completed, the subject's tendency to respond filially to other stimuli is often considerably reduced (Sluckin, 1965). Apparently, some process occurs during imprinting that restricts the control initially afforded by a wide range of stimuli to only the imprinting stimulus or the class of stimuli that includes it. Since this differentiating process occurs without prior or concurrent discrimination training, it has been described by Jaynes (1958) as "emergent discrimination."The research reported here was designed to examine the changes in filial reactions

When a 5-day-old duckling is enforcibly exposed to a novel moving object, it gradually forms a filial attachment to it. The present research documented 3 behavioral stages that transpire during such socialization: (a) The duckling initially attempts to escape, (b) the tendency to escape wanes and aggression toward the object emerges, and (c) aggressive behavior wanes and filial behavior becomes evident. Additional experiments illustrated how the presence and absence of a previously imprinted stimulus controlled aggression and the necessity of enforcing proximity between the subject and the novel object in order to obtain aggression. By doing so, these experiments provide support for the hypothesis that this aggression is fear mediated and that a low but nonzero level of fear maximizes its occurrence.

Three experiments examined the effects of aversive stimulation in the context of imprinting. In Experiment 1, newly hatched ducklings were electrically shocked in either the presence or absence of an imprinting stimulus, and both their tendency to follow the stimulus when it was presented alone and their preference for it over novel stimuli were subsequently assessed. The results indicated that the shock enhanced the subjects' tendency to follow the imprinting stimulus in the nonchoice situation regardless of whether the ducklings were shocked in its presence or absence. If, however, the electrical shocks were associated with the presence of the imprinting stimulus, they also reversed the ducklings' usual preference for the imprinting object over the novel stimuli. Comparable effects were obtained in the second experiment in which 5-day-old ducklings were electrically shocked in either the presence or absence of a novel, fear-eliciting imprinting stimulus. In Experiment 3, electrical shock was administered to newly hatched ducklings independently of their exposure to an imprinting stimulus, and it was found that, relative to older ducklings that received the same treatment, the newly hatched subjects subsequently displayed less emotionally in fear-eliciting situations. The results were interpreted as suggesting that there are at least three distinct effects of aversive stimulation in the context of imprinting-motivational, associative, and developmental.

No abstract