Rats were given five shocks over a 5-min period and then observed for 20 min. Much more freezing was observed in animals that remained in the shock situation than in animals moved to another situation. Freezing, therefore, seems to be controlled primarily by external shock-related cues. Freezing appears to be also partly controlled by the inherent stimulus properties of the situation.It has been known for some time that rats tend to become immobile, that is, freeze, in situations where they receive electric shock. Freezing is usually unwanted behavior, behavior that must be gotten rid of if the rat is to perform properly on, say, an avoidance-learning task. But there is now a growing interest in freezing both for its own sake because it is such a prominent part of the frightened rat's behavior repertory, and because a better understanding of freezing may lead to a better understanding of defensive behavior in general. Thus, Blanchard and Blanchard (l969) proposed that an immobile posture, which they designated "crouching," could be used as an index of fear. Anisman and Waller (l971) explained different effects in active and passive avoidance in terms of immobility responses, such as freezing, acquired during prior shock exposure and they were able to manipulate such responses experimentally. Bolles and Riley (l973) found that freezing was very rapidly acquired as an avoidance response.Bolles and Riley (l973) reported that when rats were shocked every 15 min, they froze during 77% of the iritershock interval. Blanchard, Dielman, and Blanchard (1968) found some elevation in the level of crouching hours after the administration of a single shock. How are we to explain this great persistence of immobility following aversive stimulation? Is it due to reverbration of some sort in the autonomic nervous system? Is it due to arousal of the pituitary-adrenal system? The present study supports the Blanchards' conclusion that the persistence of freezing, and crouching, is not primarily due to any such endogenous factor, but is, rather, due to the continued presence of external cues that predict shock. Our evidence for the external control of immobility is, like theirs, the relatively low incidence of freezing in rats that are shocked in one situation and tested in another. METHOD SubjectsThe subjects were 40 naive Wistar rats, approximately 120 days of age.Supported by Research Grant GB·40314 from the National Science Foundation. Requests for reprints should be addressed to the first author, Department of Psychology. University of Washington, Seattle. Washington 98195. ApparatusTwo boxes were used for conditioning and testing. One. designated the "long" box, was 76 cm long, 20 cm wide, and 18 ern high. It was placed in a sound-attenuating chamber and lighted by a 15-W bulb. The floor of the box was constructed of I-ern stainless steel bars spaced 2.5 em center to center; the sides were also stainless steel. Background noise of approximately 76 dB was provided by a ventilation fan.The second box was designated the "square" b...

Ten-and lfi-day-old rat pups were trained with two procedures to approach an anesthetized mother, and then were punished for approaching. Both ages of subjects exhibited increased latencies to reapproach the mother, indicating passive inhibitions, but only the older pups retreated. All but one of the younger pups eventually reached the mother within 3 min after the punishment, while only half of the older pups did so. In a second experiment examining the development of locomotor avoidance reactions, 5· to 20-day-old rats were shocked without the mother present. Fifteen-and 20-day-old rats significantly decreased their activity patterns in reaction to shock and spent significantly less time in the shock area than either of the younger aged pups. These results suggest that flight reactions are components of a rat's defensive repertoire that appear very rapidly between 10 and 15 days of age.Although information on the effects of punishment in very young organisms is limited (Walters & Grusec, 1977), we do have a certain amount of information available with respect to the behaviors of preweanling rats (younger than 20 days) in step-through and step-down passive avoidance situations (Riccio, Rohrbaugh, & Hodges, 1968;Riccio & Schulenburg, 1969). In such situations, very young rats are slower to reach a high criterion of performance than weanling or older subjects. What has not been clear is whether the results imply that young animals are slower to learn aversive responses, or whether they do learn the contingencies of punishments, but nonetheless fail to avoid because of their innate reactions to punishing events (Campbell, Riccio, & Rohrbaugh, 1971). The question is further confounded in that the data on preweanling rats have been restricted to the examination of punishment on only untrained, highly probable responses (e.g., crossing to the dark side of an alley or stepping off a platform).The limitation to untrained response assessment has been due, in part, to a paucity of adequate techniques for studying aversively controlled, as well as appetitively controlled, behaviors in preweanling subjects. However, recent evidence (e.g., Amsel, Burdette, & Letz, 1976;Kenny & Blass, 1977) indicates that very young rats can be trained to approach an anesthetized mother for dry suckling as a reward. By punishing such a learned approach response, it would now be possible to examine further the characteristics of aversive stimulation during the very early Requests for reprints should be sent to the first author, Laboratory of Comparative and Physiological Psychology, The Ohio State University, 1314Kinnear Road, Columbus, Ohio 43212. Portions of this research were supported by PHS Grant 5 T32 MH14608-02 to The Ohio State University while Joe1le Mast and Carrie-Ellen Jacobs were NRS Fellows. stages of development. Therefore, a major concern in the present investigations was to study the effects of punishment on young pups' behaviors in approachavoidance conflict situations. Also, we were interested in examining the unlearned res...

Behavioral observations were made on the reaction to tailshock and footshock in 5- to 20-day-old hooded rats. For detection thresholds, age differences were found for footshock but not for tailshock. During intershock intervals, more generalized activity and freezing were elicited by footshock, whereas more responding directed to the shock source was elicited by tailshock. The unconditional responding to shock indicated that the older animals had a larger behavioral repertoire of defensive reactions and responded differentially to different shock intensities. The younger animals appeared to have a more limited and stereotyped repertoire of defensive reactions and to be more sensitive to the same nominal shock level. These normative data should prove useful in evaluating motivational and response repertoire problems when assessing learning processes developmentally.

That organisms cannot remember events occurring during infancy may be the result of common forgetting processes exacerbated by the organism's increase in size during development or a unique process such as insufficient neurological development at the time of the early experience. To establish the uniqueness of infantile forgetting, size change was made irrelevant by exposing infant rats to "off-baseline" Pavlovian fear conditioning and assessing the effect of an apparatus-free conditioned stimulus upon independently established bar pressing. In Experiment 1, bar pressing by rats exposed to Pavlovian contingencies when 20-22 days old was substantially suppressed by the conditioned stimulus both 1 and 42 days after conditioning. In Experiment 2, pups conditioned when 17-19 and 20-22 days old again showed excellent retention, whereas pups conditioned when 11-13 and 14-16 days old showed total forgetting 42 days later. In Experiment 3, pups conditioned when 14-16 days old remembered well after 5 days, less well 10 days later, and not at all after 20 days. These findings suggest that size change may contribute to the forgetting of events occurring late in development, but that neurological immaturity may underly the forgetting of earlier events.

Trace procedures. with a gap between the conditioned stimulus (CS) and the unconditioned stimulus (US). often produce weaker conditioning than procedures with contiguous CS and US. Such deficits were found in this experiment using rat subjects and conditioned suppression techniques to assess the strength of fear conditioning. But the deficit was greatly reduced either by filling the CS-US gap with a second "filler" stimulus or by adding a brief "safety signal" at the start of the intertrial interval.It was Pavlov (I 927) who first noted that conditioning was possible with what he termed a trace procedure, in which the conditioned stimulus (CS) is presented only briefly and is not contiguous with the unconditioned stimulus (US). Pavlov observed that trace procedures were not particularly effective, and subsequent experimenters confirmed this observation. For example, Davitz, Mason, Mowrer, and Viek (I957) and Kamin (1961) reported that CS-US gaps of just a few seconds seriously impaired fear conditioning. Kamin (I 954) and Mowrer and Lamoreaux (I 95 1) reported severe decrements in avoidance learning with short gaps between the avoidance "cs" and shock. Although conditioning decrements are not always found (e.g., Brahlek, 1968), they do seem to occur under a wide variety of conditions.The present studies examine some procedures for eliminating or reducing the trace-conditioning deficit. These particular procedures were suggested by Mowrer and Lamoreaux's (l95 1) hypothesis that the deficit results from the animal's failure to discriminate the interstimulus interval from the inter trial interval. If the animal confuses the short, dangerous "silent" part of the interstimulus interval with the long, safe silence of the intertrial interval, then it should tend not to respond during the former and to respond too much during the latter (a result that Mowrer and Lamoreaux found). It also follows that the conditioning deficit should be ameliorated by marking or designating in some way one silent period or the other so that they are not so likely to be confused. We examined two ways of doing this ; first, we tried to mark the short interstimulus interval, then we attempted to mark the long intertrial interval. Both approaches were successful. EXPERIMENT 1Our first approach to the problem was to fill the gap in the interstimulus interval with a filler, a second CS, to make the CS-US gap discriminably different from the intertrial interval. MethodSubjects. The animals were 36 naive female rats, of Wistar descent, with initial body weight of approximately 300 g.Apparatus.The animals were trained and tested in Foringer Skinner boxes. Fear conditioning was carried out in wooden boxes of approximately the same dimensions, with the same type of grid floor, and illumination and masking noise (exhaust fan) conditions. Shock was scrambled, of .5-sec duration and I-rnA intensity . The CS was a white noise of approximately 75 dB in both apparatuses, and the additional stimulus was a 3,300-Hz tone, also about 75 dB.Procedure. The a...