Perceptual consequences of early social experience in precocial birds

185

Behavioural Processes, 30 (1993) 185-200 0 1993 Elsevier Science Publishers B.V. All rights reserved 03766357/93/$06.00 BEPROC

00500

Review

Perceptual

consequences

experience Robert Lickliter, Department

of early social

in precocial

Antoinette

birds

B. Dyer and Thomas McBride

of Psychology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, USA (Accepted 9 June 1993)

Abstract This

article

reviews

recent work

with

precocial avian hatchlings

demonstrating

the

important contribution that social experience with conspecifics can make to the development of species-typical perceptual preferences. In particular, experiments on the role of sibling

social

interaction

in the development

of early

auditory

and visually

directed

maternal preferences are surveyed. Results reveal that young hatchlings denied the opportunity for direct social experience with siblings consistently display auditory and visual preferences different from those shown by hatchlings allowed ongoing experience with their broodmates during the period immediately following hatching. Taken together, the studies reviewed here demonstrate that the perceptual preferences underlying the process of filial imprinting, long thought to be simply ‘innate’ or ‘instinctive’, are sensitive to an array of social factors present in the young bird’s posthatching environment. The findings also provide support for the view that the minimum unit for the developmental analysis of species-typical behavior must be the developmental system, developing organism and its specific stimulative environment.

Key words: Developmental

systems;

comprised

of both the

Early perceptual preferences; Filial imprinting;

Socio-

genesis

Introduction The study of the relationship

between an organism and its environment

been relegated to the biological discipline

Correspondence University,

to: R. Lickliter,

Department

Blacksburg, Virginia 24061-0436,

of Psychology, USA.

has traditionally

of ecology. Over the last several decades a

Virginia Polytechnic

lnsitute

and State

186

growing

number

of biologists

and psychologists

have argued for a more

‘ecological’

approach to the study of behavior, an approach that explicitly focuses on the organism-environment relationship and, in particular, on the specific features of normally occurring stimulation available in the organism’s natural habitat during development (Brunswik, 1952, 1985,

1956; 1989;

Gibson, 1966, 1979; Cottlieb, 1971, 1976; Johnston, 1982, 1985; Oyama, Valsiner, 1987; West and King, 1987). For example, Khayutin (1985) has

argued that “the study of behavior must include both the analysis of the environment and of processes inside the organism, as well as an analysis of interaction between the organism and its environment”. While few contemporary

biologists or psychologists would argue against this view, an

empirical concern for the structure and nature of an organism’s particular environment continues to remain overlooked or undercharacterized in many contemporary conceptions of species-typical behavioral development. Many investigators continue to view the environment as largely unpredictable and potentially infinite in its range of stimulus conditions, rather than organized and specific as far as young organisms of every species are concerned. The former perspective has resulted in a general lack of appreciation of species-typical environments as providers of basic, necessary elements for development and as a positive, informative and constructive force in the achievement of phenotypic outcomes. In particular, many investigators continue to view a young organism’s effective environment

as essentially

‘supportive’ or ‘disruptive ’ in nature and therefore

less than

equal partners with genetic factors involved in phenotypic development (Gould, 1982; Smith-Gill, 1983). Indeed, as West et al. (1988) and Harper (1989) have observed, students of behavioral development are really just beginning to analyze what constitutes ‘the environment’ for any particular animal species. While there are a number of reasons for this conceptual and empirical shortcoming (see Lickliter and Berry, 19901, we believe it is due, at least in part, to the fact that the term ‘environment’ is used in several different ways in both psychology and biology. On the one hand, it is commonly used to refer to the potentially variable, largely unpredictable features of the physical world. This usage refers to the biosphere, the environment outside the organism in the broadest sense. In this use the ‘environment’ is typically viewed as something the organism must be protected from or buffered against, because of its potentially perturbing or disruptive nature (see Smith-Gill, 1983). West and King (1987) have referred to this use of the term environment as ‘environment-as-ecotope’. On the other hand, the term environment

can also be used to refer to the reliable,

dependable, and often unique developmental context of a given species. West and King (1987) have coined the term ‘environment-as-habitat’ to refer to this use of the term environment, referring to those aspects of the physical, biological, and social surroundings with which the active, developing organism actually interacts. This usage is well illustrated in the following quote from John King (1968): “Individuals of a species are usually raised by parents of the same species in the environment which has been occupied by that species for many generations. This continuity of early experience from one generation to the next envelopes the young of each species in an environment as characteristic of the species as its genotype.” For example, an ubiquitous feature of life for nearly all birds and mammals is that the developing individual is embedded in and interacts with a social environment. Across an array of diverse species and ecologies, the typical circumstances of development for most avian and mammalian species reliably includes conspecifics, especially parents and siblings.

187

Like other aspects of the developmental

milieu,

these social companions

can be viewed

as

experiential resources to the developing individual and, as this review will attempt to demonstrate, can play a significant role in the achievement of species-typical behavior. In particular, parents, siblings, and other conspecifics appear to facilitate the often rapid perceptual and social adaptations often required during early development. However, the role of conspecifics (other than the mother) in the achievement of species-typical behavior has tended to be overlooked or undercharacterized in much of the animal behavior literature. In particular, the role of siblings and other conspecifics present in an organism’s early social environment has not typically been subjected to systematic, developmental analysis until

relatively

recently.

Indeed, only within the last several decades have researchers begun to specifically focus their empirical efforts on how the particular experiential requirements of phenotypic development are related to the specific social contexts typically encountered by developing individuals (e.g. Alberts and Cubernick, 1983; Baptista and Petrinovich, 1984; Blaich and Miller, 1986; Galef, 1985; Cottlieb, 1971, 1980, 1991 b; Hofer, 1978, 1987; Khayutin, 1988; Petrinovich, 1990; 1985; Kruijt et al., 1983; Mason, 1978; Mason and Capitanio, Wang and Novak, 1992; West et al., 1988). The results of these studies indicate that empirical attention to the experiential stimulation present during the course of both prenatal and postnatal development is required if researchers are to give an appropriate interpretation to the results of experimental manipulations concerned with that development. The investigation of the phenomena of filial imprinting illustrative

example

of this conceptual

and methodological

insight.

What

understanding serves as one follows

is in no

way intended to be a comprehensive review of the literature on filial imprinting; rather, our intention is to utilize the phenomena of imprinting as a vehicle to highlight the important role of conspecifics to the perceptual and social adaptations often required during

early development.

The study of imprinting As is widely recognized, the young of many precocial preference for the first moving, conspicuous object they initial

preference

appears

to

develop

without

the

bird species can form a social encounter after hatching. This

presence

of any

obvious

external

reinforcement, is achieved relatively rapidly, and often shows enduring stability. Over the last fifty years, research concerned with the nature of imprinting has generated a large and diverse

literature

and has directly

influenced

thinking

about

the nature

of the mechanisms

and processes underlying early social attachment in animals, including humans (see Bateson, 1966; Bolhuis, 1991; Bowlby, 1969; Hess, 1973; Sluckin, 1973 for reviews). The fact that the phenomemon of imprinting is used in both biology and psychology as an exemplar of the mechanisms and processes underlying social preferences and attachment seems surprising and somewhat paradoxical, however, since much of the imprinting research performed over the last 50 years has generally ignored tance of the developing bird’s typical or usual social environment maintenance of its early filial behavior.

or obscured the imporin the establishment and

Specifically, in most studies of imprinting avian hatchlings have routinely been reared in social and even visual isolation throughout the course of the experiment. This isolation rearing, while convenient to the experimenter, is in marked contrast to the typical context of the process of species identification as it occurs in the naturally occurring environment of

188

the young bird. For example, under species-typical circumstances, the precocial hatchling would receive an array of visual, auditory, and social stimulation from its surroundings, especially its mother and siblings, in the period following hatching. The young chick certainly would not be socially or visually isolated during the first days of postnatal life. Many investigators of the imprinting process have also employed highly artificial, species-atypical maternal surrogates in their experiments, often from the assumption that young birds should be equally sensitive to any salient visual object present in their posthatching environment. The range of items used as maternal surrogates has varied widely across studies, and has included such objects as flashing lights, milk bottles, balloons, sponges, boxes, and balls. Interestingly, the results of these types of imprinting studies have led some researchers to conclude that a propensity for visual imprinting is somehow ‘innate’. In other words, the learning abilities underlying the imprinting process are often viewed as somehow genetically preprogrammed into the organism and to automatically unfold during the course of maturation (see Lorenz, 1965; Eibl-Eibesfeldt, 1970; Gould, 1982; Suboski, 1990 for examples). However, when some investigators began to turn

their

attention

away from

such

artificial social conditions a different and more complicated story of early species identification began to emerge (see Gottlieb, 1965, 1971, 1973). For example, to determine whether the results obtained from studies using artificial imprinting objects could be generalized to the hatchlings’ typically encountered perceptual and social conditions, Johnston and Cottlieb (I 981a) examined the effects of using taxidermically prepared (‘stuffed’) adult hens as maternal surrogates. Incubator reared mallard ducklings (Arm platyrhynchos) were allowed a 20 min imprinting trial at 24 h following hatching, during which time they followed one of five different models. Two of the models employed were highly artificial social objects (a green ball and a red and white striped box). The other three models were stuffed adult female ducks of different species (Fig. I). On the following day, each subject was given a IO min test trial, during which time it was presented with a simultaneous choice between the familiar imprinting object (which it had followed at 24 h of age) and a different, unfamiliar object. Results revealed that when one of the test objects was of the highly artificial type typically used in most imprinting experiments (i.e. the ball or the box), the ducklings showed a strong preference for the familiar object (i.e. the object the birds had been trained with). In contrast, when the models used were the natural stuffed hens, most of the birds did not show a preference for the familiar model, even if the length of the training (exposure) trial was extended to 24 continuous hours. Based on these results, Johnston and Gottlieb (1981a) argued that many imprinting studies have greatly overestimated hatchlings’ abilities to visually discriminate among naturally occurring social objects. This overestimation was the result, at least in part, of the use of artificial objects to evaluate the animal’s discrimination abilities, and then incorrectly generalizing those results to the bird’s natural situation.

Early social experience

and visually

imprinted

preferences

Related research has also served to reveal that species-typical social experience can affect the perceptual and social preferences underlying early filial behavior. For example, Johnston and Gottlieb (1985) found that if young ducklings are allowed social experience with siblings following their initial exposure to a natural stuffed model of a maternal hen, their subsequent visually imprinted preferences were markedly different from those of

189

Fig. 1. Visual models used in the imprinting experiments. From left to right: red and white striped box, green ball, pintail (Anas acuta) hen, mallard (Anas platyrhynchos) hen, redhead (Aythya americana) hen. Scale shown by green ball (I5 cm in diameter). isolate-reared ducklings. Specifically, the experience of social rearing with siblings following imprinting to a mallard hen model was found to enable ducklings to exhibit a preference for the familiar mallard (Anas p/atyrhynchos) hen over an unfamiliar redhead duck (Aythya

americana)

hen

model

at 48

h after hatching, a visual

discrimination

that

isolate-reared ducklings are unable to make (Johnston and Cottlieb, 1985). This finding that aspects of the natural social context of ducklings can serve to enhance the development of visual imprinting

led to a series of related studies utilizing

domestic

mallard ducklings as subjects (Lickliter and Gottlieb, 1985, 1986, 1987, 1988). The general methods employed were virtually the same in these studies. In brief, incubator hatched ducklings were used as subjects. Each mallard duckling in each experiment was given a single 30 min imprinting trial at 24 h after hatching with a stuffed mallard hen emitting a recording of the mallard maternal assembly call. The assembly call is the call that duck hens use in calling their young from the nest and continue to use during the posthatching period to keep the young together (Gottlieb, 1971; Miller and Cottlieb, 1978). Subsequently, two test trials, each 10 min in length, were given to each subject, one at 48 h and one at 72 h after hatching. During testing, the familiar stuffed mallard hen and an unfamiliar stuffed redhead hen were moved silently around the test arena and the duckling’s visual preference was assessed by recording the latency and duration of following each hen model. An individual preference was assigned to any subject that followed one hen for more than twice as long as it followed the other hen in a test. The

first

study

in the series (Lickliter

and Gottlieb,

1985)

demonstrated

that when

ducklings are: (1) reared in isolation but allowed to see and hear (but not physically interact with) one sibling; (2) reared directly with one sibling; or (31 reared in a group of siblings but denied direct social interaction, no visual preference for the familiar mallard hen is seen. In contrast, ducklings allowed unrestricted social interaction with 8-l 2 siblings displayed a visual preference for the familiar mallard hen at both 48 h and 72 h after hatching (Fig. 2). In other words, if young ducklings are exposed to a stuffed model of an adult hen and are subsequently asked to choose between that familiar hen and an unfamiliar hen of another species, they appear unable to make this visual discrimination when reared in varying degrees of social isolation. On the other hand, if the young birds are reared in groups of 8-12 broodmates following hatching, as would normally occur in

190

Reared isolation

in

Reared Sibling

with

Reared Group

Social

A

-

One

e

in Social

-

A

TrHa$ed

l

Tested: Mallard Redhead

* A

with

Significant for Mallard

i*

Mallard

vs.

Preference

I

4’8

;4

0

A

j2

Hours Posthatching Fig. 2. Summary

nature,

familiar

and

of the design and outcomes

tested

under

mallard

similar

hen during

of experiments

conditions,

they

exhibit

the days immediately

on effects of isolation

a reliable

visual

following

and social rearing.

preference

for the

hatching Oohnston

and

Cottlieb, 1985; Lickliter and Cottlieb, 1985). Lickliter and Cottlieb (1987) also showed that when the social experience with siblings occurs can influence its effectiveness in contributing to visually controlled maternal imprinting. Social interaction with siblings must occur after the initial maternal imprinting experience to be effective; social rearing with siblings before the maternal imprinting trial was found to be ineffective. Ducklings that were reared with siblings before the maternal imprinting trial (at 24 h after hatching) but not after the trial did not show a visual preference for the familiar mallard hen in subsequent testing. In contrast, birds reared with siblings before and after the imprinting trial or only after the imprinting trial were able to visually discriminate and prefer the familiar mallard hen (Fig. 3). A subsequent study examined the species-specificity of social experience required to foster species-specific visual imprinting in young hatchlings (Lickliter and Gottlieb, 1988). In this study, no visual preference for the familiar mallard hen was found if mallard ducklings were allowed social experience from hatching with a group of same-aged, nonconspecific domestic chicks (Callus domesticus) or muscovy ducklings (Cairina moschata). In contrast, mallard ducklings reared with other domesticated or wild-type mallard ducklings did exhibit a visual preference for the familiar mallard hen at both 48 h and 72 h following hatching. Thus, for young mallard ducklings to show a visually imprinted preference for a natural maternal model, they must be allowed social interaction with broodmates of their own species. Social experience with other precocial birds, even ducklings of another species, was ineffective (Lickliter and Gottlieb, 1988). Taken together, these results underscore the notion that surprisingly detailed information about the experiential stimulation present during the normal course of development is

Reared Before

Socially Training

Reared Socially After Tralnlng Reared Before Training

A

Trained Hen

0

Tested: Mallard Redhead *

Socially and After

-

0

with

Slgnlficant for Mallard

Mallard

vs.

Preference

A

;4

w

4’8

f2

Hours Posthatching Fig. 3. Summary

of the design and outcomes

of experiments

on the timing

of social experience.

191

required if investigators are to successfully design experiments

to understand that develop-

ment. Of course, a sensitivity to the experiential features present in the social environment is but one step in an attempt to capture the complex structure of the young organism’s experiential milieu.

The phenomenon

of peer imprinting

Despite the very large literature on imprinting, remarkably little information is available on the effects of social stimulation available in the precocial bird’s brood situation following hatching. A few studies have recently focused their efforts in this direction. For example, Johnston and Cottlieb (1985) found that ducklings that had been imprinted to a mallard hen model at 24 h after hatching and then allowed social interaction with a group of siblings after training preferred the familiar mallard model over another unfamiliar pintail duck (Anas acuta) hen model at a 48 h choice test, but not at a subsequent 72 h retest. These findings imply that after the initial formation of a visually imprinted maternal preference, subsequent maintenance of the preference may require further exposure to the mother.

In other words, under the social conditions

in which the development of social

preferences normally occurs (i.e. at the nest with mother and siblings), social experience with broodmates may somehow interfere with the maintenance of the maternal social bond if the young hatchiings are not otherwise exposed to the mother hen. If this supposition is correct, then the relative permanence of initial social preferences reported in the imprinting literature might only be an artifact produced by the conventional use of isolation rearing between training (imprinting) and later testing. Several related studies have examined this question. In general, these studies closely followed the general methods of those described above. Socially reared mallard ducklings were individually given a single 30 min training trial with a stuffed mallard hen at 24 h following hatching, after which they were returned to their social group of siblings. Two test trials, each 10 min in length, were given to each subject, one at 48 h and one at 72 h after hatching. During these simultaneous choice tests, subjects were asked to choose between the silent familiar mallard hen and 4 stuffed natural models of young mallard ducklings. Utilizing these methods, Lickliter and Gottlieb (1986a) found that social rearing with siblings results in a visual preference for the group of siblings over the familiar maternal hen. Not surprisingly,

a preference for the familiar mallard hen’over the group of unfamiliar

stuffed ducklings was found only when subjects were reared in social isolation for the duration of the experiment (i.e. from hatch through 72 h). Imprinted ducklings that were reared in a social group containing same-age siblings or the stuffed mallard hen and same-age siblings displayed a visual preference for the stuffed ducklings over the familiar mallard hen at both 48 and 72 h tests (Fig. 4). Further, rearing ducklings in a social condition that allowed them to see other ducklings but denied them the opportunity for direct contact and reciprocal social interaction was also found to be sufficient to interfere with the exhibition of an imprinted preference for the familiar mallard (but was not sufficient to redirect their social preference to the sibling models). That is, birds allowed social interaction with siblings preferred the duckling models over the familiar mallard hen model, whether or not they were provided with continual exposure to the mallard hen during rearing. In contrast, ‘imprinted’ birds denied the opportunity for direct social interaction with siblings did not show a visual preference for either the hen or duckling models (Lickliter and Cottlieb, 1986a).

192

Hours Posthatching Fig. 4. Summary

of the design and outcomes

of experiments

on the effects of peer imprinting.

These results indicate that under the social conditions in which maternal imprinting normally occurs, social interaction with siblings can interfere with the maintenance of the maternal social bond, suggesting that continued interaction with the mother is necessary to maintain it. Of course, under normal conditions such social contact with the mother is routinely available, since the duckling remains with its hen throughout the posthatching period (Bjarvall, 1967). Dyer, Lickliter, and Gottlieb (1989) went on to investigate whether active following of the maternal hen during the posthatching period is essential to the induction of a visually controlled, species-typical maternal preference. In this study, mallard ducklings were not given the opportunity to actively follow a mallard hen model during an imprinting trial. Rather, subjects were passively exposed to a vocalizing mallard hen model with live siblings present. Subjects were housed with the hen model

and broodmates

for either 24, 48, or 72

h following hatching. They were then tested for their visual preference for the familiar mallard hen or an unfamiliar redhead duck hen (Fig. 5). Subjects were found to be largely unresponsive to either hen in the choice test, even when they were passively exposed to the mallard hen for as long as 72 h prior to testing. These results clearly suggest that active following of the hen is a necessary experiential component of the development of a visual maternal preference. However, since subjects were reared with both the hen replica and a group of broodmates prior to testing, it could also be that the lack of a passively induced maternal preference was due to the influence of siblings during the subjects’ rearing experience. That is, the ducklings may have become imprinted to each other during the social rearing

Passive Mallard Siblings

Exposure Hen and

to

Passive Mallard Siblings

Exposure Hen and

to

Passive Siblings

Exposure Only

to

Passive Exposure Mallard Only

to

0

0

2:4

4’8

Tested: Mallard Redhead

“s.

Tested: Mallard Siblings

vs.

*

Significant Preference for Mallard

v

Significant Preference for Siblings

7.2

Hours Posthatching Fig. 5. Summary

of the design and outcomes

of experiments

on the effects of passive exposure.

193

experience and thus were unresponsive were reared individually

to the hen. To address this question,

hatchlings

with the stuffed vocalizing mallard hen; results of testing revealed

that these individually reared ducklings were also unresponsive to the silent mallard hen in the choice test (Fig. 5). This finding again indicates that some type of postnatal experience other than passive exposure to the hen is required to foster maternal imprinting. The question next arose as to whether passive exposure to siblings could induce a preference for them (i.e. peer imprinting). Accordingly, Dyer et al. (1989) reared duckings together in groups of eight for 48 h following hatching. One group was reared with only broodmates present, and another group was reared with both the mallard hen and broodmates present. The subjects from each group were then individually tested for their visual preference for the mallard hen model or four stuffed ducklings. Results revealed that ducklings from both groups preferred the visual characteristics of siblings over the mallard hen, whether or not the hen was present during rearing (Fig. 5). One plausible explanation for this rather surprising finding is that the noninteractive nature of the stuffed mallard hen was qualitatively different from that of live, interactive siblings. However, a subsequent experiment demonstrated that passively induced ‘peer imprinting’ occurred even when ducklings were reared individually with seven stuffed ducklings

(Dyer

et al., 1989).

That is, ducklings

never exposed to live siblings

prior to

testing also showed a preference for the stuffed siblings over the familiar mallard hen in simultaneous choice tests. These findings indicate that even minimal experience with siblings is sufficient for the passive induction of a visual preference for broodmates (i.e. peer imprinting), whereas prolonged passive exposure maternal hen does not result in maternal imprinting.

to the visual characteristics of a

Of course, under natural conditions, if ducklings are to learn the specific visual features of their hen during the posthatching period, they must be capable of learning these features in the presence of other ducklings. Lickliter and Cottlieb (198613) thus focused on the influence ducklings

of siblings during a maternal imprinting trial. In this study, isolate-reared were allowed to follow a vocalizing mallard hen replica for 30 min at 24 h

following

hatching. However,

rather than following

the hen individually

(as in the previ-

ously described studies) hatchlings were exposed to the mallard hen in groups of four. Because each duckling was reared in isolation from hatching to testing, each subject received only 30 min of simultaneous exposure to both the hen model and three same-age siblings (during the training trial). Subjects were subsequently individually tested for their preference for the familiar mallard hen over an unfamiliar pintail duck (Anas acuta) hen. It is important to note that isolate-reared and socially reared ducklings which have been trained individually with the mallard hen reliably demonstrate a preference for the familiar mallard hen in this test situation (Lickliter and Cottlieb, 1986b). In contrast, the subjects trained with the mallard hen in groups of four failed to show a preference for the familiar mallard hen over the unfamiliar pintail hen when tested 48 and 72 h following hatching (Fig. 6). Furthermore, when presented with a choice between the familiar mallard hen and a group of stuffed ducklings, group trained ducklings exhibited a significant preference for the stuffed siblings over the mallard hen (Fig. 6). Taken together, these results indicate that a mere 30 min of social experience with siblings (concurrent with exposure to a hen) during a training trial is sufficient not only to prevent the establishment of an imprinted preference to a maternal hen model but also to induce an imprinted preference for siblings. In sum, it appears that under the conditions in which the formation of the maternal bond normally occurs, social contact with siblings (and the resulting peer imprinting) may

194

Reared in Social Isolation

A w

Reared Group

-

in Social

A

v

in Sibling

v

‘,

.

T;;&;t;

Mallard

,

T~;;ghllard

-

?) *

Significant Preference for Mallard

D

Significant Preference for Siblings

d8

Hours Posthatching 6. Summary of the design and outcomes of experiments

serve to impede the development

vs.

vs.

m

v

0 Fig.

Alone

;;J;;d

&7

A

Reared in Social Isolation

Trained

+

i

A

Reared in Social Isolation

A T

A w

on the effects of group training.

of a maternal preference. Consequently,

active and

ongoing maternal involvement may be necessary to both establish and maintain the social attraction to the hen on the part of precocial avian hatchlings. This presumptive requirement of repeated and relatively continuous exposure to the mother hen stands in contrast to most conventional interpretations of imprinting, which have posited that a brief exposure to the mother is sufficient to form a stong and enduring maternal preference. While

it is clear that socially isolated hatchlings exposed briefly to a maternal surrogate

shortly after hatching can learn to prefer that familiar object over other novel objects in choice tests, the relevance of such a process to the usual course of species and individual recognition in nature seems suspect in light of the studies reviewed above. indeed, in all situations that we have studied, ducklings readily learn and prefer the visual characteristics of their siblings over those of an adult hen (see also Lickliter, 1989). Visually imprinted preferences for a maternal hen were found only if the ducklings were reared in social isolation or when siblings were not one of the social stimuli choice tests.

Social mechanisms

of species-specific

audio /visual

presented in the simultaneous

responsiveness

The results of more recent work with bobwhite quail chicks provide further evidence for the important contribution that social experience with broodmates makes to the perceptual preferences shown by young precocial hatchlings. Under normally occurring circumstances, the newly hatched precocial chick receives ongoing stimulation in both the auditory and visual modalities from mother and siblings and must process that information appropriately to control its behavior in an adaptive manner. Nonetheless, the nature of the relationship between auditory and visual cues routinely available in the posthatch environment, and the ways in which these two sensory modalities interact in the course of early development has received relatively little research attention (but see Bolhuis and Van Kampen, 1992; Cottlieb, 1971, 1973; Johnston and Cottlieb, 1985). To begin to investigate the role of combined auditory and visual stimulation in the control of early filial behavior, incubator-hatched bobwhite quail chicks were tested naively (without benefit of a training or imprinting trial) at either 24, 48, 72, or 96 h following hatching. During testing, the quail chicks were presented with a simultaneous choice test between species-typical auditory and visual cues versus species-atypical auditory and visual

195

cues. Choice, latency and duration

of response to these stimuli

were scored as in the

duckling studies described earlier. Results revealed a hierarchy in the functional priority of the auditory and visual systems in the days immediately following hatching (Lickliter and Virkar, 1989). At 24 and 48 h of age, species identification was found to depend on the auditory component of maternal stimulation (the species-typical maternal call>. At 72 and 96 h of age, combined auditoryvisual maternal stimulation was necessary to control bobwhite chicks’ filial behavior. However, when quail chicks were denied the opportunity for social interaction with siblings following hatching (i.e. reared in isolation), they sustained a naive auditory preference for the bobwhite maternal call into these later stages of postnatal development (through 96 hr of age). That is, chicks reared in social isolation exhibited decelerated development of intersensory functioning when compared to chicks allowed ongoing social interaction with broodmates. These data indicate that in the course of normal development, integrated auditory-visual functioning will appear in bobwhite quail hatchlings by about 72 h of age, but its appearance depends, at least in part, on experiential factors that are a reliable part of the natural social context in which development normally occurs. A more recent study has found similar results for early species-specific visual responsiveness (McBride and Lickliter, 1993). In this study, the social context was manipulated by rearing bobwhite chicks in one of four different post-hatch rearing conditions. In the ‘group’ condition, chicks were reared with same-age bobwhite chicks. These chicks had the opportunity to see, hear, and physically interact with a group of conspecifics from hatching to testing. In the ‘non-conspecific’ condition, bobwhite chicks were reared with same-aged scaled quail chicks. These subjects also had the ongoing opportunity to see, hear, and physically interact with these non-conspecific chicks from hatching to testing. In the ‘isolation’ condition, each chick was reared in separate compartment. These chicks were thus unable to see or physically interact with other chicks. They were, however, able to hear the vocalizations of other bobwhite chicks present in the rearing room. In the ‘partial isolation’ condition, chicks were reared separately, but they had the opportunity to see and hear one other bobwhite chick through a glass divider. This condition allowed ongoing visual and auditory exposure to another bobwhite chick, but prevented direct social interaction. Chicks were subsequently tested at either 48 h or 72 h following hatching for their preference between the bobwhite maternal call paired with a stuffed model of a bobwhite hen versus the bobwhite maternal call paired with a stuffed model of a scaled quail hen. This configuration required subjects to direct their social preferences on the basis of available visual cues, as the identical maternal auditory cues did not allow a basis for decision. The results of this study revealed that only chicks reared in a group with same-aged conspecifics exhibited a species-specific visual preference for the stuffed natural model of a bobwhite hen (when presented with maternal auditory cues>. Subjects that were reared in isolation, partial isolation, or with scaled-quail chicks did not demonstrate this speciesspecific visual preference in the simultaneous audio-visual choice tests (McBride and Lickliter, 1993). These findings indicate that visual experience and social interaction with conspecifics is necessary to foster or facilitate the development of this type of visual responsiveness for maternal visual cues in bobwhite chicks. These findings provide further support for the view that the minimum unit for developmental analysis must be the developmental system, comprising both the animal and its stimulative environment. This methodological insight has been highlighted several times over the last decade (see Johnston, 1982; Cottlieb, 1991a; Lickliter and Ness, 1990;

196

Oyama, 1985>, and a number of biologists and psychologists have cogently argued that the relation between the organism and its environment, rather than simply the organism itself, is the appropriate focus for the study of problems of perception (Brunswik, 1952, 1956; Gibson, 1966, 1979; Johnston and Turvey, 1980; Turvey and Shaw, 1979).

Conclusions

and a look to the future

The results of the studies concerned with the effects of the social environment

on early

perceptual preferences reviewed here emphasize the important connection between species-typical surroundings and the expression of species-typical behavior. In particular, the studies reviewed here reveal the nonarbitary connection between the developing organism’s species-typical social context and species-typical behavior and underscore the need for increased experimental attention to naturally occurring social factors in any description or analysis of perceptual or behavioral development. In this light, it is unfortunate that many studies concerned with aspects of perceptual development continue to devote little attention to features of the subject’s usual rearing environment, despite our growing knowledge that rather subtle alterations in a subject’s social environment can alter the expression of a variety of species-typical behaviors (see Hofer, 1987; Mason and Capitanio, 1988; Porter, 1990 for examples from mammalian species; see Baptista and Petrinovich, 1984; Logan, 1992 for examples from altricial avian species). These results stress the importance of the empirical task of defining the relevant developmental features of an organism’s environment, of delineating the actual physical, biological, and social aspects of the surround

with which the developing organism actually

interacts (see Alberts and Cramer, 1988; Khayutin, 1985; McBride and Lickliter, 1993; West and King, 1988 for examples). Heuristically, this emphasis on the animal-context system serves to move investigators from a predominantly gene-centered view of the heritable determiners of behavioral phenotypes to a view that includes a large class of extragenetic ‘hereditary’ variables that have often been omitted from or overlooked in analyses or explanations

of behavior development (Lewontin,

1991;

Lickliter

and Berry,

1990; Lickliter and Ness, 1990). This conceptual and methodological shift is founded on a somewhat unconventional view of heredity, one that recognizes that what is inherited in reproduction are genes and the developmentally

relevant features of the organism’s species-typical environment.

In

other words, it is developmental systems, including not only the genes but also the features of the environment that influence development, that are transmitted between generations (Johnston and Gottlieb, 1990; Oyama, 1985, 1988, 1989; West and King, 1987). This view emphasizes that what is passed on or made available in reproduction is both a structured genome and a structured segment of the world (Lickliter, 1993). By characterizing not only the genes but also the stimulative developmental context as partners in providing the basic, necessary elements for behavioral development, this expanded view of heredity makes the animal-context transaction process the explicit object of study, thereby including a large class of extragenetic variables that have often been omitted from analyses or explanations of species-typical behavior (see Lewontin, 1991 for a more comprehensive statement of this shift). In addition, although many students of behavior have tended to attribute invariance and transgenerational phenotypic stability only to genetic factors, the expanded view of heredity favored here recognizes that there are many extragenetic influences to which

197

organisms are subjected to during their prenatal and postnatal development that are also relatively invariant (see Emlen, 1972; Mason, 1978; Hofer, 1978; Cottlieb, 1980; Lickliter and Virkar, 1989; Johnston and Gottlieb, 1990). Like genes, these influences also contribute to phenotypic regularity across generations (i.e. ‘species-typicality’). For example, the presence of conspecifics is virtually guaranteed in the context of normal development of precocial avian hatchlings and, as we have shown in this review, early social experience with these conspecifics is essential for the normal development of early species-typical perceptual and social preferences. West and King (1987) have proposed the concept of ‘ontogenetic niches’ to represent the ecological and social legacies transmitted to the young of a species. They argue that a formal name for “the set of ecological and social circumstances surrounding organisms engaged in the business of development (p. 550)” is necessary to give extragenetic inheritance equal status with genetic inheritance in theories of phenotypic development (see also Johnston and Cottlieb, 1990). Regardless of terminology, it seems clear that an expanded view of heredity, one which recognizes that what is inherited from one generation to the next are genes and the developmentally relevant features of the organism’s specific environment, can afford investigators a valuable heuristic in advancing our understanding of the development of species-specific perception.

Acknowledgements The writing of this article was supported by a National Institute of Mental Health Grant (#MH48949) awarded to Robert Lickliter. We thank Gilbert Gottlieb for his constructive comments on an earlier version of the manuscript. We followed both national and institutional guidelines for the care and use of animal subjects.

References Alberts,

J.R. and Cramer, C.P.,

developmental

1988.

change. In: E.M.

Ecology and experience: Blass (Editor),

Handbook

Sources of means and meaning of of Behavioral

Neurobiology,

Plenum

Press, New York, pp. l-39. Alberts,

J.R. and Gubernick,

parent-offspring Offspring Baptista,

Interactions,

L.F.

hypothesis

In:

1966.

Bjarvall, A., 1967.

and H.

Plenum Press, New York,

pp. 7-45.

L.,

Behaviour,

1986.

J.J., 1991.

Bolhuis,

J.J. and Van Kampen,

Chicago.

sensitive

Behav., 32: 172-l

Alarm call responsivity 100:

Symbiosis

model of in

Parent

phases and song templates 81.

Biol. Rev., 41:

177-220.

H.S.,

of mallard ducklings:

IV. Effects of social

401-405.

Mechanisms of avian imprinting:

120:

Bowlby, J., 1969.

(Editors),

28: 141-147. D.B.,

Bolhuis,

E.,

Social interaction,

The characteristics and context of imprinting.

experience. J. Comp. Psychol.,

Brunswik,

1984.

Moltz

The critical period and the interval between hatching and nest exodus in rnallard

Blaich, C.F. and Miller,

Behaviour,

Reciprocity and resource exchange: A symbiotic Rosenblum

and Petrinovich,

L.A.

in the white crowned sparrow. Anim.

Bateson, P.P.G., ducklings.

D.J., 1983.

relations.

1992.

A review. Biol. Rev., 66: 3033345.

An evaluation of auditory

learning in filial imprinting.

195-230. Attachment and Loss: Vol. 1. Attachment.

1952.

The

Conceptual

Foundations

Basic Books, New York.

of Psychology.

University

of Chicago Press,

198 Brunswik,

E.,

1956.

University

Perception

Dyer, A.B.,

Lickliter,

R. and Gottlieb,

under experimentally Eibl-Eibesfeldt, Emlen,

and the

Representative

Design

of Experiments

in

Psychology.

of California Press, Berkeley.

I., 1970.

ST.,

1972.

ontogenetic development

G.J. Jacobs and R.F.

NASA SP, Washington, Galef, B.C.,

1985.

Maternal and peer imprinting

in mallard ducklings

Dev. Psychobiol.,

22: 463-475.

Ethology: The Biology of Behavior. Rinehart and Winston,

The

Schmidt-Koenig,

G., 1989.

simulated natural social conditions.

Belleville

of orientation (Editors),

capabilities.

Animal

New York. In:

Orientation

S.R.

Galler,

DC, pp. 262.

Social learing in wild norway rats. In: T.D.

Issues in the Ecological Study of Learning, Erlbaum,

Johnston and A.T.

Hillsdale,

Pietrewicz (Editors),

NJ, pp. 143-166.

Gibson, J.J., 1966.

The Senses Considered as Perceptual Systems.

Houghton-Mifflin,

Boston.

Gibson, J.J., 1979.

The Ecological Approach to Visual Perception. Houghton-Mifflin,

Boston.

Gottlieb, C., 1965.

Imprinting

Comp. Physiol. Cottlieb,

G.,

Psychol.,

1971.

K.

and Navigation,

in relation to parental and species identification by avian neonates. J.

59: 345-356.

Development

of Species Identification

in Birds.

University

of Chicago Press,

Chicago. Gottlieb, G., 1973. Psychol. Bull.,

Neglected developmental variables in the study of species identification in birds, 79: 362-372.

Gottlieb, G., 1976.

The roles of experience in the development of behavior and the nervous system.

In: G. Gottlieb (Editor),

Development of Neural and Behavioral Specificity, Academic Press, New

York pp. 25-54. Gottlieb,

G.,

1980.

Development

of species

identification

in ducklings:

VI.

Specific embryonic

experience required to maintain species-typical perception in Peking ducklings. J. Comp. Physiol. Psychol.,

94: 579-587.

Gottlieb, G., 1991 a. Experiental canalization of behavioral development:

Theory.

Dev. Psychol.,

27:

35-39. Gottlieb, G., 1991 b. Social induction of malleability in ducklings. Anim. Gould, J.L., Harper,

L.V.,

Hess, E.H., Hofer,

1982.

P.H.

1989.

1973.

M.A.,

The Nature of Human Behavior. Ablex, Norwood, Imprinting.

1978.

Hofer, M.A.,

1987.

1985.

Pietrewicz

New York.

New York.

A psychobiologist’s

In: P.P.G.

Press, New York,

Bateson and pp. 135-166.

view. Child Dev., 58: 633-647.

Learning and the evolution of developmental systems. In: H.C. Plotkin (Editor),

Learning, Development, A.T.

Reinhold,

Perspectives in Ethology, Vol. 3, Plenum

Early social relationships:

1982.

Johnston, T.D.,

Van Nostrand

Norton,

NJ.

Hidden regulatory processes in early social relationships.

Klopfer (Editors),

Johnston, T.D.,

Behav., 41: 953-962.

Ethology: The Mechanisms and Evolution of Behavior. W.W.

and Culture,

Wiley,

New York, pp. 41 I-442.

Conceptual issues in the ecological study of learning. In: T.D.

(Editors),

Issues in the Ecological Study of Learning,

Erlbaum,

Johnston and

Hillsdale,

NJ, pp.

1-24. Johnston,

T.D.

and Gottlieb,

G., 1981.

Development

What is the role of imprinting? Anim. Johnston,

T.D.

and Gottlieb,

G., 1985.

preferences in Peking ducklings. Johnston,

T.D.

evolution.

and Gottlieb,

Effects of social experience on visually

Dev. Psychobiol.,

G., 1990.

J. Theoretical

Biol.,

147:

and Turvey,

M.T.,

1980.

Johnston, T.D.

In: G. Bower (Editor),

of visual species identification

in ducklings:

Behav., 29: 1082-1099. imprinted

maternal

18: 261-271.

Neophenogenesis:

A developmental

theory of phenotypic

471-495. A sketch of an ecological metatheory for theories of learing.

The Psychology of Learning and Motivation. Vol. 14, Academic Press, New

York. pp. 147-205. Khayutin, S.N., 15: 105. King, J.A., 1968.

1985.

Sensory factors in the behavioral ontogeny of altricial birds. Adv. Study Behav.,

Species specificity and early experience. in: G. Newton and S. Levine (Editors),

Experience and Behavior, Thomas,

Springfield,

pp. 42-64.

Early

199 Kruijt, J.P., Ten Cate, C.J. and Meeuwissen,

C.B.,

1983.

The influence of siblings on the development

of sexual preferences of male zebra finches. Dev. Psychobiol., Lewontin, Lickliter,

R.C., 1991. R., 1989.

Biology as Ideology. Harper Collins,

16: 233-239.

New York.

Species-specific auditory preference of bobwhite quail chicks (Colinus

is altered by social interaction with siblings. J. Comp. Psychol., Lickliter,

R., 1993.

Structured

the construction

organisms and structured

103:

environments:

of learning capacity. In: J. Valsiner

and H.G.

virginianus)

221-226. Developmental

Voss (Editors),

systems and

The Structure

of

Learning Processes, Ablex, Norwood. Lickliter,

R. and Berry, T.D.,

tion of evolutionary Lickliter,

1990.

The phylogeny fallacy: Developmental

R. and Gottlieb, C., 1985.

species-specific

psychology’s misapplica-

theory. Dev. Rev., 10: 322-338. Social interaction with siblings is necessary for visual imprinting of

maternal preferences in ducklings (Anas

J. Comp. Psychol.,

platyrhynchos).

99:

371-379. Lickliter,

R. and Gottlieb, G., 1986a. Training ducklings in broods interferes with maternal imprinting.

Dev. Psychobiol., Lickliter,

R.

19: 555-566.

and Gottlieb,

G.,

198613.

Visually

redirected by social interaction with siblings. Lickliter,

R. and Gottlieb, G., 1987. R. and Gottlieb, R. and Ness, J.W.,

Comp. Psychol., Lickliter,

104:

R. and Virkar,

1990.

21 I-21

Social specificity:

Logan, C.A., 1992. Lorenz,

K., 1965.

M.

Dev. Psychobiol.,

Intersensory

functioning

in bobwhite quail chicks:

analysis in behavioral systems:

The case of bird song. Ann. NY of Chicago Press, Chicago.

The

Development

of Behavior:

Comparative

and Lickliter,

107:

In: G.M. Burghardt and

and Evolutionary

Aspects,

pp. 233-251. 1988.

Formation

and expression

R., 1993.

of filial attachment in rhesus

Dev. Psychobiol.,

21: 401-430.

Social experience with siblings fosters species-specific

siveness to maternal visual cues in bobwhite quail chicks (Colinus Miller,

Early sensory

22: 651-667.

and Capitanio, J.P.,

T.C.

21: 31 I-321.

and comparative psychology: Status and strategy. J.

monkeys raised with living and inanimate mother substitutes. McBride,

40-46.

Social experience and primate cognitive development,

(Editors),

W.A.

101:

Interaction with own species is necessary to

Evolution and Modification of Behavior. University

Garland Press, New York, Mason,

is

102-117.

1978.

Bekoff

in ducklings

8.

Developmental

Acad. Sci., 662: Mason, W.A.,

Domestication

P., 1989.

dominance. Dev. Psychobiol.,

preference

19: 265-277.

in ducklings. J. Comp. Psychol.,

G., 1988.

foster species-specific maternal preference in ducklings. Lickliter,

maternal

Retroactive excitation: Posttraining social experience with siblings

consolidates maternal imprinting Lickliter,

imprinted

Dev. Psychobiol.,

virginianus).

reson-

J. Comp. Psychol.

320-327. D.B.

and Gottlieb,

G., 1978.

Maternal vocalizations

of mallard ducks. Anim.

Behav., 26:

1178-1194. Oyama, S., 1985. University

The Ontogeny of Information:

Oyama, S., 1988.

Statis, development, and heredity. In: M.W.

ary Processes and Metaphors, John Wiley, Oyama, S., 1989.

Systems and Evolution.

Cambridge

Systems and Development, L.,

Dewsbury

Ho and S.W.

Fox (Editors),

Evolution-

New York, pp. 255-274.

Ontogeny and the central dogma: Do we need the concept of genetic program-

ming in order to have an evolutionary Petrinovich,

Developmental

Press, New York.

1990.

(Editor),

Erlbaum,

perspective? In: M.R.

Hillsdale,

Avian song development: Contemporary

Gunner

and E. Thelen

(Editors),

NJ, pp. 1-34. Methodological

and conceptual issues.

Issues in Comparative Psychology,

Sinauer,

In:

D.A.

Sunderland,

MA,

pp.340-359. Porter, R.H., 1990. Littermate influences on behavioral and physiological development in spiny mice. In: D.A. Dewsbury (Editor), Contemporary Issues in Comparative Psychology, Sinauer, Sunderland, MA, pp. 300-316. Sluckin,

W.,

1973.

Imprinting

and Early Learning. Aldine, Chicago.

200 Smith-Gill, tion.

S.J., 1983.

Am. Zool.,

Suboski,

M.D.,

Turvey,

M.T.

1990. and

perception

for

Research, Valsiner, Wang,

A. and

Comp. West,

M.J.

M.,

106:

King,

vs. phenotypic

modula-

A.P.,

Psychol.

of perceiving: L.C.

Nilsson

Rev., 97: 271-284.

An

ecological

(Editors),

reformulation

Perspectives

on

of

Memory

Influence

of Children’s

Action.

of social environment pennsylvanicus)

John Wiley, on parental

and prairie

New York. behavior

voles (M.

and

pup

ochrogaster).

J.

71.

1987.

Settling

nature

and

nurture

into

an ontogenetic

niche.

Dev.

20: 549-562.

males. Nature, M.J.,

In:

voles (Microtus

West, M.J. and King, A.P., 1988. West,

conversion

NJ, pp.221 -265.

1992.

163-l

learning.

primacy

and the Development

of meadow

and

Developmental

recognition The

memory.

Hillsdale,

Culture

Psychol.,

Psychobiol.,

R., 1979.

understanding

Novak,

development

plasticity:

Releaser-induced

Shaw,

Erlbaum,

J., 1987.

Developmental

23: 47-55.

334:

King, A.P.

Visual displays of female

cowbirds

affect the development

of song in

244-246. and Arberg, In:

A.A.,

legacies

in ontogeny.

Plenum

Press, New York, pp. 41-62.

E.M.

1988.

Blass (Editor),

The inheritance Handbook

of niches:

of Behavioral

The role of ecological Neurobiology,

Vol.

9,