animal behavior: a continuing synthesis...animal behavior and closer ties to neuroscience. changes...

Annu. Rev. Psychol. 1993. 44:675-708Copyright © 1993 by Annual Reviews Inc. All rights reserved

ANIMAL BEHAVIOR: A ContinuingSynthesis

William Timberlake

Department of Psychology and Center for the Integrative Study of Animal Behavior,Indiana University, Bloomington Indiana 47405

KEYWORDS: comparative methods, ethology, behavior systems, animal cognition, motivation

CONTENTSINTRODUCTION ..................................................................................................................... 676

The Historical Synthesis of Animal Behavior .................................................... 676Current Status of the Synthesis .......................................................................... 677A Reconsideration of the Synthesis .................................................................... 678

PERSISTENT RESEARCH DIFFERENCES ........................................................................... 679Field vs. Laboratory Methods ............................................................................ 679Genes vs. Environment ...................................................................................... 680Rules Guiding Research ..................................................................................... 680

METHODS OF COMPARING BEHAVIOR ........................................................................... 682Protoevolutionary Comparisons ........................................................................ 683Phylogenetic Comparisons ................................................................................ 687Ecological Comparisons .................................................................................... 689Microevolutionary Comparisons ....................................................................... 691Integration ......................................................................................................... 692

THE DYNAMICS AND STRUCTURE OF BEHAVIOR ....................................................... 694Motivation .......................................................................................................... 694Learning ............................................................................................................. 695Regulatory and Structural Behavior Systems ..................................................... 695

COGNITIVE ABILITIES .......................................................................................................... 696Comparative Cognition ...................................................................................... 697Mental Life and Consciousness .......................................................................... 698Animal Welfare and Rights ................................................................................ 699

TOWARD CONTINUING SYNTHESIS ................................................................................. 700

6750066-4308/93/0201-675502.00

www.annualreviews.org/aronlineAnnual Reviews

Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.

http://www.annualreviews.org/aronline

676 TIMBERLAKE

INTRODUCTION

As a synthetic discipline, animal behavior lives by the wits, good will, andagreement of the scientists creating it. Current interest in animals is high, andpowerful new techniques are available to establish the mechanisms, develop-ment, evolution, and function of behavior. Yet the coherence of the disciplineis low, its conceptual center increasingly little more than a weighted average ofnew interests and old allegiances. Some of this lack of coherence can be tracedto incomplete integration of parts of the original synthesis of the field. Thepresent review considers four areas of conflict and truncated development forwhich further integration should contribute to a more stable base: (a) researchdifferences, (b) methods of comparing behavior, (c) the dynamics and struc-ture of behavior, and (d) cognitive abilities.

The Historical Synthesis of Animal Behavior

The discipline of animal behavior emerged in the late 1950s and early 1960sfrom the synthesis of comparative psychology and ethology. The initial inter-action of these fields was dominated by sharp disagreement over the relativeimportance of instinct and learning in determining behavior (Lehrman 1953).Ethologists emphasized functional stereotyped motor movements and stimulussensitivities that occurred without specific experience, presumably based ongenetic programming. Psychologists emphasized the malleability of develop-ment, often focusing on nonfunctional outcomes as evidence of the importanceof learning. A concise illustration of this conflict is Grohmann’s (1939) dem-onstration that pigeons reared in confined boxes flew at the first opportunity,pitted against Dennis’s (1941) discovery that vultures raised in similarly con-stricted circumstances failed to fly and fell to the ground when placed on aperch.

The unexpected outcome of this apparently intractable conflict was synthe-sis of the new and energetic discipline of animal behavior (Hinde 1966; Marler& Hamilton 1966). A major factor promoting this synthesis was the surprisingnumber of interests shared by ethology and comparative psychology, includingconcern with development, motivation, evolution, adaptation, physiologicalmechanisms, small stimulus-response units, and stimulus control. Othershared concerns included a strong empirical orientation and a distaste forvitalism. A critical role in the synthesis also was played by personal tiesbetween individual researchers, such as between the ethologist Lorenz and thepsychologist Lehrman. The contributions of Frank t~each along these lines aredescribed in a brief memorial (Dewsbury 1989b), while Tinbergen (posthu-mously) and Robert Hinde were recently honored with essay collections docu-menting their influence (Bateson 1991; Dawkins et al 1991).


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 677

Current Status of the Synthesis

The synthesis of comparative psychology and ethology has been quite success-ful. Authors of the last three chapters on this subject in the Annual Review ofPsychology have documented the robustness of animal behavior research(Dewsbury 1989a; Mason & Lott 1976; Snowdon 1983). Opportunities forprofessional publication continue to increase with the expansion of old jour-nals and the introduction of several new ones (Journal of Biological Rhythms,Behavioral Ecology, and Journal of Cognitive Neuroscience are recent exam-ples). To stay current with just the titles of relevant books and chapters pub-lished each year requires a significant amount of time.

There are, however, some mixed indicators. The relative success of the fieldin securing government support for basic research is uncomfortably low.Though training funds in animal behavior recently received a slight increase,American government agencies typically fund only 10-15% of research pro-posals in this area (see Myers 1990, for an account of the process from theapplicant’s view). University positions related to animal behavior appear tohave decreased. Research with dogs, cats, and rabbits has declined steadily(Viney et al 1990). Paranoid feelings remained high for years in reaction to O. Wilson’s (1975) argument that comparative psychology had died and ethol-ogy and physiological psychology would follow by the year 2000, cannibal-ized by integrative neurophysiology, sociobiology, and behavioral ecology.Though concern has subsided, self-doubt continues. Bateson & Klopfer (1989)edited a volume entitled "Whither Ethology?" Demarest (1981) presided forseveral years over a self-examining newsletter on comparative psychology.Leger (1988: ix) noted, "I know of no discipline that has devoted so muchtime, effort, and printer’s ink to a lengthy debate concerning its territoriallimits .... "

At an accelerating pace important discoveries relevant to animal behaviorcome from outside the discipline. These discoveries provide fascinating oppor-tunities for assimilation but also may bring into question the adequacy andstability of the synthesis. Recent advances in allometry (Gittleman 1989b) andthe hormonal control of natural behavior (Wingfield & Moore 1988) havebeen dominated by physiologists. Molecular neuroscience has assumed amajor portion of the search for basic mechanisms of learning, kin recognition,hormonal control of behavior, and rhythmicity (see Becker et al 1992;Martinez & Kesner 1991). Research in motivation has been co-opted by ani-mal welfare concerns (Dawkins 1990; Hughes & Duncan 1988). The task modeling animal behavior has leaned heavily on approaches developed inother fields. These approaches include: optimality, from economics and ap-plied mathematics (Alexander 1982); connectionism, from cognitive psychol-ogy and computer science (Grossberg 1988); and stochastic dynamic modeling(Mangel & Clark 1988) and nonlinear dynamical systems (Beltrami 1987),from mathematics and physical systems.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


678 TIMBERLAKE

Relevant new fields, such as artificial life, spring up without warning,complete with conferences, a newsletter (Alife Digest), and journals (AdaptiveBehavior and Artificial Life). Slightly older fields, such as sociobiology andbehavioral ecology, continue to pursue different visions of causation (thoughsee the suggestion of Krebs & Davies 1991 and Bell 1991 that behavioralecology return to mechanism). Other fields, such as neuroethology (Ingle Crews 1985) and developmental psychobiology (Blass 1986) waver betweenanimal behavior and closer ties to neuroscience. Changes in fundamentalideas, like the nature of evolution (Eldridge 1989), reinterpretations of themodern evolutionary synthesis (Provine 1988), and the active role of genes daily behavior (Rusak et al 1990) provide continuing challenges to the task constructing a coherent account. Because animal behavior remains a syntheticdiscipline, the critical issue for the field is maintaining an historicallygrounded coherence while assimilating the flood of new developments.

A Reconsideration of the Synthesis

In several respects the original synthesis was incomplete. For some the gapbetween comparative psychology and ethology has closed (Dewsbury 1990).But others still see marked differences in training, research focus, compari-sons, procedures, and models~ifferences sufficiently large to promote sepa-rate professional allegiances. For example, the International Ethology Confer-ence and the journal Ethology maintain a separate identity, as do the Interna-tional Society of Comparative Psychology, the Journal of Comparative Psy-chology, and the International Journal of Comparative Psychology. Some

even feel the name "animal behavior" is an issue because it might be seen asexcluding humans. We are all animals here and we have a worthy heritage.

Much of the historical coherence of animal behavior has been provided byTinbergen’ s (1951) four questions concerning evolution, function, mechanism,and development of behavior. These questions explicitly encourage a balanceof approaches and causal levels, establish boundaries to the field, and helphighlight research concerns overshadowed by rapid developments in othertopic areas (e.g. Barlow 1989; Bateson & Klopfer 1989; Dawkins 1989).However, Tinbergen’s questions by themselves do not provide a unifyingpicture of the fit between animal and environment. The clarity of theethologist’s original picture has been lost in the ensuing forty years. Thepurpose of the present review is to reconsider some unresolved conflicts andundeveloped strengths of the original synthesis in the hope that their furtherdevelopment and integration might help provide a more coherent center for thestudy of animal behavior. Even with this regrettable restriction it was possibleto sample a only small portion of the thousands of excellent publications sincethe last review.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 679

PERSISTENT RESEARCH DIFFERENCES

Field vs. Laboratory Methods

In the extreme view, comparative psychologists have been associated with alaboratory approach to research--the use of arbitrary stimuli, "artificial" labo-ratory-bred animals, automatic recording of arbitrary response elements, andextensive statistical analysis. Ethologists have been associated with a fieldapproach to research--including extensive observation of animals in uncon-trolled field conditions, a focus on highly complex behaviors, and im-pressionistic reporting. These stereotypes, always of doubtful accuracy, areinappropriate now that laboratory and field approaches have become increas-ingly difficult to distinguish (Blanchard et al 1989; Dewsbury 1990; Gibbonset al 1992). However, differences remain in how frequently individual investi-gators use the full range of research options.

Some researchers argue that field work must precede laboratory analysis(external validity first, see Kamil 1988). Others feel that phenomena must established in the laboratory before dealing with the complexity of the field(internal validity first). The best long-term strategy is to include a range approaches, thereby providing both sorts of validity and avoiding obstaclesthat arise simply from pursuing one approach exclusively. For example, a highdegree of environmental control of stimuli allowed remarkable insight into thedevelopment of structure in birdsong, but the recent demonstration that livetutors produce different effects than tapes highlights the importance of alsochecking environmental influences on behavior in less constrained circum-stances (see Marler 1991; Petrinovich 1990).

There is no single correct environment or species for research or observa-tion, just the need to avoid downplaying either the contribution of the animalor the contribution of the environment. Different environments and speciesprovide different opportunities to analyze the fit between animal and environ-ment. Heidiger (1950) drew on this perception in arguing that the behavior animals in zoos often was determined by so-called fight and flight distances. Ina similar vein, Tinbergen & Perdeck (1950) and Hailman (1967) explored possibilities of using artificial stimuli to analyze the mechanisms underlyingbegging in young birds. Even quite restricted environments can reveal charac-teristic animal/environment relations. For example, Olson (1991) used artifi-cial learning problems to explore memory differences in corvids related tofood caching, and Timberlake & Washburne (1989) used artificial movingstimuli to explore how the feeding ecology of rodent species affects condi-tioned reactions to movement predicting food.

Timberlake (1990) argued that behavior in even the most restricted labora-tory environments is related to evolutionary determinants: first, because ani-mals do not have a separate evolved repertoire for dealing with restrictedenvironments; second, because even the most rigorous laboratory scientistscarefully modify their procedures, apparatus, and measures to deal with the


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


680 TIMBERLAKE

specific species under study (see Skinner’s 1938, 1959 account of developingthe leverpress for rats). To recapitulate, there need be no fundamental conflictbetween laboratory and field, provided we view them as different opportuni-ties for reconstructing and predicting the fit between animal and environ-ment.

Genes vs. Environment

The basic question of whether behavior is determined by genes or environmentappeared to be resolved in the original synthesis. The concept of epigenesiscaptured the notion that development is not directed toward a preformed end,but occurs as an interaction between genes and environment (Oppenheim1982). However, the nature of this interaction has proved difficult to analyze.Notions such as open instincts (Mayr 1974) and learning-instinct intercalation(Lorenz 1965) are not particularly epigenetic in the scope of interaction theyallow for. Development does not occur as the programmed or even as thestatistical interaction of genes and environment. Instead it seems to consist ofindividual, largely self-organizing processes involving genes, cells, physiolog-ical systems, and the "outer" environment (Thelen 1990; Wikler & Finlay1989). Like evolution, development is not goal-directed but based on "spit andbailing wire"; that is, on making do with what is available. It is a vector thathas direction in terms of its constituent processes (Oyama 1985).

Perhaps the most difficult point to grasp is that genes are not causal entities(Bateson 1988). Information about the "finished" animal does not reside ex-clusively in the genes any more than it resides in the environment. Identifyinggenes as causal agents remains a useful fiction because it makes possiblesimple manipulations and predictions of behavior in a particular environment.The fiction is made more attractive because proposed alternatives to simplecausation often invoke maximal dialectical complexity (everything affectseverything), a stance that gives only vague guidance to research (Kuo 1976).At present, the most rapid progress toward an understanding of developmentappears to come from using experimental manipulations to open windowsduring ontogeny that reveal how developmental processes work (Blass 1988;Moore 1990; Sinervo & Huey 1990; Thelen & Ulrich 1991).

Rules Guiding Research

Many successful scientists, just like successful authors and parents, oftenreconstruct their lives to yield a set of rules to follow (see Dewsbury 1985).Some recommend starting with first principles and deducing predictions thatpit one hypothesis against another (strong inference). Others advise their stu-dents to observe extensively and without preconception, drop everything tofollow a new lead, and then persist in the face of adversity. In practice mostadept scientists appear to be more like game theorists, using conditional ratherthan absolute rules--deduction and induction, observation and manipulation,strong inference and demonstration, one species and many (Maynard-Smith


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 681

1982). Scientists also differ in individual abilities and predilections. Beer(1980) suggested that ethologists tend to be visually oriented, leading to theirconcern with behavioral form. Conversely, many psychologists began in phys-ics and engineering, making them comfortable with the construction of appara-tus and simple causal models.

Three additional suggestions may be useful in guiding integrative research:(a) assume "a modicum of ignorance" (a phrase of Nottebohm’s), (b) view things from the animal’s perspective, and (c) be aware of the functionalsystems that underlie behavior. On the issue of ignorance, it is often useful(after considerable training in an area) to return to being a little naive, to testthe interpretation so obvious it’s almost not worth checking, to question tradi-tional procedures, to ask why the form of a response varies over time. Re-searchers inevitably accrue methods, measures, and models that constrain theirquestions and interpretations. A well-placed, simple question can reveal animportant issue, e.g., Petrinovich (1990).

As to taking the animal’s view, most scientists remain cautious because ofthe risk of falling into anthropomorphism--the projection of human thoughtsand feelings into other animals. Though some have argued for its heuristicvalue and complexity (Fisher 1991), there is little evidence that research hasbeen advanced in the long run by assuming that animals are fundamentallyhuman in their thinking and behavior (Staddon 1989). The term theromorph-ism--meaning animal-centered as opposed to human-centered (yon Uexkull1934; see also Burghardt 1991; Simmons 1989)--better captures the presentidea. Based on knowledge of an animal’s sensory and motor equipment, itsintegrating and processing capabilities, and its motivational structure, dynam-ics, and decision rules, one attempts to enter into the animal’s view of thecircumstances. This is usually a difficult bootstrap endeavor involving askingrepeated experimental and observational questions about an animal, and con-sidering carefully the answers. It provides, though, an integration of intuitionand experimental results that is essential both in setting up reasonable experi-mental manipulations and in interpreting and modeling more freely occurringbehavior.

Finally, a recurrent theme in this discussion is that animals are sets offunctional systems that operate within and influence the context of environ-ment, physiology, behavior, social relations, and evolution. A simpler causalapproach often dominates because researchers appreciate the power of analyz-ing and modeling the effects of one or two variables at a time. The problemwith this simpler approach is the danger that our "picture" of the animal---ourmeans of organizing knowledge--will become dependent primarily on ourexperimental procedures and simple causal models. Such a picture often is nottransferable to new situations or even to different manipulations. For example,careful manipulation and measurement of copulation in single pairs of ratsproduced initial insight into the underlying physiological mechanisms (Rose1990). However, broader comparisons of mammalian copulatory patterns


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


682 TIMBERLAKE

(Langtimm & Dewsbury 1991) introduced considerations of phylogeny andadaptation. Examining the social context of mating in rats clarified the roles offemale and male choice as well as the selection pressures on the mechanisms(McClintock 1984). The hormonal analysis of reproduction provided furtherinsight into evolutionary processes and mechanisms (Moore 1990).

Taking a systems approach has enabled researchers to organize informationto develop a view of animal and environment that is conceptually and experi-mentally more tractable and heuristic.

METHODS OF COMPARING BEHAVIOR

Comparison has been touted as the heart of both comparative psychology andethology (e.g. Dewsbury 1990; Lorenz 1950). Comparisons provide informa-tion about evolution and ecology, as well as suggesting and testing hypothesesabout mechanism. Yet nowhere in animal behavior is the absence of a satisfy-ing synthesis clearer than in the lack of agreement about the nature of appro-priate comparisons. Thoughtful discussions have been offered by both psy-chologists and evolutionary biologists (Beer 1980; Bell 1989; Clutton-Brock& Harvey 1984; Dewsbury 1990; Gittleman 1989b; Hailman 1988; Harvey &Pagel 1991; Riley & Langley 1993). Given the diversity of answers, it istempting to steal a page from philosophers of science and claim that thecomparative method is whatever is used by scientists in making comparisons(see Adkins-Regan’s 1990 review of comparative work on sex hormones andbehavior).

In the absence of a single compelling rationale, it may be useful to catego-rize comparisons of behavior based on levels of concern with dimensions ofgenetic and ecological relatedness. Figure 1 shows four resultant categories:protoevolutionary comparisons (those not focused on either genetic or ecolog-ical relatedness), phylogenetic comparisons (focused on genetic but not eco-logical relatedness), ecological comparisons (focused on ecological but notgenetic relatedness), and microevolutionary comparisons (concerned with bothgenetic and ecological relatedness).

Historically, comparative psychology focused on comparisons of mentallife and physiology (protoevolutionary comparisons) combined with behaviorgenetics and development (microevolutionary comparisons). Ethologists fo-cused on phylogenetic comparisons of behavior patterns in limited groups(homology), and ecological comparisons based on the form and function behavior (convergence or divergence). Given these differences, it is not sur-prising that ethologists and psychologists often talked past each other (seeLorenz’s, 1950, lament that comparative psychologists didn’t do comparativeresearch). As we shall see though, separations have blurred and empiricalmethods have emerged that cut across all categories (Harvey & Pagel 1991).


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 683

Protoevolutionary Comparisons

Faced with a large number of diverse species, most scientists are driven toimpose some form of comparative order. A few scientists strive to add evenmore species for perspective and uniqueness (Adkins-Regan 1990). In neithercase is the predominant focus on phylogeny or ecology, but on abstract rela-tions such as trends, scales, and universal laws. I refer to such comparisons asprotoevolutionary because their focus is not on the course of evolution butevolution is their broad concern.

CLASSIFICATION, TRENDS, GRADES, AND LEVELS Comparisons across speciesin psychology readily can be traced to the interests of Darwin (1871) andRomanes (1884) in establishing a continuum of mental life from simpler animalsto humans. Comparative psychologists developed more careful measures andreliable tests to document trends, grades, and levels in intelligence (e.g. Aronson1984; Bitterman 1965; Krushinski 1965; Razran 1971) and in sensory-motorcapabilities (Warden et al 1935). Recent examples of this analytic approachinclude development of a scale of categorization complexity (Herrnstein 1990)and the use of Piagetian stages of object permanence in human infants to gradespecies [e.g. primates (Parker 1990), dogs (Gagnon & Dore 1992), cats (Dumas& Dote 1991), and Psittacine birds (Pepperberg & Funk 1990)].

Biologists also have a history of classifying and grading species in a rela-tively nonevolutionary fashion, primarily with respect to morphology. For

Concern withEcological

Relatedness

High

Low

Concern with GeneticRelatedness

Low High

Ecological

Convergence

Divergence

Protoevolutionary

Classifications,Trends, & Grades

Scaling Functions &AIIometry

Universal Laws

Microevolutionary

Genetics &Development

Within-SpeciesStrategies

Micro-Model Systems

Phylogenetic

Homologies

Series

Animal Models

Figure I Types of comparison


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


684 TIMBERLAKE

example, animals have been classified grossly on the basis of radial vs. bilat-eral symmetry and polarized vs. unpolarized synapses, and more precisely onthe basis of single cone vs. three cone retinas, and eusocial vs. asocial behav-iors. Trends across groups have been noted, such as the decrease in geneticrelatedness and increase in aggression in Wilson’s pinnacles of sociality(1975), or changes from two-chambered to four-chambered hearts acrossclasses of vertebrates (Hodos & Campbell 1990). Within smaller groups,behavior characteristics have been ordered in grades relating to differences incomplexity of function or design features; for example, grades of cooperationin nest spinning ants (Holldobler & Wilson 1990), or sociality in primates(Martin 1974), or in bees (Michener 1974).

The major objection to such comparisons is the tendency to interpret trends,grades, and levels as progressive and as representing the actual course ofevolution. Hodos & Campbell (1969, 1990) pointed out that many psycholo-gists assumed they were documenting the progressive evolution of mental lifefrom lower animals up to humans, an assumption often more clearly related tothe scala naturae of Aristotle than to the course of evolution. Biologists, too,have inappropriately seen evolution and progress in some trends and grades(Gottlieb 1984; Nitecki 1988). For example, the three-chambered heart foundin amphibians could be viewed as an intermediate evolutionary stage betweenthe two-chambered heart found in fish and the four-chambered heart of mam-mals and birds. Instead, the three-chambered heart appears to be an adaptationof a four-chambered heart to the skin-aeration possibilities of an amphibianecology (Hodos & Campbell 1990). The most conservative approach does notassume that classification schemes, trends, grades, and levels represent evolu-tionary history or progress. Instead, these comparisons must be reconciledindividually with the course of evolution by working out the relations amongmechanism, selection, and phylogeny.

SCALING FUNCTIONS AND ALLOMETRY The process of scaling is based onfitting a function through a scatter plot relating variables such as body mass andbasal metabolism rate (McNab 1989) or brain volume and body weight (Jerison& Jerison 1988). The result typically is a power function plotted as a straightline in log-log coordinates. The development of these relations is often referredto as allometry (Schmidt-Nielsen 1984; Thiessen 1990), because it typicallyinvolves the relation between body size and other variables, such as brain size,neural complexity, learning ability, metabolism, efficiency of locomotion, orreproductive success (e.g. Clutton-Brock & Harvey 1984; Gittleman 1989a). most cases, each point is a species, but in recent comparative work manyresearchers (e.g. Gittleman 1989b) have made each point a genus in order avoid influencing the function by correlated phylogenetic or ecological vari-ables. Allometric functions are presumed to be determined by general physicalrelations and limitations, and thus should be present in all animals, thoughdifferent taxa may have different exponents relating the variables.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 685

Taken by themselves scaling functions deal with correlation rather thancausation. They represent broad statements about the evolutionary landscapethat make little contact with specific evolutionary history, development, ormechanisms. Recent research, though, has used deviations from allometricrelations to indicate the effects of particular mechanisms and environments(e.g. McNab 1989). Also the use of convergent correlations and experimentalmanipulations can facilitate causal interpretations of allometric relations. Forexample, Sinervo & Huey (1990) tested the mechanisms underlying size-re-lated locomotion differences in related lizard species. By directly manipulatingadult body size through partial removal of yolk from the eggs of one species,they showed that burst speed but not stamina was affected by body size.

UNIVERSAL LAWS A primary goal of science is to develop general principlesor laws, preferably expressed as mathematical functions. In physics, the laws ofmechanics hold true almost without regard to the entities involved. Laws ofchemistry and biology more frequently require that the nature of the entity betaken into account. For example, to predict chemical reactions it is necessary toknow the elements involved.

It is probably unfortunate that many behavior researchers have tried so hardto emulate physics (Bolles 1988) when chemistry would have made a bettermodel. The fit of billiard ball and environment is not the same as the fit of ananimal and its environment. Something like a dynamic "periodic table" relat-ing animals, environments, and behavior is necessary before more accurategeneral laws are possible. Still, the development of general laws is a naturalgoal of science. Below are two examples of the search for laws of animalbehavior.

Beginning in the 1930s many comparative psychologists focused attentionon the development of universal laws of Pavlovian and operant conditioning.As more species were examined it became clear that similarities in learningwere greater than differences (e.g. Macphail 1985; Skinner 1959). To be sure,similarities were highlighted by tuning apparatus and procedures to minimizedifferences attributable to the sensory, motor, and motivational aspects of aspecies (Timberlake 1990). Still, the similarities were remarkable, even acrossmarkedly divergent phyla (e.g. Maier & Schneirla 1935). More recently,Bitterman and his colleagues have shown that honey bees demonstrate amajority of the specific learning effects found in laboratory rats (e.g. Bitterman& Couvillon 1991). Neuroscientists have related universal learning laws togeneral cellular mechanisms (Hawkins & Kandel 1984). Some psychologistshave attributed universal learning laws to the convergent evolution of associa-tive processes (Dickinson 1980).

In biology, the past 20 years has seen a universal law approach emerge inbehavioral ecology based on the principle of optimality. The basic assumptionis that evolved behavior maximizes net benefit to the animal, both immediately(in terms of, for example, energy) and ultimately (in terms of individual


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


686 TIMBERLAKE

reproduction and inclusive fitness). The principle of optimality is presumed todrive behavior regardless of ecology and without consideration of underlyingmechanisms (Grafen 1991). Cleverly applied, it can predict a wide range phenomena including foraging behavior (Stephens & Krebs 1986), group size(Pulliam & Caraco 1984) territoriality (Davies & Houston 1984), mating tems (Davies 1991), and parental care (Clutton-Brock & Godfray 1991). selection (a form of optimality based on maximizing reproductive success atthe level of the gene rather than the individual) has been used to explainaltruistic behavior (Grafen 1991).

Most general principles of behavior, such as those concerned with learningand optimality, arise initially from a relatively narrow base of experimentation.Integration of concerns about mechanism, phylogeny and ecology comes later.To their credit, researchers in optimal foraging have shown increasing interestin the mechanisms (so-called "rules of thumb") that produce near-optimalbehavior for a species in particular circumstances (Stephens & Krebs 1986).

For example, a bird may leave a patch based on the rule "search elsewhereafter 20 seconds without food," or it may choose between alternative patchesbased only on the immediate probability of payoff (a hill-climbing rule). some environments these rules produce optimal behavior, but in other environ-ments they may fail miserably as optimal strategies.

Comparative learning researchers also have been concerned with the mech-anisms that underlie learning. But, with a few exceptions they tend to regarddeviations from expected results as constraints on general learning principlesrather than as new information about specific adaptive mechanisms. In con-trast, the research of Garcia and his coworkers (e.g. Garcia et al 1989) hasconsistently analyzed taste aversion learning as a specific functional adapta-tion.

MODEL SYSTEMS A fourth class of protoevolutionary comparison is the modelsystems approach, an approach used frequently by physiologists to chooseanimals suitably qualified to test particular phenomena. The assumption under-lying a model system is that a general substrate of action exists for manytaxonomic groups and that its ruling principles can be investigated readily inparticular preparations. Thus physiologists studied nerve conduction in squidbecause a giant axon was readily accessible in this species. More recentlyscientists have taken advantage of the relatively small number of pathways andneurons in simple marine molluscs to study the nature of learning (e.g. Alkonet a11987). The many uses of chickens as a model system have been documentedby Andrew (1991).

Ethology, sociobiology, and behavioral ecology have also begun to use dataand general principles derived from work with other animals to analyze humanbehavior (e.g. Groebel & Hinde 1989; Hinde 1991 on Tinbergen). Wilson(1975) and those immediately following him bore the brunt of public dismaythat the same analysis could be applied to human and nonhuman animals. But


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 687

work has continued in this vein, appearing frequently in journals such asEthology and Sociobiology (see Mulder 1991). For example, Daly & Wilson(1988a,b) melded an anthropological approach with that of genetic relatednessto predict patterns of parental solicitude and homicide in humans. Thornhill &Thornhill (1991) have examined the relation of characteristics of human rapeto variables that have been shown to influence reproductive fitness in otheranimals. The evidence is compelling that some of the variance in humanbehavior is controlled by the same variables that determine the behavior ofother species.

In sum, the precise procedures, generality, lawfulness, and heuristic valueof protoevolutionary comparisons are the source of their attractiveness. Draw-backs often include lack of attention to phylogeny, specific selection pres-sures, mechanism, and development. Considerable care is required to avoidviewing all trends as progressive, attending to other species only as they relateto humans, or ignoring important species differences in the search for univer-sal laws. Still, there seems little reason to shun protoevolutionary comparisonsprovided they are seen not as an endpoint, but as a starting place.

Phylogenetic Comparisons

Ethology began in this century with biologists comparing the form and devel-opment of motor acts in related species (Wheeler in ants; Heinroth in Euro-pean songbirds and ducks; Whitman in pigeons and doves--see Thorpe 1979).Including the study of behavior was a natural extension of the taxonomictradition in biology. Though enthusiasm for tracing phylogenetic relationswaned for a bit, recent technical advances such as DNA finger-printing andmultivariate statistics have provided some insight and a glut of informationabout phylogenetic relations among species (Brooks &McLennan 1991).

BEHAVIORAL HOMOLOGIES Behavioral homologies refer to similarities in thetopography (form) and sequencing of behavior based on phylogenetic linkage.The lack of a completely satisfying definition of behavioral homology hascaused some investigators to dismiss the concept (Atz 1970). But the dataacquired in the study of behavioral homologues appear potentially useful infilling in the phylogeny of behavioral and brain mechanisms, plus serving asanother source of evidence about the overall course of evolution. For example,Lorenz (1950) used patterns of shared similarities and differences in courtshipbehavior to infer the course of evolution in a family of ducks (anatidae). Theextent to which all species of ducks showed the same displays marked theircloseness phylogenetically. Subgroups of ducks sharing unique displays wereviewed as having split off from the root stock over evolutionary time. Later workcombined evidence from behavior patterns with that based on morphological orphysiological analysis to produce convergent evidence about phylogeny [e.g.Van Tets’ (1965) work on Pelicaniformes; and Archibald’s thesis work unison calls in cranes, reported extensively in Grier & Burk (1992)].


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


688 TIMBERLAKE

A more recent form of specifying homologous relations is based on the usesof multivariate classification statistics. Typically, combinations of morphol-ogy and/or behavior are used to differentiate related groups of animals. Forexample, Eberhard (1982) combined data from web-building with informationon predatory and courtship behavior to produce a taxonomy of related spiders.Losos’ (1990) analysis of 13 species of lizards indicated that limb proportionsand locomotor behavior evolved together, while Langtimm & Dewsbury(1991) showed that similarities in copulatory behavior among the Sigmodonti-ane rodents were traceable to a common ancestor.

BEHAVIORAL AND NEUROPHYSIOLOGICAL SERIES A second method of phylo-genetic comparison is to arrange living species in a sequence based on small andapparently systematic differences in behavior or neurophysiology. Such asequence can present a compelling hypothesis of evolutionary development,though determining the validity of the sequence requires independent datarelating the species. In behavior, a classic example is Kessel’s (1955) arrange-ment of the courtship behavior of empid flies in a sequence beginning withcourting males presenting the female with captured prey and culminating witha species in which the males present empty gossamer puffs. Similar series canbe generated at the level of neurophysiology. A well-known example is theargument by MacLean (1990) for the triune evolution of control circuitry in thebrain (first brain stem, then paleocortex, and finally neocortex). Masterton et (1969) showed changes in brain development across branching phylogeneticgroups of mammals, as did Kroodsma & Konishi (1991) in comparing songbirdswith their suboscine relatives. Robinson (1991) warned that a large inductivedatabase with many intermediate forms is necessary to infer a series that allowsa distinction between evolutionary pressures and phylogeny. Brooks &McLennan (1991) argue strongly for a more computational approach in estab-lishing such series.

ANIMAL MODELS The (nonhuman) animal model approach is predicated on thenotion that basic physiological substrates and behavioral qualities of a modelspecies are homologous to those of humans. This approach has become acornerstone of modern medical research (Alter et al 1991). Common laboratoryanimals are used as stand-ins for humans in determining, for example, thepotency of a particular tranquilizer, or the physiological processes of aging (e.g.Gold & Stone 1988). Considerable objections have been raised (e.g. Regan1983) to the extent and type of commercial product testing using animal models,resulting in changes in procedures and decreases in the amount of testing. Themost sweeping objection, though, that work with nonhuman animals has norelevance to humans, is wrong, and even speciesist. Though no scientist wouldquestion the worth of continued research on determinants of the generalizabilityof results from animal models, specific results and principles stemming fromanimal models have long been useful in work with humans (for a nondisease


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 689

example see Squire’s 1992 review of converging human, primate, and rat datarelating the hippocampus and memory).

A common interest in the use of animal models is in relating the behavior ofother mammals to that of humans. For example, Young & Thiessen (1991)documented similarities between the cephalo-caudal organization of washingin humans and grooming in rodents. A less provocative finding is that allfemale mammals show a high degree of similarity in the hormonal regulationof their reproductive behavior (Rosenblatt 1989). The most explicit phyloge-netic comparisons are between humans and the great apes, especially chimpan-zees. For example, Tomonaga & Matsuzawa (1992) found that humans andchimpanzees showed the same dominance among perceptual categories in amatching to sample task. Outer contour elements were perceived the mostreadily and straight-line elements the least. The relation of language leamingin apes and humans has been the source of fascinating data and continueddebate (Gardner et al 1989; Savage-Rumbaugh 1988). In social behavior Man-son & Wrangham (1991) have reported parallels in intergroup aggression chimpanzees and humans. Some of the difficulties with behavioral homologiesare considered at the end of the next section.

Ecological Comparisons

Ecological comparisons focus on the importance of selection pressures inproducing divergence or convergence of behavior among species.

CONVERGENCE The classic ecological comparison is based on behavioral andmorphological similarities among species with dissimilar genetic make-up butsimilar environmental pressures. Usually an attempt is made to relate thesimilarities in a face-valid way to assumed selection pressures. Early work doneby Hailman (1965) and others showed that unrelated cliff nesting shorebirds,such as gannets and gulls, exhibit quite similar behavior and morphology,presumably on the basis of the common selection pressures of the cliff environ-ment. More recently, Logue (1988) provided evidence for a convergence strategies among vertebrates to avoid poisonous foods, and Vander Wall (1990)discussed ecological similarities among diverse food hoarders. Sherry et al(1989) found that families of food-storing passerines show a larger hippocampusrelative to the telencephalon and body weight than non-food-storing families.

LIFE HISTORY THEORY In this view, selection for optimal strategies predictsgeneral types of ecological convergence that approach the breadth of universallaws. Animals are grouped on the basis of their adaptations to common ecolog-ical pressures and a configuration of their behaviors is predicted. A now classicexample is Ridley’s (1983) demonstration that precopulatory guarding of thefemale by the male in invertebrates and anurans was predictable from whetherfemales were receptive during predictable brief periods of time. More recentexamples include the suggestion of Driver & Humphries (1988) that random


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


690 TIMBERLAKE

movement by prey is the result of common predation pressure, and Davies(1991) review of the data used to support the view that mating systems followfrom the defendability of resources (food for the female, and females for themale).

DIVERGENCE Divergence comparisons involve looking for deviations fromhomologous behavior or morphology that can be related to changes in inferredselection pressures. For example, Glickman & Sroges (1966) investigated theexploratory behavior of zoo animals, looking for overall phylogenetic grades inreactivity. However, the data compelled a different interpretation relating ex-ploratory behavior to the ecological variables of food variety, predator pressure,and the importance of social communication. More recently, Dewsbury (1988)used deviations and similarities in behavioral profiles to relate reproductivestrategies to ecological circumstances, and Beecher (1990) predicted andshowed that parent-offspring recognition was better in swallows that werecolonial rather than solitary nesters. Also Brown (1989) developed evidence thatthe vocal repertoires and auditory sensitivities of old world monkeys aredetermined by their functions in conjunction with the acoustic characteristics ofthe habitats. Dukas & Real (1991) supported the prediction, based on therequirements of sociality, that a social bumblebee should show faster learningabout reward than a solitary carpenter bee.

DEVIATIONS FROM ALLOMETRIC RELATIONS This is a form of divergencecomparison based on using allometric scaling functions as a central tendencyagainst which to scale the deviations of individual species. For example,Jerison’s (1973) encephalization quotient is based on the relation of the brainsize to body weight ratio of a single species relative to the ratio typical of theirtaxonomic group. The resultant deviations can be related to ecological variables.McNab (1989) noted that the fundamental relation between basal rate of metab-olism and body mass in carnivores is a power function with an exponent of .67(Schmidt-Nielsen 1984), but it is influenced (in nonlinear ways) by the type (vertebrates, invertebrates, leaves, fruit, and mixed diets), and the basicecology (arboreal, burrowing, aquatic, arctic). Used in this way, allometricscaling becomes a type of ecological comparison (see also Harvey & Krebs1990).

Finally, adaptive correlation is a form of divergence comparison based ondemonstrating a suite of relations between morphology or behavior and pre-sumed selection pressures. Cullen (1957) provided a classic example by docu-menting the differences in agonistic and reproductive behaviors between shoreand cliff-nesting gulls and explaining them in terms of the different selectionpressures produced by the cliff habitat. Another classic example is the correla-tion among feeding niche, morphological characteristics, and behavior inDarwin’s finches (Grant 1986). A variant of this approach is the work Gaulin et al (1990) and Gaulin & Wartell (1990) showing that across species


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 691

and genders of rodents, average home range size is correlated with the spatiallearning abilities shown by animals in laboratory tests.

There is a very long history of concern with the adequacy of the concepts ofhomology and convergence, a concern that has been compounded by applyingthese concepts to behavior (see Beer 1980). One fundamental difficulty is thateach type of comparison considers only half of the evolutionary variables; oneconsiders the evolutionary history, the other the selection pressure of theenvironment. Another difficulty is that most conclusions about phylogeneticand ecological relations are not checked by experimental manipulations. Stud-ies such as Hailman’s (1965) attempted to improve the accuracy of causalinferences by controlling rearing environments, but this approach appears tohave gone out of favor. Recent statistical techniques provide alternative meansof separating phylogenetic and ecological influences (e.g. Harvey & Pagel1991).

Microevolutionary Comparisons

Microevolutionary comparisons are concerned with the contribution of bothphylogeny and ecology to behavior. Historically, comparative psychologistsfocused on artificial selection (Plomin et al 1990) and ethologists focused ritualization--the evolutionary process in which motor patterns become spe-cialized for communication (Eibl-Eibesfeldt 1975). Both were concerned withgeneral principles of development and the nature of critical periods (e.g. Im-melman et al 1981).

GENETICS AND DEVELOPMENT Much of the work on behavior genetics contin-ues to focus on simple causal models, often statistical in nature. For example,Wheeler et al (1991) were able to change the rate of wing vibrations in-Drosophila by the exchange of a single gene. An important window on therelation of genes and selection pressures has been provided by the use of DNAfingerprinting to link members of a population (Everitt et al 1991). DNA-basedinvestigations of bees have shown that hormonal control of tasks differs withinsubpopulations in a hive (Robinson et al 1989). DNA-based investigations birds have contradicted our assumption that many species are exclusivelymonogamous (Weatherhead & Montgomerie 1991).

Changes in behavior and development increasingly are treated as importantcontributors to evolution (Bateson 1988). Because development stands be-tween genes and their expression, it can be selected for as a critical mediator ofevolution. Arnold (1990) noted that development is a dimension of the pheno-type and genotype instead of an alternative to direct inheritance. King & West(1990) proposed the concept of inherited niches to account for the effects differences in ecological pressures on song development in cowbird sub-populations. A similar explanation may underlie the demonstration of Gold-thwaite et al (1990) that ground squirrels do not show specialized defensivereactions to snakes in arctic populations that are now free from snakes.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


692 TIMBERLAKE

Gottlieb (1992) argued that because of the large amount of "silent" DNA all phyla, persistent changes in phenotype can occur in response to changes inthe environment, with no change in genotype. For the more statistically ori-ented, allometric functions can be plotted with individuals (or individuals atdifferent times) as the points, to look for similarities and differences in func-tional relations across development (Gittleman 1989a). Finally, as indicatedpreviously, many processes of development have proved remarkably accessi-ble to experimental investigation (Miller 1988).

BEHAVIORAL STRATEGIES WITHIN SPECIES Tinbergen and his students pro-vided evidence that populations show ranges of behavioral strategies associatedwith different reproductive success [e.g. Patterson (1965) on group vs. isolatenesters in gulls]. Work has expanded exponentially on within-species differ-ences in life history strategies--such as body size at maturation, courtshipmethod, mate choice, and parental care. Lessells (1991) reviewed a considerablebody of literature relating life-history strategies to reproductive success andevolution. These strategies are often modified by different environmental con-ditions, and are even affected by the relative distributions of the strategies ofother animals in the population. The notion of evolutionarily stable strategies(Maynard-Smith 1982) and the ideal free distribution approach (Milinski Parker 1991) are designed to capture the conditional, frequency-based nature ofappropriate behavior.

MICRO-MODEL SYSTEMS The division between protoevolutionary and micro-evolutionary comparisons is not always clear, particularly in the areas of generalprinciples and model systems. The assumption in both types of comparison isthat careful analysis of the behavior of a small number of subjects will revealinformation of general applicability. One distinction is that the microevolution-ary approach focuses on comparisons within the same or closely related species.For example, Ketterson & Nolan (1992) and Marler & Moore (1991) hormones to engineer new phenotypes in free-living male birds and lizards.These phenotypes were used to illustrate the effects of behavioral variation onfitness, and to infer general evolutionary trajectories within a population.

Integration

Explicit comparisons provide a powerful technique for disentangling the deter-minants of behavior, but it should be apparent that these four general catego-ries of comparison are not fixed. They are intended to provide a frameworkacknowledging diversity while encouraging a more coherent approach. Theremay be considerable advantage in studying the same topic across all fourcategories. For instance, behaviors related to kin selection can be dealt with asa protoevolutionary trend, as a homologue, as examples of ecological conver-gence, and as the product of the genetic makeup of a particular population.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 693

Similarly, the same empirical technique (e.g. allometric scaling, profiles, ideal free distribution) can be used in comparisons of any sort.

Should one of these types of comparison be preferred? In a recent paper Hailman (1988) argued that an adequate comparison requires including multi- ple species in all four categories of a table similar to the present one. Homol- ogy is inferred only if related animals in different niches show it and unrelated animals in the same niches do not. Convergence is indicated only if unrelated animals show similarities that their relatives in different circumstances do not. This is a worthy approach, but it focuses only on establishing homologues and convergence and the price is high in terms of subjects required.

Careful work in any single category provides useful data. Robinson (1991) cited several examples of how the painstaking accumulation of data from unrelated insects eventually led to insights concerning adaptation and phylogeny (Eberhard 1980; Robinson 1985). However, using several types of comparison probably will advance knowledge more rapidly than focusing on a single type. Phylogenetic or ecologically-based explanations often overlook simple proximal causes (Barlow 1989). On the other side, an exclusive focus on proximal determinants can produce general principles unconnected with function and evolution. For example, the general principles of learning established by laboratory psychologists are not very helpful in explaining or predicting naturally occurring phenomena such as song learning in birds (Timberlake & Lucas 1989). Finally, an exclusively ecological comparison may ignore the importance of general processes and exaptations (Riley & Langley 1993).

One effective approach to comparison would be to begin with protoevolutionary observation and analyses of the regularities and functional relations among the stimuli, responses, and states involved, move on to ecological and phylogenetic comparisons, and cross to microevolutionary analysis of mechanism and development. For an example of such a shift, consider that the study of learning began with the protoevolutionary approach of trends and grades, moved through general principles and model systems, into ecological comparisons (Balda & Kamil 1989) and the beginning of microevolutionary considerations (Arnold 1981; Bateson 1988).

It is assuredly not necessary that all research in animal behavior be explicitly comparative. For example, the sequence of research above is appropriate for any study of animal behavior whether comparative or not. In terms of Tinbergen’s questions, this research example begins with an interest in mechanism, incorporates function and evolution, and returns to mechanism at a more profound level. However, this particular sequence of research types is not critical in advancing the field. What is important is the development of a picture of the animal and environment that helps integrate the results of different types of research.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


694 TIMBERLAKE

THE DYNAMICS AND STRUCTURE OF BEHAVIOR

Motivation

At the time of the initial synthesis, the study of motivation was a critical substrate of both comparative psychology and ethology. Motivation fell out of favor with psychologists because neither deficit motivation nor incentive effects could be tied firmly to physiology, and the relations between deprivation manipulations and behavior differed with both the type of manipulation and the measure of behavior used. Perhaps most importantly, motivation was considered superfluous because, given a sentient organism with a few reflexes, researchers felt able to construct the form and dynamics of new behavior by employing operant and Pavlovian conditioning techniques. Motivation per- sisted only as a broad causal principle related to general arousal and attraction.

Ethologists from Tinbergen (195 1) through Baerends & Drent (1970) began with a more complex view of motivation that combined regulation with hierarchical structures of states, releasers, and action patterns spread across levels of organization. However, despite its central position in ethological thinking, motivation declined in popularity for reasons similar to those in comparative psychology. Drive was not unitary (Hinde 1966), and the concept of action-specific energy failed to account for many examples of the initiation and cessation of behavior (e.g. Dethier 1976). Motivation became an unneces- sary ghost in the machine.

To the surprise of many, motivation seems to be making a comeback. First Toates (1986) and then Colgan (1989) wrote small primers. The former sur- veyed motivational systems, the latter considered three basic research topics. The first topic was the motivational systems developed by ethologists, which they inferred from the timing, sequencing, and organization of behavior. The second topic was regulatory physiological systems for which control circuitry is inferred from lesions, stimulation, and measurement of hormonal levels and metabolic indicators (e.g. Stricker 1990; see also Mrosovsky 1990). The third topic was the adaptation and regulation of behavior from the viewpoint of optimality and game theory (e.g. Krebs & Davies 1991).

Three more research topics might have been added: one focusing on social contexts of motivation, particularly the developmental, ecological, and strate- gic aspects, including kin selection (Cheney & Seyfarth 1990; Slobodchikoff 1988); another focusing on the generation of response components and stimulus processing rules for particular states (Gallistel 1990); a final chapter could have dealt with recent work on computer simulations of animals. Imaginary animals can be taught to categorize inputs, learn sequential dependencies, and filter noise for signal (Grossberg 1988). Mechanical “insects” wander through their environments (Beer 1990), and complete worlds of computerized repro- ducing animals can be turned loose to evolve in sometimes wildly disparate ways (Langton et a1 1991).


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 695

Learning

The study of learning very early became a protoevolutionary endeavor, one tied intimately to apparatus, procedures, and general principles. Within the last 20 years, a part of this massive literature (e.g. Spear et a1 1990) has moved slowly in the direction of a functional ecological approach. Bolles & Beecher (1988), Gallistel (1990), Gould (1986), Kamil & Roitblat (1983, Kamil et a1 (1987), and Zentall & Galef (1988) deal with the role of learning in solving ecological problems in both restricted and relatively unrestricted circumstances. Other researchers have focused on laboratory versions of ecological learning problems, ranging from examples of perceptual learning (Suboski 1989), to the learning of the time and location of food availability by garden warblers (Biebach et a1 1989), and how gouramis defend territories (Hollis 1990).

One of the most fascinating continuing research stories concerns memory for stored foods in the bird families of corvids and parids. An interesting recent outcome is that field differences in the memory of caching and noncaching corvids show up in tests of memory in an analogue of the radial arm maze (Balda & Kamil 1989), and also in even more constrained laboratory tests (Olson 1991; see Shettleworth 1990). Species that depend heavily on memory to retrieve caches in the field do better on even arbitrary laboratory tests of memory. Similar differences have been obtained for storing and nonstoring parids (Krebs et a1 1990). Such evidence raises the possibility of using related species in studies of ecology and phylogeny (Kamil 1988). Similar opportunities appear to be developing in the study of bird song (Kroodsma & Konishi 1991; Nottebohm 1991).

Regulatory and Structural Behavior Systems

As the study of both motivation and learning moves toward ecological and microevolutionary comparisons, it will be necessary to provide an animal- based rather than a procedure- or apparatus-based framework. Such a framework may serve as a basis for integrating other research as well. Behavioral ecologists working on the determinants of foraging have begun to confront effectively the ways in which animals distribute their energy resources across feeding alternatives and the demands of different systems (McNamara & Houston 1990). Caraco & Lima (1987) and Lucas & Walter (1991) have examined the influence of metabolic balance on sensitivity to risk. Laboratory investigations have become concerned with the regulation of feeding by non- metabolic influences such as local cost and time windows (Collier & Johnson 1990; Cuthill et a1 1990; Plowright & Shettleworth 1991). Other researchers have pressed for the consideration of the circadian and ultradian rhythmicity of behavior (Brady 1988; Silver 1990).

An approach compatible with much of this work is the concept of behavior systems deriving in part from traditional ethology (Davey 1989; Davis 1984;


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


696 TIMBERLAKE

Fanselow & Lester 1988; Heiligenberg 1991; Timberlake & Lucas 1989). This approach combines hierarchical motivational systems with a dimension of appetitive to consummatory motivational modes related to the physical and temporal proximity of incentives. Learning occurs in terms of integrating sensory-motor control circuitry, tying it to motivational states, and relating these elements to the environment. Such a systems approach encourages analysis of the levels of regulation (e.g. Cools 1985; Fentress 1991), consideration of interactions among different motivational states: for example, fear and thermoregulation in chicks (Rovee-Collier et a1 1991), and comparison of regulation and learning in sexual systems (e.g. Domjan & Hollis 1988, Everitt 1990), and social systems relating to feeding in rats (Galef 1990, Timberlake 1983).

A behavior systems approach also makes ready contact with development (e.g. Hall 1990; Hogan 1988; Hogan et a1 1991), and with the neurophysiological underpinnings of particular behaviors-for example, the reproductive behavior of reptiles and mammals (Bronson 1989; Crews 1988), and fear and aggression of rats in semi-natural environments (Blanchard & Blanchard 1990; see also Brain et a1 1990). Fanselow (1989) and Fanselow & Lester (1988) have shown that avoidance and escape behavior is controlled by a sensory dimension of predatory imminence, associated with varying response probabilities, stimulus control, and brain structures. Lammers et a1 (1988a,b) have distinguished specific areas of the hypothalamus of rats related to social grooming, attack, teeth-chattering, flight, and escape jumps.

In short, amplifying the concept of behavior systems provides many opportunities for producing a more coherent and heuristic picture of the fit between animals and their envrionments. A system of behavior can provide a framework for integrating the answers to Tinbergen’s questions and the results of different methods of comparing behavior. Finally, a systems approach can provide an organization for dealing with issues of animal cognition and animal welfare.

COGNITIVE ABILITIES

Even for normally reserved scientists, it is fascinating and delightful that chimpanzees sign (Gardner et a1 1989), vervet monkeys warn of specific predators (Cheney & Seyfarth 1990), starlings mimic and rearrange human sounds and music (West & King 1990), rats and gray parrots count (Davis & Perusse 1988; Pepperberg 1990, and pigeons distinguish between slides of cats and chairs (Wasserman et a1 1988).

Compared with the enthusiastic rebirth of interest in animal cognition, the behaviorist period of strict avoidance of the attribution of mental life appears at best a long fallow period (Wasserman 1993), and at worst an anomalous dark age that has finally been set right by the cognitive revolution. Neverthe- less, behaviorism made a critical contribution to the study of animal cognition


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 697

because it compelled scientists to try to ignore the projection of their ownpsychology onto animals long enough to begin to discern determinants ofbehavior in clearly defined external variables. The development of both com-parative psychology and ethology required the rejection of poorly definedcausal agents--mentalism and instincts in psychology, vitalism in biology--toallow science to emerge.

Comparative Cognition

Comparative cognition is the domain of scientists interested in how differentspecies process and integrate stimuli (Boysen & Capaldi 1992; Honig Fetterman 1992; Kesner & Olton 1990; Ristau 1991; Roitblat 1987). However,the field is split along much the same lines as traditional learning, into pro-toevolutionary approaches (anthropocentric) and ecological approaches(Shettleworth 1993). The anthropocentric stance is fueled partly by interest the classic continuum of mental life (Wasserman 1993) and partly by thesuccess of traditional research on universal laws of learning. Recent work onrepresentation, counting, remembering, and categorization has been summa-rized in Gallistel (1989, 1990), Wasserman (1993), and Honig & Fetterman(1992). The processing of auditory stimuli by birds and mammals has alsoreceived attention (Dooling & Hulse 1989).

Much of the work relating nonhuman and human language and intelligencefalls largely in the protoevolutionary camp (e.g. Gardner et al 1989 and Parker& Gibson 1990 on primates; Herman et al 1990 on dolphins). This work hasprofited from analytic efforts such as those pioneered by Premack (1983) break down cognitive phenomena into components that can be tested sepa-rately, an approach followed by Washburn & Rumbaugh (1991) in studyingcounting in monkeys and by Gisiner & Schusterman (1992) in studying cate-gorization in sea lions.

As for the ecological approach to cognition it is basically theromorphic,animai centered. The work of yon Uexkull (1934) on the functional fit be-tween the filtering of the sensory surround and behavior in the woodtick servesas a classic precursor. Gallistel (1990) recently reviewed studies of the ecolog-ical relevance of abilities such as memory, timing, and navigation. Yoerg(1993) also dealt generally with the relation of ecology and learning phenom-ena. The work on food storing in birds alluded to in the learning sectionprovides examples of the ecological approach (see also Kallander & Smith1990), as does the work of Cheney & Seyfarth (1990) on vervet monkeys,work on the critical evaluation of foraging "information centers" in socialspecies (e.g. Richner & Marclay 1991; Zentall & Galef 1988), and Real’s(199 I) article on choice behavior and the evolution of cognitive "architecture"in bumblebees.

I believe cognitive research will be better served in the long run by movingin the same direction as other research in animal behavior, namely towardecological, phylogenetic, and microevolutionary concerns. In maintaining an


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


698 TIMBERLAKE

anthropocentric approach that is not clearly evolutionary, scientists tend toignore the ecological and evolutionary basis of the phenomena they study,focusing instead on general principles and concerns of definition and experi-mental procedure. The result can be an unforseen limit on research proceduresand interpretation. For example, the eight-arm radial maze has been usedextensively to study memory, based on the presumption that rats were re-warded for efficient search by finding the maximum number of pellets in eightchoices. However, Timberlake & White (1990) showed that rats efficientlysearch maze arms in the absence of any food at all, presumably based onevolved mechanisms facilitating foraging efficiency and increasing environ-mental familiarity in the absence of food.

Mental Life and Consciousness

For many scientists, the return to a concern with animal consciousness (e.g.Bateson & Klopfer 1991; Griffin 1981) is a return to a problematic approach toanimals that it took centuries to escape (Burghardt 1985). Given there is littleconvincing data that the introspectively revealed contents of human conscious-ness play a primary causal role in most human behavior, how do we explainthe fascination with animal consciousness? Beer (1992) has suggested that ourinterest in mental life is based on our "folk psychology," basically a "toolkitfor coping with the cognitive and conative demands and tensions of humanlife." In other words, our interest in and ideas about consciousness reflect ahuman strategy for interacting with the world. Because of the questionablestatus of folk psychology and its intimate ties to human language, it is difficultto defend pouting large resources into a frontal attack on animal conscious-ness. The direct approach has failed badly in the past, and has yet to pay largedividends even in analyzing human behavior, despite high current interest andfirsthand access to data.

There is more support for exploring mental states such as intentions (e.g.Dennett 1987). However, care must be taken. The use of such intentional termsas "Machiavellian" or "’deceptive" to describe the behavior of nonhuman ani-mals often has been more successful in stirring controversy than in specifyingthe determinants of behavior. Most scientists agree that mental terms must betreated experimentally as intervening variables for which we need to provideconvergent behavioral, physiological, and comparative analysis.

Given the number of people concerned with mental states, some progresswould be expected from an experimental approach analyzing components ofprocessing and motivation. For example, Martin et al (1991) showed a separa-tion in rats between determinants of discriminative and affective effects ofopiates. The work of Povinelli et al (1991) suggests that rhesus monkeys arenot able to infer the knowledge states of humans, though in similar circum-stances chimpanzees apparently can (see also Whiten 1991).

In a way, the notion of deception is not necessarily more amorphous thanthe concept of, say, timing. Both involve stimulus processing and behavioral


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 699

results, The difference is that the processes involved in timing are betteranalyzed and the concept has only a little explanatory value and interest byitself. In contrast, the concept of deception has compelling popular explanatorypower for behavior, even though the concept is still clearly in transition. Forexample, Gyger & Marler (1988) inferred that deception in food calling is usedby male domestic fowl to attract hens, but Moffatt & Hogan (1992) argued thathens so rapidly track the relation of calls to food that deception would not beeffective. Research into both deception and timing would be served well byconceptualizing them within a functional system, and analyzing the compo-nent processes that comprise them (e.g. Adams & Caldwell 1990).

Animal Welfare and Rights

The historical inability of scientists to agree on how to study the mental life ofnonhuman animals and its continuity with the mental life of humans, has comeback to challenge the field of animal behavior. Freely ascribing mental statesto other animals piques general interest, but it also allows activist philosophersto freely claim the existence of suffering in nonhuman animals without theconvergent evidence and careful analysis that would be required for humans(e.g. Regan 1983; Singer 1990). Bekoff & Jamieson (1991) argue that becausethe criticisms of animal research raised by these philosophers are similardespite their differing "schools," the criticisms must point to a central truth.But an obvious alternative is that these activist philosophers began with theirconclusions in place and reasoned backwards to produce a fit with their firstprinciples.

A large number of scientists have become engaged in the laudable attemptto bring together concepts of animal welfare and research (e.g. Dawkins 1990;see Carlstead et al 1991 for application of a systems approach to the reductionof stereotyped behaviors). There are now many journals focused on animalwelfare and conservation issues (e.g. Anthropozoos, Applied Animal BehaviorScience, Zoobiology) while others (e.g. American Journal of Primatology)devote considerable space to and publish special issues on these topics.

Less fortunately, oversight committees and bureaucracies are increasing atevery level, all with rules that can take on a life of their own independent of thegood intentions with which they were created. For example, the expense ofstainless steel caging and hospital-like stainless steel covering on all surfacesin animal housing rooms has unaccountably become a rcquirement of goodcare. The increased concern with animal welfare has in some ways improvedthe state of animals in science, but in too many cases it is still difficult todetermine whether the importance and size of improvments outweighs the costto research. The issues are many and complex (Novak & Petto 1991).

Equally problematic for research is the inability of scientists to communi-cate with the public as effectively as do animal rights groups. The knowledgeof scientists ought to be more influential with the environmental andbiodiversity movements and pet lovers than the anti-science stance of many


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


700 TIMBERLAKE

animal fights groups. Yet too often important and well-intentioned remarks ofvisible scientists, such as Jane Goodall, fail to emphasize that our increasedknowledge about animals and their cognitive processes comes from researchnot from intuition. Intuition alone usually produces contradictory and poorlyfounded beliefs.

In a recent radio interview, a leader of the largest animal fights organizationin the United States stated with great feeling that her goal was to let everyanimal run free through meadows in the sunlight. Her sentiment was compel-ling but she utterly misunderstood a critical point. Given alternatives of coveror darkness most animals would prefer not to run through sunlit meadows, noteven mammals, the small cladistic group she probably was thinking of [40%of mammals are rodents, and another 20% are bats (Vaughan 1978)]. cannot allow such projections to represent nonhuman animals to the public orto governing officials. It is demeaning and even dangerous to us all.

TOWARD CONTINUING SYNTHESIS

A primary path of science is toward specialization, inevitably accompanied byfragmentation. Specialization is a sign of progress, but it is also both a problemand an opportunity for a synthetic field like animal behavior. The key toprogress in such a field is continued synthesis. Wilson (1975) was basicallyfight in his vision that sociobiology, behavioral ecology, and integrative neu-robiology would combine with the classic study of behavior. What he failed tosee was that behavior and its fit with the environment is the basis for thisintegration.

In any continuing synthesis Tinbergen’s four questions are useful in en-couraging balance in the ways phenomena are considered. The different meth-ods of comparison reviewed above facilitate relating phenomena to trends,scales, general laws, and phylogenetic, ecological, and microevolutionaryanalyses. What ultimately defines a continuing synthesis, though, is not bal-ance or methods, but the parsing and integration of the resultant knowledge bymeans of a flexible and heuristic picture of the fit between animal and environ-ment. A picture I find attractive stems from modifying the basic behaviorsystems approach of the ethologists to emphasize greater roles for learning anddevelopment, stimulus processing and integration, response organization andcoordination, and the co-regulation of hierarchical and interacting motiva-tional states.

This view also treats behavior (including perception) as a bidirectional linkbetween animal and environment in both local and ultimate senses. Locally,behavior creates environments as much as local environments create behavior.This point is particularly salient in highly social species. In an ultimate sense,behavior is the ambassador of environments to the genes as well as the repre-sentation of the genes in environments. Filling in this picture will requirecontinuing research that combines elements of laboratory and field by provid-


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.

ANIMAL BEHAVIOR 701

ing sufficient stimulus support to engage processing mechanisms, responsecomponents, and motivational states relevant to the functional behavior systemunder study, but in a way that allows manipulation and measurement of thedeterminants of the fit between animal and environment.

Finally, continuing synthesis demands hard-nosed, innovative collaborationbetween disciplines, an ideal that faces important realities of finance anddefended territories. J. P. Scott (1973:34) addressed both the promise and theproblems of such research:

Anyone who works on the new frontiers of science finds that the conventionaldisciplinary boundaries disappear, and become important only when we con-sider university organization and finance ... scientific progress is broughtabout through cooperation between involved individuals ... and segregationbased on separation of disciplines may be just as harmful as that based on race.

ACKNOWLEDGMENTS

Preparation of this review was supported by NIMH Grant 37892 and NSFGrant IBN-9121647. I thank Ellen Ketterson, Meredith West, Gary Lucas, andLeslie Real for comments, and Holly Stocking for patience.

Literature Cited

Adams, E. S., Caldwell, R. L. 1990. Deceptivecommunication in asymmetric fights of thestomatopod crustacean Gonodactylusbredini. Anita. Behav. 39:706-16

Adkins-Regan, E. 1990. Is the Snark still aBoojum? The comparative approach to re-productive behavior. Neurosci. Biobehav.Rev. 14:243-52

Alexander, R. M. 1982. Optima for Animals.London: Edward Arnold

Alkon, D. L., Disterhoft, J., Coulter, D. 1987.Conditioning-specific modification of post-synaptic membrane currents in mollusc andmammal. In The Neural and MolecularBases of Learning, ed. J. P. Changeux, M.Konishi, pp. 205-38. New York: Wiley

Alter,W. A. III, Hartgraves, S. H., Wayner, M.J. 1991. A review of animal to human ex-trapolation: issues and opportunities. Phys-iol. & Behav. 5:1-184

Andrew, R. J. 1991. Neural and BehaviouralPlasticity: The Use of the Domestic Chickas a Model. Oxford: Oxford Univ. Press

Aronson, L. R. 1984. Levels of integration andorganization: a reevaluation of the evo-lutionary scale. In Behavioral Evolutionand Integrative Levels, ed. G. Greenberg,E. Tobach, pp. 57-81. Hillsdale NJ:Erlbaum

Arnold, S. J. 1981. The microevolution offeeding behavior. In Foraging Behavior:Ecological, Ethological; and Psychologi-cal Approaches, ed. A. C. Kamil, T.D. Sargent, pp. 409-53. New York: Gar-land

Arnold, S. T. 1990. Inheritance and the evolu-tion of behavioral ontogenies. In Develop-mental Behavior Genetics: Neural,Biomedical, and Evolutionary Ap-proaches, ed. M. E. Hahn, J. K. Hewitt, N.D. Henderson, R. Benno, pp. 167-89. NewYork: Oxford Univ. Press

Atz, J. 1970. The application of the idea ofhomology to behavior. In Developmentand Evolution of Behavior, ed. L. R. Aron-son, E. Tobach, D. S. Lehrman, J. S.Rosenblatt, pp. 53-74. San Francisco:Fl’eelnan

Baerends, G. P., Drent, R. H. 1970. The her-ring gull’s egg. Behaviour 17(Suppl.):l-416

Balda, R. P., Kamil, A. C. 1989. A compara-tive study of cache recovery in three corvidspecies. Anim. Behav. 38:486-95

Barlow, G. W. 1989. Has sociobiology killedethology? See Bateson & Klopfer 1989, pp.1-45

Bateson, P. P. G. 1988. The active role of be-haviour in evolution. In Evolutionary Pro-cesses and Metaphors, ed. M. W. Ho, S.W. Fox, pp. 191-207. London: John Wiley& Sons

Bateson, P. P. G. 1991. The Development andIntegration of Behaviour: Essays in Hon-our of Robert Hinde. Cambridge: Cam-bridge Univ. Press

Bateson, P. P. G, Klopfer, P. H. 1989. Per-spectives in Ethology. Vol. 8. WhitherEthology? New York: Plenum

Bateson, P. P. G., Klopfer, P. H. 1991. Per-


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.

702 TIMBERLAKE

spectives in Ethology Vol. 9. Human Un-derstanding and Animal Awareness. NewYork: Plenum

Becker, J. B., Breedlove, S. M., Crews, D.1992. Behavioral Endocrinology. Cam-bridge, MA: MIT Press

Beecher, M. D. 1990. The evolution of parent-offspring recognition in swallows. SeeDewsbury 1990, pp. 360-80

Beer, C. G. 1980. Perspectives on animal be-havior comparisons. In Comparative Meth-ods in Psychology, ed. M. H. Bomstein,pp. 17~64. Hillsdale, NJ: Erlbanm

Beer, C. G. 1992. Conceptual issues in cogni-tive ethology. Adv. Stud. Behav. 21:6%110

Beer, R. D. 1990. Intelligence as Adaptive Be-havior: An Experiment in ComputationalNeuroethology. New York: Academic

Bekoff, M., Jamieson, D. 1991. Reflectiveethology, applied philosophy, and themoral status of animals. See Bateson &Klopfer 1991, pp. 1-47

Bell, G. 1989. A comparative method. Am.Nat. 133:553-71

Bell, W. J. 1991. Searching Behaviour: TheBehavioural Ecology of Finding Re-sources. London: Chapman Hall

Beltrami, E. 1987. Mathematics for DynamicModeling. New York: Academic

Biebach, H., Gordijn, M., Krebs, J. R. 1989.Time-and-place learning by garden war-blers, Sylvia borin. Anita. Behav. 37:353-

6OBitterman, M. E., 1965. Phyletic differences in

learning. Am. Psychol. 20:39(~410Bitterman, M. E., Couvillon, P. A. 1991. Fail-

ures to find evidence of adaptive special-ization in the learning of honey bees. InThe Behavior and Physiology of Bees ed.L. J. Goodman, R. C. Fisher, pp. 288-305.Wallingford UK: CAB International

Blanchard, D. C., Blanchard, R. J. 1990. Thecolony model of aggression and defense.See Dewsbury 1990, pp. 410-30

Blanchard, R. J., Brain, P. F., Blanchard, D. C.,Parmigiani, S., eds. 1989. Ethoexperimen-tal Approaches to the Study of Behavior.Dordrecht: Kluwer Academic

Blass, E. M., ed. 1986. Handbook of Behav-ioral Neurobiology Vol. 8. DevelopmentalPsychobiology and Developmental Neuro-biology. New York: Plenum

Blass, E. M., ed. 1988. Developmental Psycho-biology and Behavioral Ecology. Hand-book of Behavioral Neurobiology, Vol. 9.New York: Plenum

Bolles, R. C. 1988. Nativism, naturalism, andniches. See Bolles & Beecher 1988, pp.1-15

Bolles, R. C., Beecher, M. D., ed. 1988. Evolu-tion and Learning. Hillsdale, N J: Erlbaum

Boysen, S. T., Capaldi, E. J. 1992. The Devel-opment of Numerical Competence: Animaland Human Models. Hillsdale, N J:Erlbaum

Brady, J. 1988. The circadian organization ofbehavior: time-keeping in the Tsetse fly, a

model system. Adv. Stud. Behav. 18:153-91

Brain, P. F., Parmigiani, S., Blanchard, R. J.,Mainardi, D., eds. 1990. Fear and De-fence. London: Harwood Academic

Bronson, F. H. 1989. Mammalian Reproduc-tive Biology. Chicago: Univ. Chicago Press

Brooks, D. R., McLennan, D. A. 1991. Phylog-eny, Ecology, and Behavior. Chicago:Univ. Chicago Press

Brown, C. H. 1989. The acoustic ecology ofEastAfrican primates and the perception ofvocal signals by grey-checked mangabeysand blue monkeys. See Dooling & Hulse1989, pp. 210-39

Burghardt, G. 1985. Animal awareness: cur-rent perceptions and historical perspective.Am. Psychol. 40:905-19

Burghardt, G. M. 1991. Cognitive ethologyand critical anthropomorphism: a snakewith two heads and hog-nose snakes thatplay dead. See Ristau 1991, pp. 53-90

Caraco, T., Lima, S, L. 1987. Survival, energybudgets, and foraging risk. In QuantitativeAnalyses of Behavior, ed. M. L. Commons,A. Kacelnik, S. J. Shettleworth, 6:1-20

Carlstead, K., Seidensticker, J., Baldwin, R.1991. Environmental enrichment for zoobears. Zoo-Biology 10:3-16

Cheney, D. L., Seyfarth, R. M. 1990. HowMonkeys See the World: Inside the Mind ofAnother Species. Chicago: Univ. ChicagoPress

Clutton-Brock, T. H., Harvey, P, H. 1984,Comparative approaches to investigatingadaptations. See Krebs & Davies 1984, pp.7-29

Clutton-Brock, T., Godfray, C. 1991. Parentalinvestment. See Krebs & Davies 1991, pp.234-62

Colgan, P. 1989. Animal Motivation. London:Chapman & Hall

Collier, G., Johnson, D. F. 1990. The time win-dow of feeding. Physiol. Behav. 48:771-77

Cools, A. R. 1985. Brain and behavior: hierar-chy of feedback systems and control ofinput. In Perspectives in Ethology, Vol. 6.Mechanisms, ed. P. P. G. Bateson, P.H. Klopfer, pp. 109-68. New York: Ple-num

Crews, D., ed. 1988. Psychobiology of Repro-ductive Behavior. Englewood Cliffs, NJ:Prentice-Hall

Cullen, E. 1957. Adaptations in the kittiwaketo cliff nesting. Ibis 9:275-302

Cuthill, 1. C., Kacelnik, A., Krebs, J. R.,Haccou, P., et al. 1990. Starlings exploitingpatches: the effect of recent experience onforaging decisions. Anim. Behav. 40:625-40

Daly, M., Wilson, M. 1988a. Homocide. Haw-thorne, NY: Aldine de Gruyter

Daly, M., Wilson, M. 1988b. Psychology ofdiscriminative parental solicitude. SeeLeger 1988, pp. 51-144

Darwin, C. 1871. The Descent of Man andSelection in Relation to Sex. London: John


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 703

MurrayDavey, G. 1989. Ecological Learning Theory.

New York: RoutledgeDavies, N. B. 1991. Mating systems. See

Krebs & Davies 1991, pp. 263-94Davies, N. B., Houston, A. I. 1984. Territory

economics. See Krebs & Davies 1984, pp.148-69

Davis, H., Perusse, R. 1988. Numerical com-petence in animals: definitional issues, cur-rent evidence, and a new research agenda.Behav. Brain Sci. 11:561-615

Davis, W. J. 1984. Motivation and learning:Neurophysiological mechanisms in a"model" system. Learn. Motiv., 15, 377-93

Dawkins, M. S. 1989. The future of ethology,how many legs are we standing on? SeeBateson & Klopfer 1989, pp. 47-54

Dawkins, M. S. 1990. From an animal’s pointof view: motivation, fitness and animalwelfare. Behav. Brain Sci. 13:1~51

Dawkins, M. S., Halliday, T. R., Dawkins, R.,eds. 1991. The Tinbergen Legacy. London:Chapman Hall

Demarest, J. 1981. Comparative PsychologyNewsletter. W. Long Branch, NJ: Mon-mouth Coll.

Dennett, D. C. 1987. The Intentional Stance.Cambridge, MA: MIT Press

Dennis, W. 1941. Spalding’s experiment on theflight of birds repeated with another spe-cies. J. Comp. Psychol. 31:337-48

Dethier, V. G. 1976. The Hungry Fly: A Physi-ological Study of the Behavior Associatedwith Feeding. Cambridge, MA: HarvardUniv. Press

Dewsbury, D. A. 1985. Studying Animal Behavior. Chicago: Univ. Chicago Press

Dewsbury, D. A. 1988. The comparative psy-chology of monogamy. See Leger 1988,pp. 1-50

Dewsbury, D. A. 1989a. Comparative psychol-ogy, ethology, and animal behavior. Annu.Rev. Psychol. 40:581---602

Dewsbury, D. A. 1989b. Frank AmbroseBeach 1911-1988. Am. J. Psychol.102:414-20

Dewsbury, D. A. 1990. Contemporary Issuesin Comparative Psychology. Sunderland,MA: Sinauer

Dickinson, A. 1980. Contemporary AnimalLearning Theory. Cambridge: CambridgeUniv. Press

Domjan, M., Hollis, K. L. 1988. Reproductivebehavior: a potential model system foradaptive specializations in learning. SeeBolles & Beecher 1988, pp. 213-37

Dooling, R. J., Hulse, S. H. 1989. The Com-parative Psychology of Audition. Hillsdale,NJ: Erlhaum

Driver, P. M., Humphries, D. A. 1988. ProteanBehavior: The Biology of Unpredictability.Oxford: Clarendon Press

Dukas, R., Real, L. A. 1991. Learning foragingtasks by bees: A comparison between so-cial and solitary species. Anim. Behav. 42:269-76

Dumas, C., Dore, F. Y. 1991 Cognitive devel-opment in kittens (Felis catus ): an obser-vational study of object permanence andsensorimotor intelligence. J. Comp. Psy-chol. 105:357-65

Eberhard, W. G. 1980. Horned beetles. Sci.Am. 242:166-82

Eberhard, W. G. 1982. Behavioral characteris-tics for the higher classification of orb-weaving spiders. Evolution 36:1067-95

Eibl-Eibesfeldt, I. 1975. Ethology: the Biologyof Behavior. New York: McGraw-Hill

Eldredge, N. 1989. Macroevolutionary Dy-namics: Species, Niches, and AdaptivePeaks. New York: McGraw-Hill

Everitt, B. J. 1991. Sexual motivation: a neuraland behavioural analysis of the mecha-nisms underlying appetitive and copulatoryresponses of male rats. Neurosci.Biobehav. Rev. 14:217-32

Everitt, J., Hurst, J. L., Ashworth D., Bamard,C. J. 1991. Aggressive behaviour amongwild-caught house mice, Mus domesticusRutty, correlates with a measure of geneticsimilarity using DNA fingerprinting. Anim.Behav. 42:313-16

Fanselow, M. S. 1989. The adaptive functionof conditioned defensive behavior: an eco-logical approach to Pavlovian stimulus-substitution theory. See Blanchard et al1989, pp. 151-66

Fanselow, M. S., Lester, L. S. 1988. A func-tional behavioristic approach to aversivelymotivated behavior: predatory imminenceas a determinant of the topography of de-fensive behavior. See Bolles & Beecher1988, pp. 185-212

Fentress, J. C. 1991. Analytical ethology andsynthetic neuroscience. See Bateson 1991,pp. 77-120

Fisher, J. W. 1991. Disambiguating anthropo-morphism: an interdisciplinary review. SeeBateson & Klopfer 1991, pp. 49-85

Gagnon, S., Dore, F. Y. 1992. Search behaviorin various breeds of adult dogs (Canisfamiliaris): object permanence and olfac-tory cues. J. Comp. PsychoL 106:58-68

Galef, B. G. 1990. An adaptationist perspectiveon social learning, social feeding, and so-cial foraging in Norway rats. See Dews-bury 1990, pp. 55-79

Gallistel, C. R. 1989. Animal cognition. Annu.Rev. PsychoL 40:155-89

Gallistel, C. R. 1990. The Organization ofLearning. Cambridge, MA: MIT Press

Garcia, J., Brett, L. P., Rusiniak, K. W. 1989.Limits of Darwinian conditioning. In Con-temporary Learning Theories: Instrumen-tal Conditioning Theory and the Impact ofBiological Constraints on Learning, ed. S.B. Klein & R. R. Mowrer. Hillsdale N J:Erlbaum

Gardner, R. A., Gardner, B. T., Van Cantfort.T. E. 1989. Teaching Sign Language toChimpanzees. Albany, NY: SUNY Press

Gaulin, S. J., FitzGerald, R. W., Wartell, M. S.1990. Sex differences in spatial ability and


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


704 TIMBERLAKE

activity in two vole species (Microtus oc-hrogaster and M. pennsylvanicus). J.Comp. PsychoL 104:88-93

Ganlin, S. J., Wartell, M. S. 1990. Effects ofexperience and motivation on symmetrical-maze performance in the prairie vole(Microtus ochrogaster). J. Comp. Psychol.104:183-89

Gibbons, E. G. Jr., Wyers, E. J., Waters, E.,Manzel, E. W. Jr. 1992. Naturalistic Envi-ronments in Captivity for Animal BehaviorResearch. Albany, NY: SUNY Press

Gisiner, R., Schusterman, R. J. 1992. Se-quence, syntax, and semantics: responsesof a language-trained sea lion (Zalophyscalifornianus ) to novel sign combinations.J. Comp. P~,chol. 106:78-91

Gittleman, J. L., ed. 1989a. Carnivore Behav-ior, Ecology, and Evolution. Ithaca, NY:Comell Univ. Press

Gittleman, J. L. 1989b. The comparative ap-proach in ethology: aims and limitations.See Bateson & Klopfer 1989, pp. 55-83

Glickman, S. E,, Sroges, R. W. 1966. Curiosityin zoo animals. Behaviour 26:151-88

Gold, P. E., Stone, W. S. 1988. Neuroendo-crine effects on memory in aged rodentsand humans. Special issue: experimentalmodels of age-related memory dysfunctionand neurodegeneration. Neurobiol. Aging9:709--17

Goldthwaite, R. O., Coss, R. (3, Owings, D. H.1990. Evolutionary dissipation of an anti-snake system: differential behavior by Cal-ifornia and Arctic ground squirrels inabove- and below-ground contexts.Belutviour 112:246-69

Gottlieb, G. 1984. Evolutionary trends and ev-olutionary origins: relevance to theory incomparative psychology. Psychol. Rev.91:448-56

Gottlieb, G. 1992. Individual Development andEvolution: The Genesis of Novel Behavior.New York: Oxford Univ. Press

Gould, J. L. 1986. The biology of learning.Annu. Rev. Psychol. 37:163 92

Grafen, A. 1991. Modelling in behaviouralecology. See Krebs & Davies 1991, pp. 5-31

Grant, P. 1986. Ecology and Evolution ofDarwin’s Finches. Princeton: PrincetonUniv. Press

Grier, J. W., Burk, T. 1992. Biology of AnimalBehavior. St. Louis: Mosby

Griffin, D. R. 1981. The Question of AnimalAwareness. New York: Rockefeller Univ.Press. 2nd ed.

Groebel, J., Hinde, R. A., eds. 1989. Aggres-sion and War: Their Biological and SocialBases. Cambridge: Cambridge Univ. Press

Grohmann, J. 1939. Modifikation oderFunktionsreifung? Ein Beitrag zur Kl~irungder Wechselseitigen Beziehungenzwischen lnstinkthandlung und Erfahrung.Z. Tierpsychol. 2:132-44

Grossberg, S., ed. 1988. Neural Networks andNatural Intelligence. Cambridge, MA:

MIT PressGyger, M., Marler, P. 1988. Food calling in the

domestic fowl, Gallus gallus: the role ofexternal referents and deception. Anim.Behav. 36:358--65

Hailman, J. P. 1965. Cliff-nesting adaptationsof the Galapagos swallow-tailed gull. Wil-son Bull. 77:346-62

Hailman, J. P. 1967. The ontogeny of an in-stinct. Behaviour (Suppl.) 15:1-159

Hailman, J. P. 1988. Operationalism, optimal-ity and optimism: suitabilities versus adap-tations of organisms. In EvolutionaryProcesses and Metaphors, ed. M. W. Ho,S. W. Fox, pp. 85-116. London: JohnWiley & Sons

Hall, W. G. 1990. The ontogeny of ingestivebehavior: changing control of componentsin the feeding sequence, See Stricker 1990,pp. 77-123

Harvey, P. H., Krebs, J. R. 1990. Comparingbrains. Science 249:140-46

Harvey, P. H., Pagel, M. D. 1991. The Com-parative Method in Evolutionary Biology.Oxford: Oxford Univ. Press

Hawkins, R. D., Kandel, E. R. 1984. Is there acell-biological alphabet for simple forms oflearning’? Psychol. Rev. 91:376-91

Heidiger, H. 1950. WiM Animals in Captivity.London: Butterwortlas

Heiligenberg, W. 1991. The neural basis of be-havior: a neuroethological view, ~4nnu.Rev. Neurosci. 14:247--67

He,-man, L. M., Morrel-Samuels, P., Pack, A.A. 1990. Bottlenosed dolphin and humanrecognition of veridical and degraded videodisplays of an artificial gestural language.J. Exper. Psychol.: Gen. 119:215-30

Herrnstein, R. J. 1990. Levels of stimulus con-trol: a functional approach. Cognition37:133-66

Hinde, R. A. 1966. Animal Behaviour. NewYork: McGraw-Hill

Hinde, R. A. 1991. From animals to humans.SeeDawkins 1991 et al, pp. 31 39

Hodos,W., Campbell, C. B. G. 1969. The scalanaturae: why there is no theory in compara-tive psychology. Psychol. Rev. 76:337-50

Hodos, W., Campbell, C. B. G. 1990. Evolu-tionary scales and comparative studies ofanimal cognition. See Kesner & Olton1990, pp. 1-20

Hogan, J. A. 1988. Cause and function in thedevelopment of behavior systems. SeeBlass 1988, pp. 63-106

Hogan, J. A., Honrado, G. I., Vestergaard, K.1991. Development of a behavior system:dustbathing in the Burmese red junglefowl(Gallus gallus spadiceus ): II. Internal fac-tors. J. Comp. Psychol. 105:269-73

Holldobler, B., Wilson, E. O. 1990. The Ants.Harvard: Harvard Univ. Press

Hollis, K. L. 1990. The role of Pavlovian con-ditioning in territorial aggression and re-production. See Dewsbury 1990, pp.197-220

Honig, W. K., Fetterman, J. G. 1992. Cognitive


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 705

Aspects of Stimulus Control. Hillsdale, NJ:Edbaum

Hughes, B. O., Duncan, I. J. 1988. The notionof ethological "need," models of motiva-tion and animal welfare. Anim. Behav.36:1696-1707

Immelmann, K., Barlow, G. W., Petrinovich,L., Main, M. 1981. Behavioral Develop-ment: The Bielefeld Interdisciplinary Proj-ect. Cambridge: Cambridge Univ. Press

Ingle, D., Crews, D. 1985. Vertebrate neu-roethology: definitions and paradigms.Annu. Rev. Neurosci. 8:457-94

Jerison, H. J. 1973. Evolution of the Brain andIntelligence. New York: Academic

Jerison, H. J., Jerison I., eds. 1988. Intelligenceand Evolutionary Biology. New York:Springer-Verlag

Kallander, H., Smith, H. G. 1990. Food storingin birds: an evolutionary perspective. Curr.Ornithol. 7:147-208

Kamil, A. C., Roitblat, H. L. 1985. The ecol-ogy of foraging behavior: implications foranimal learning and memory. Annu. Rev. ofPsychol. 36:141-69

Kamil, A. C. 1988. A synthetic approach to thestudy of animal intelligence. See Lager1988, pp. 257-308

Kamil, A. C., Krebs, J. R., Pulliam, H. R. 1987.Foraging Behavior. New York: Plenum

Kesner, R. P., Olton, D. S., eds. 1990. Neuro-biology of Comparative Cognition.Hillsdale, N J: Erlbanm

Kessel, E. L. 1955. Mating activities of balloonflies. Syst. Zool. 4:97 104

Ketterson, E. D., Nolan, V. Jr. 1992. Hor-mones and life histories: an integrative ap-proach. Am. Nat. In press

King, A. P., West, M. J. 1990. Variation inspecies-typical behavior: a contemporaryissue for comparative psychology. SeeDewsbury 1990, pp. 321-39

Krebs, J. R., Davies, N. B. 1984. BehaviouralEcology: an Evolutionary Approach Sun-derland, MA: Sinauer. 2nd ed.

Krebs, J. R., Davies, N. B. 1991. BehaviouralEcology: An Evolutionary Approach. Ox-ford: Blackwell. 3rd ed.

Krebs, J. R., Healy, S. D., Shettleworth, S. J.1990. Spatial memory of Paridae: compaii-son of a storing and a nonstoring spccics,the coal tit, Parus ater, and the great tit, P.Major. Anita. Behav. 39:1127-37

Kroodsma, D. E., Konishi, M. 1991. A subos-cine bird (eastern phoebe, Sayornis phoebe) develops normal song without auditoryfeedback. Anim. Behav. 42:47-7-87

Krushinski, L. V. 1965. Solution of elementarylogical problems by animals on the basis ofextrapolation. Prog. Brain Res. 17:280-308

Kuo, Z. Y. 1976. The Dynamics of BehaviorDevelopment: An Epigenetic .View NewYork: Plenum. Enlarged ed.

Lammers, J. H. C. M., Kruk, M. R., Meelis,W., van der Poel, A. M. 1988a. Hypotha-lamic substrates for brain stimulation-in-

duced attack, teeth-chattering and socialgrooming in the rat. Brain Res. 449:311-27

Lammers, J. H. C. M., Kruk, M. R., Meelis,W., van der Poel, A. M. 1988b. Hypotha-lamic substrates for brain stimulation-in-duced patterns of locomotion and escapejumps in the rat. Brain Res. 449:294-310

Langtimm, C. A., Dewsbury, D. A. 1991. Phy-logeny and evolution of rodent copulatorybehaviour. Anim. Behav. 41:217-25

Langton, C. L., Farmer, J. D., Rasmussen, S.,Taylor, C., eds. 1991. Artificial Life I1.New York: Addison-Wesley

Leger, D. W., ed. 1988. Nebraska Symposiumon Motivation, 1987: Comparative Per-spectives in Modem Psychology. Lincoln,NE: Univ. Nebraska Press

Lehrman, D. S. 1953. A critique of KonradLorenz’s theory of instinctive behavior. Q.Rev. Biol. 28:337~53

Lessells, C. M. 1991. The evolution of life his-tories. See Krebs & Davies 1991, pp. 32-68

Logue, A. W. 1988. A comparison of tasteaversion learning in humans and other ver-tebrates: evolutionary pressures in com-mon. See Bolles & Beecher 1988, pp.97-116

Lorenz, K. 1965. The Evolution and Modifica-tion of Behavior. Chicago: Univ. ChicagoPress

Lorenz, K. Z. 1950. The comparative methodin studying innate behaviour patterns.Symp. Soc. Exp. BioL, IV. 4:221-68

Losos, J. B. 1990. Concordant evolution of lo-comotor behaviour, display rate and mor-phology in Anolis lizards. Anim. Behav.39:879-90

Lucas, J. R., Walter, L. R. 1991. When shouldchickadees hoard food? Theory and experi-mental results. Anim. Behav. 41:579-601

MacLean, P. D. 1990. The Triune Brain inEvolution: Role in Paleocerebral Func-tions. New York: Plenum

Macphail, E. M. 1985. Vertebrate intelligence:the null hypothesis. In Animal Intelligence,ed. L. Weiskrantz, pp. 37-51. Oxford:Clarendon Press

Maier, N. R. F., Schneirla, T. C. 1935. Princi-ples of Animal Psychology. New York:McGraw-Hill

Mangel, M., Clark, C. W. 1988. DynamicModeling in Behavioral Ecology.Princeton, NJ: Princeton Univ. Press

Manson, J. H., Wrangham, R. W. 1991. Inter-group aggression in chimpanzees and hu-mans. Curr. Anthropol. 32:369-90

Marler, C.A., Moore, M. C. 1991. Supplemen-tary feeding compensates for testosterone-induced costs of aggression in malemountain spiny lizards, sceloporusjarrovi.Anim. Behav. 42:209-19

Marler, P. 1991. Song-learning behavior: theinterface with neuroethology. Trends Neu-rosci. 14:100-206

Marler, P., Hamilton, W. J. 1966. Mechanismsof Animal Behavior. New York: Wiley


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


706 TIMBERLAKE

Martin, G. M., Bechara, A., Van der Kooy, D.1991. The perception of emotion: parallelneural processing of the affective and dis-criminative properties of opiates. Psycho-biology 19:147-52

Martin, R. D. 1974. The biological basis ofhuman behavior. In The Biology of Brains,ed. W. B. Broughton, pp. 215-50. NewYork: Wiley

Martinez, J. L. Jr., Kesner, R. P. 1991. Learn-ing and Memory: A Biological View. NewYork: Academic

Mason, W. A., Lott, D. F. 1976. Ethology andcomparative psychology. Annu. Rev. Psy-chol. 27:129 54

Masterton, R. B., Heffner, H., Ravizza, R.1969. Evolution of human hearing. J.Acoust. Soc. Am. 45:966-85

Maynard-Smith, J. 1982. Evolution and theTheory of Games. Cambridge: CambridgeUniv. Press

Mayr, E. 1974. Behavior programs and evolu-tionary strategies. Am. Sci. 62:650-59

McClintock, M. K. 1984. Group mating in thedomestic rat as a context for sexual selec-tion: consequences for the analysis of sex-ual behavior and neuroendocrineresponses. In Advances in the Study of Be-havior, ed. J. Rosenblatt, C. Beer, R.Hinde, pp. 1-15. New York: Academic

McNab, B. 1989. Basal rate of metabolism,body size, and food habits in the order Car-nivora. See Gittleman 1989a, pp. 335-54

McNamara, J. M., Houston, A. I. 1990. Thevalue of fat reserves and the tradeoff be-tween starvation and predation. ActaBiotheor. 38:37~51

Michener, C. D. 1974. The Social Behavior ofthe Bees. Cambridge, MA: Harvard Univ.Press

Milinski, M., Parker, G. A. 1991. Competitionfor resources. See Krebs & Davies 1991,pp. 137~58

Miller, D. 1988. Development of instinctivebehavior: an epigenetic and ecological ap-proach. See Blass 1988, pp. 415-44

Moffatt, C. A., Hogan, J. A. 1992. Ontogeny ofchick responses to maternal food calls inthe Burmese Red Junglefowl (Gallus gal-lus spadiceus). J. Comp. Psychol. 106:92-96

Moore, C. L~ 1990. Comparative developmentof vertebrate sexual behavior: levels, cas-cades, and webs. See Dewsbury 1990, pp.278-99

Mrosovsky, N. 1990. Rheostasis: The Physiol-ogy of Change. Toronto: Univ. TorontoPress

Mulder, M. B. 1991. Human behavioural ecol-ogy. See Krebs & Davies 1991, pp. 69-98

Myers, G. 1990. Writing Biology: Texts in theSocial Construction of Scientific Knowl-edge. Madison: Univ. Wisconsin Press

Nitecki, M. H., ed. 1988., Evolutionary Prog-ress. Chicago, IL: Univ. Chicago Press

Nottebohm, F. 1991. Reassessing the mecha-nisms and origins of vocal learning in

birds. Trends Neurosci. 14:206-11Novak, M. A., Petto, A. J. 1991. Through the

Looking Glass: Issues of PsychologicalWell-being in Captive Nonhuman Pri-mates. Washington, DC: American Psy-chological Association

Olson, D. J. 1991. Species differences in spa-tial memory among Clark’s nutcrackers,scrub jays, and pigeons..L Exp. Psychol.:Anim. Behav. Process. 17:363-76

Oppenheim, R. W. 1982. Preformation andepigenesis in the origins of the nervous sys-tem and behavior: issues, concepts, andtheir history. In Perspectives in Ethology,ed. P. P. G. Bateson, P. H. Klopfer, 5:101-131. New York: Plenum

Oyama, S. 1985. The Ontogeny of Information.Cambridge: Cambridge Univ. Press

Parker, S. T. 1990. Origins of comparative de-velopmental evolutionary studies of pri-mate mental abilities. See Parker & Gibson1990, pp. 3~64

Parker, S. T., Gibson, K. R., ed. 1990. "Lan-guage" and Intelligence in Monkeys andApes. Cambridge: Cambridge Univ. Press

Patterson, I. J. 1965. Timing and spacing ofbroods in the black-headed gull, Larusridibundus. Ibis 107:433-59

Pepperberg, I. M. 1990. Some cognitive capac-ities of an African Grey parrot (Psittacuserithacus). Adv. Stud. Behav. 19:357~-09

Pepperberg, I. M., Funk, M. S. 1990. Objectpermanence in four species of psittacinebirds: an African Grey parrot (Psittacuserithacus), an Illiger mini macaw (Aramaracana ), a parakeet (Melopsitmcus un-dulatus), and a cockatiel (Nymphieushollandicus). Anita. Learn. Behav. 18:97-108

Petrinovich, L. 1990. Avian song development:methodological and conceptual issues. SeeDewsbury 1990, pp. 340-59

Plomin, R., DeFries, J. C., McClearn, G. E.1990. Behavioral Genetics. San Francisco:Freeman

Plowright, C. M., Shettleworth, S. J. 1991.Time horizon and choice by pigeons in aprey-selection task. Anim. Learn. Behav.19:103-12

Povinelli, D. J., Parks, K. A., Novak, M. A.1991. Do rhesus monkeys (Macaea mul-atta) attribute knowledge and ignorance toothers. J. Comp. Psychol. 105:318-25

Premack, D. 1983. The codes of men andbeasts. Behav. Brain Sci. 6:125--67

Provine, W, B. 1988. Progress in evolution andmeaning in life. See Nitecki 1988, pp. 49-74

Pulliam, H. R., Caraco, T. 1984. Living ingroups: Is there an otpimal group size? SeeKrebs & Davies 1984, pp. 122-47

Razran, G. 1971. Mind in Evolution. NewYork: Houghton Mifflin

Real, L. A. 1991. Animal choice behavior andthe evolution of cognitive architecture. Sci-ence 253:980-86

Regan, T. 1983. The Case for Animal Rights.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


ANIMAL BEHAVIOR 707

Berkeley, CA: Univ. California PressRichner, H., Marclay, C. 1991. Evolution of

avian roosting behaviour: a test of the in-formation centre hypothesis and of a criti-cal assumption. Anita. Behav. 41:433-38

Ridley, M. 1983. The Explanation of OrganicDiversity. Oxford: Clarendon Press

Riley, D. A., Langley, C. M. 1993. The logicof species comparison. PsychoL Sci. Inpress

Ristau, C. A., ed. 1991. Cognitive Ethology:The Minds of Other Animals: Essays inHonor of DonaM R. Griffin. Hillsdale, NJ:Erlbaum

Robinson, G. E., Page, R. E., Strambi, C.,Strambi, A. 1989. Hormonal and geneticcontrol of behavioral integration in honeybee colonies. Science 246:109-12

Robinson, M. H. 1985. Predator-prey interac-tions, informational complexity, and the or-igins of intelligence. J. Washington Acad.Sci. 75:91 104

Robinson, M. H. 1991. Niko Tinbergen, com-parative studies and evolution. SeeDawkins et al 1991, pp. 100-28

Roitblat, H. L. 1987. Introduction to Compar-ative Cognition. San Francisco: Freeman

Romanes, G. J. 1884. Mental Evolution in Ani-mals. New York: Appleton

Rose, J. D. 1990. Forebrain influences onbrainstem and spinal mechanisms of copu-latory behavior: a current perspective onFrank Beach’s contribution. NeuroscLBiobehav. Rev. 14:207-16

Rosenblatt, J. S. 1989. The physiological andevolutionary background of maternal re-sponsiveness. New Direct. Child Dev.1989:15-30

Rovee-Collier, C., Kupersmidt, J., O’Brien, L.,Collier, G. 1991. Behavioral thermoregula-tion and immobilization: conflicting de-mands for survival. J. Comp. Psychol.105:232-42

Rusak, B., Robertson, H. A., Wisden, W.,Hunt, S. P. 1990. Light-pulses that shiftrhythms induce gene-expression in thesuprachiasmatic nucleus. Science248:1237-40

Savage-Rumbaugh, S. 1988. A new look at apelanguage: comprehension of speech andsyntax. See Lager 1988, pp. 201-55

Schmidt-Nielsen, K. 1984. Scaling: Why IsAnimal Size so Important? Cambridge:Cambridge Univ. Press

Scott, J. P. 1973. The organization of compara-tive psychology. Ann. NY Acad. Sci.223:7-40

Sherry, D. F., Vaccarino, A. L., Buckenham,K., Herz, R. S. 1989. The hippocampalcomplex of food-storing birds. BrainBehav. Evol. 34:308-17

Shettleworth, S. J. 1990. Spatial memory infood-storing birds. Philos. Trans. R. Soc.London Ser. B 329; 143-51

Shettleworth, S. J. 1993. Where is the comparison in comparative cognition? Psychol.Sci. In press.

Silver, R. 1990. Biological timing mechanismswith special emphasis on the parental be-havior of doves. See Dewsbury 1990, pp.252-77

Simmons, J. A. 1989. A view of the worldthrough the bat’s ear: the formation ofacoustic images in echolocation. Cognition33:155-99

Sinervo, B., Huey, R. 1990. Allometric engi-neering: testing the causes of inter-populational difference in performance.Science 278:1106-9

Singer, P. 1990. The significance of animalsuffering. Behav. Brain. Sci. 13:9-12

Skinner, B. F. 1938. The Behavior of Organ-isms. New York: Appleton-Century-Crofts

Skinner, B. F. 1959. A case history in the sci-entific method. In Psychology: The Studyof a Science, ed. S. Koch, 2:359-79. NewYork: McGraw-Hill

Slobodchikoff, C. N., ed. 1988. The Ecology ofSocial Behavior. New York: Academic

Snowdon, C. T. 1983. Ethology, comparativepsychology, and animal behavior. Annu.Rev. Psychol. 34:63-94

Spear, N. E., Miller, J. S., Jagielo, J. A. 1990.Animal memory and learning. Annu. Rev.Psychol. 41:169-211

Squire, L. R. 1992. Memory and the hippocam-pus: a synthesis from findings with rats,monkeys, and humans. Psychol. Rev.99:195-231

Staddon, J. E. R. 1989. Animal psychology: thetyranny of anthropocentrism. See Bateson& Klopfer 1989, pp. 123-34

Stephens, D. W., Krebs, J. R. 1986. ForagingTheory. Princeton: Princeton Univ. Press

Stricker, E. M. 1990. Handbook of BehavioralNeurobiology Vol. 10. Neurobiology ofFood and Fluid Intake. New York: Plenum

Suboski, M. D. 1989. Recognition learning inbirds. In Bateson & Klopfer 1989, pp. 137-1

Thelen, E. 1990. Dynamical systems and thegeneration of individual differences. In In-dividual Differences in Infancy: Reliabil-ity, Stability, and Prediction, ed. J.Colombo, J. W. Fagen, pp. 77-117. Engle-wood Cliffs, NJ: Erlbaum

Thelen, E., Ulrich, B. D. 1991. Hidden skills.Monogr. Soc. Res. Child Dev. 561:1-106

Thiessen, D. 1990. Body size, allometry, andcomparative psychology: locomotion andforaging. See Dewsbury 1990, pp. 80-100

Thornhill, N. W., Thornhill, R. 1991. An evo-lutionary analysis of psychological painfollowing human (Homo sapiens) rape: IV.The effect of the nature of the sexual as-sault. J. Comp. Psychol. 105:243-52

Thorpe, W. H. 1979. The Origins and Rise ofEthology. New York: Praeger

Timberlake, W. 1983. The functional organiza-tion of appetitive behavior: behavior sys-tems and learning. In Advances in Analysisof Behavior. Biological Factors in Learn-ing, ed. M. D. Zeiler, P. Harzem, 3:177-221. Chichester: Wiley


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


708 TIMBERLAKE

Timbedake, W. 1990. Natural learning in labo-ratory paradigms. See Dewsbury 1990, pp.31-54

Timberlake, W., Washbume, D. L. 1989. Feed-ing ecology and laboratory predatory be-havior toward live and artificial movingprey in seven rodent species. Anim. Learn.Behav. 17:2-11

Timbedake, W., Lucas, G. A. 1989. Behaviorsystems and learning: from misbehavior togeneral laws. In Contemporary LearningTheories: Instrumental Conditioning The-ory and the Impact of Biological Con-straints on Learning, ed. S. B. Klein, R. R.Mowrer, pp. 237-75. Hillsdale, NJ:Erlbaum

Timberlake, W., White, W. 1990. Winningisn’t everything: rats need only food depri-vation and not food reward to efficientlytraverse a radial ann maze. Learn. Motiv.21:153~53

Tinbergen, N. 1951. The Study of Instinct. Ox-ford: Clarendon Press

Tinbergen, N., Perdeck, A. C. 1950. On thestimulus situation releasing the begging re-sponse in the newly hatched herring gullchick (Larus argentatus argentatus Pont.)Beha viour 3:1-39

Toates, F. 1986. Motivational Systems. Cam-bridge: Cambridge Univ. Press

Tomonaga, M., Matsuzawa, T. 1992. Percep-tion of complex geometric figures in chim-panzees (Pantroglodytes) and humans(Homosapeins): analyses of visual similar-ity on the basis of choice reaction time. J.Comp. Psychol. 106:43-52

Uexkull, J. von 1934. Streifztige durch die Um-welten von Tieren und Menschen. 13erlin:Springer. Transl. in Schiller, C. H. 1957.Instinctive Behavior: The Development ofa Modern Concept. New York: Interna-tional Universities Press

Van Tets, G. F. 1965. A comparative studyof some social communication patterns inthe Pelecaniformes. Ornithol. Monogr.2:1-88

Vander Wall, S. T. 1990. Food Hoarding inAnimals. Chicago: Univ. Chicago Press

Vaughan, T. A. 1978. Mammalogy. Philadel-phia: Saunders

Viney, W., King, D., Berndt, J. 1990. Animalresearch in psychology: declining or thriv-ing? J. Comp. Psychol. 104:322-25

Warden, C. J., Jenkins, T. N., Warner, L. H.1935. Comparative Psychology: A Corn-

prehensive Treatise, Vol. 1. New York:Ronald

Washburn, D. A., Rumbaugh, D. M. 1991. Or-dinal judgments of numerical symbols bymacaques (Macaca mulatta ). Psychol. Sci.2:190-93

Wasserman, E. A. 1993. Comparative cogni-tion: beginning the second century of thestudy of animal intelligence. Psychol. Bull.In press

Wasserman, E. A., Kiedinger, R. E., Bhatt, R.S. 1988. Conceptual behavior in pigeons:categories, subcategories, and pseudo cate-gories. J. Exp. Psychol.: Anita. Behav.Process. 14:235-46

Weatherhead, P. J., Montgomerie, R. D. 1991.Good news and bad news about DNA fin-gerprinting. Trends Ecol. Evol. 6:173-4

West, M. J., King, A. P. 1990. Mozart’s star-ling. Am. ScL 78:106-14

Wheeler, D. A., Kyriacou, C. P., Greenacre,M. L., Yu, Q., Rutila, J. E. 1991. Moleculartransfer of a species-specific behavior fromDrosophila simulans to Drosophilamelanogaster. Science 251:1082-85

Whiten, A., ed. 1991. Natural Theories ofMind: Evolution, Development and Simu-lation of Everyday Mindreading. Oxford:Basil Blackwell

Wilder, K. C., Finlay, 13. L. 1989. Develop-mental heterochrony and the evolution ofspecies differences in retinal specializa-tions. In Development of the VertebrateRetina, ed. B. L. Finlay, D. R. Sengelaub,pp. 227-46. New York: Plenum

Wilson, E. O. 1975. Sociobiology: The NewSynthesis. Cambridge, MA: Harvard Univ.Press

Wingfield, J. C., Moore, M. C. 1988. Hor-monal, social, and environmental factors inthe reproductive biology of free-livingmale birds. See Crews 1988, pp. 149-75

Yoerg, S. I. 1993. Ecological frames of mind:the role of cognition in behavioral ecology.Q. Rev. BioL In press

Young, R. K., Thiessen, D. D. 1991. Washing,drying, and anointing in adult humans(Homo sapiens): commonalities withgrooming sequences in rodents. J. Comp.Psychol. 105:340-44

Zentall, T. R., Galef, B. G. Jr. 1988. SocialLearning: Psychological and BiologicalPerspectives. Hillsdale. NJ: Erlbaum


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.


Ann

u. R

ev. P

sych

ol. 1

993.

44:6

75-7

06. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by I

ND

IAN

A U

NIV

ER

SIT

Y -

Blo

omin

gton

on

02/2

4/09

. For

per

sona

l use

onl

y.

animal behavior: a continuing synthesis...animal behavior and closer ties to neuroscience. changes...

Documents