Earlier students of animal behavior found it convenient to classify behaviors as either innate or learned. "Innate" refers to behaviors that have a strong genetic basis, behaviors that are influenced very little by experience. Fighting, copulation, and parental behavior seem to fit into this category.
Perhaps none of the numerous definitions of "learning" fits all behavior that we refer to as learned behavior. However, the definition of a learned behavior as "a behavior that has been changed by experience" will suffice for our purposes here.
The middle of this century witnessed lively debates between classical ethologists, who stressed the importance of innate behaviors, and behaviorists, who argued that only learned behaviors are scientifically valid or even worthy of scientific attention. Currently, most students of animal behavior feel comfortable interpreting most behaviors as products that result from the interaction of genes and environment.
For example, the basic movements of food-begging in newly hatched sparrows are certainly innate, since hatchlings could hardly have had time to learn these movements by trial and error. On the other hand, food-begging becomes more effective with experience, indicating that this essentially innate behavior clearly changes with experience. It is difficult to imagine an innate behavior that cannot be changed in some way by experience, but it is equally difficult to visualize a behavior totally uninfluenced by innately programmed movements.
Some Major Categories of Bird Behavior
All attempts to define and organize behaviors are to some degree arbitrary, but the following categories are useful when dealing with the behaviors treated in this book.
Reflexes are innate (genetically programmed) muscular contractions that are elicited by a stimulus. The most obvious reflexes are responses to mechanical, visual, and auditory stimuli, and to sudden movements of the body. Reflexes are easily overlooked because they are usually quick and subtle.
A hawk gliding to a perch makes numerous, small reflexive wing movements to counter the effects of shifting winds. Upon landing, the hawk regains its balance with reflexive movements in the leg, wing, and body. These small reflexes are of the same type as when one taps the tendon beneath the knee and the lower leg automatically kicks forward.
The innate character of reflexes is evident in newly hatched birds, which, like older birds, blink in response to blowing dust or crouch when they hear a sudden noise.
The direction and extent of a limb movement is normally beyond our control during a reflex—a fact that is obvious when we touch a hot surface and watch our arm take its own course. On the other hand, all reflexes are probably modified to some extent by experience. With training, most animals can learn to increase or decrease the intensity of a reflex and, in some cases, to abolish it.
In one common reflex, exposure to cold causes microscopic muscles at the base of the feathers to contract so that the feathers fluff up and insulate the bird. However, a different stimulus, for example, the appearance of a bird's mate, can also cause feathers to fluff up, especially the feathers on the top of the head that form the crest. In this case feather erection has nothing to do with temperature but instead most likely functions as a signal.
Some reflexes are hidden from view, permitting us to see only their effects. In gulls, when fledglings peck at their parents' bills, the parents involuntarily regurgitate partially digested food that the young birds eat. Another hidden reflex is the contraction of the muscles that surround the salivary glands in ant-eating birds such as flickers. The stimulus for this reflex is the formic acid found in ants, and the response is the secretion of saliva that neutralizes the formic acid.
Fixed Action Patterns
Perhaps the most widely studied innate behavior in animals is the fixed action pattern (FAP). A particularly instructive example of a FAP is egg retrieval in geese: an incubating goose extends her head and neck and with her bill pulls back an egg that has rolled out in front of the nest. Even geese raised in isolation do this.
Several features characterize FAPs. Besides being innate, they are stereotyped in that they are relatively invariable (i.e., geese never retrieve eggs except in this manner). FAPs also have an obvious steering component. When the egg rolls off course because of irregularities in the substrate, the goose adjusts to the egg's changing locations by modifying her head and neck movements.
The stimulus that triggers a FAP is called a releaser. An egg is a visual releaser, but releasers can also be auditory, tactile, or olfactory. Moreover, releasers are effective only in specific contexts. For example, a goose will respond to an egg placed in front of her as long as she is sitting on her nest (the appropriate context), but she is unresponsive when she is away from the nest.
FAPs differ from learned behaviors because they are innate, and they differ from reflexes because reflexes have no steering component. Interestingly, if the stimulus (the displaced egg) is removed while the FAP is in progress, the FAP continues until it reaches completion; for example, the goose continues making egg-retrieving movements even though the egg is now in the hand of the experimenter. Although theoretically all FAPs continue to completion when the stimulus is suddenly withdrawn, it is very difficult or even impossible to demonstrate this characteristic except in special cases like egg retrieval.
FAPs are always short in duration (usually only a few seconds) because the series of muscular contractions that make up a FAP must follow a particular sequence. It is not possible for the brain to hard-wire instructions for a sequence of muscular contractions that lasts more than a few seconds.
Common FAPs in birds include pecking at a seed, reaching out for prey with the talons, tearing a piece of flesh from a carcass, drinking, billwiping, preening, inserting food into a nestling's mouth, and copulating. All of these FAPs are triggered by an appropriate releaser. Another behavior that is possibly a FAP (or at least has components that are FAPs) is anting, during which birds stroke their wings, bodies, and tails with ants that they hold in their bills.
Although some investigators question the usefulness as well as the validity of the FAP as a scientific concept, FAPs as classically defined here are useful for understanding more complex behaviors like those described below.
Innate Behaviors in Conflict Situations
As a rule, birds respond to threats with innate rather than learned behaviors. For example, when a threatened goose reaches out and bites an attacker, it uses basically the same hard-wired program for muscular contractions as it does for pulling at submerged vegetation (i.e., it utilizes a FAP). One reason FAPs, rather than learned movements, are the usual building blocks of defense behavior is that birds cannot afford to err during the trial-and-error process of learning a crucial defense tactic. Moreover, there is always the risk that the bird might forget a learned response at a critical moment. The same line of reasoning explains why attack behaviors are essentially innate.
Sometimes, when birds are threatened by predators but seem unable to choose between attacking or fleeing, they respond by attacking—but by attacking an object other than the predator. Thus, they redirect their attack to a substitute object, hence the term "redirection" for this type of behavior. For example, if a turkey is threatened by a mountain lion that is standing close enough to produce anxiety but not close enough to incite fleeing, the turkey may resolve the conflict by pecking a smaller turkey nearby or even an inanimate object.
Frequently, threatened animals exhibit FAPs that appear, at least to human observers, to be irrelevant or inappropriate responses to a threat. Incubating terns face a dilemma when a person slowly approaches the nest: they could risk injury to themselves by fighting, or risk injury to their eggs by fleeing. Curiously, terns often preen vigorously under these circumstances. The act of preening in this context appears irrelevant or inappropriate.
Such irrelevant or inappropriate responses to threats are called displacement behaviors. Preening, eating, bill-wiping, stretching, and drinking movements are common displacement behaviors when they are responses to threats. Bill-wiping as a displacement behavior is commonly observed in flushed birds immediately after they alight on a perch.
Intention movements are very common responses to threats. Birds begin an attack or a fleeing response, then abruptly halt the movement.
Displays and Ritualization
Displays are innate (genetically programmed) stereotyped movements that have a communicatory or signal function. In terms of their function, they may be compared to culturally acquired human gestures, which are also stereotyped movements used to communicate. In both displays and gestures, ambiguity is reduced by exaggerating the movements as well as by performing them in a more stereotyped manner. For example, the bathing movements in a gander's precopulatory display appear to be an exaggerated and more stereotyped version of movements he uses when he actually bathes. (Likewise, the gesture of saluting in humans—although culturally acquired rather than inherited—appears to be an exaggerated and more stereotyped version of a noncommunicatory behavior, perhaps raising one's hand to shade the eyes.)
Many displays are unique to the species and are determined to some degree by the bird's anatomical and behavioral characteristics. Thus, prairie chickens would be expected to evolve terrestrial displays and Common Nighthawks, aerial displays.
Courtship displays often differ markedly among species that are closely related, since a bird that confuses its own display with that of a similar species might waste time and energy courting and inseminating the wrong species. Territorial displays tend to be more general and accordingly are recognizable by other species that could potentially invade the bird's territory.
The evolutionary origin of displays intrigued early ethologists, who reasoned that it is more efficient for a display to evolve from existing FAPs, reflexes, or intention movements than if they evolved de novo. The term "ritualization" is applied to this evolutionary process. For example, erecting the feathers—originally a reflex in response to cold—might acquire, through evolution, a signal or communicatory function to become part of a territorial display in which the bird raises its crest. Likewise, the FAP for drinking (lowering the head for water and then pointing the bill toward the sky) could evolve into a courtship display that employs the same basic movements.
Not all components of a display are ritualized reflexes, FAPs, or intention movements. The inflated neck pouch in the prairie chicken's display probably evolved uniquely for that purpose.
Obviously a bird has a greater chance of surviving if it responds appropriately to stimuli that indicate danger. Thus, a towhee feeding on the forest floor should flee when it hears a sudden rustling of leaves that might signal an approaching predator. On the other hand, the bird will waste valuable time and energy if it flies away each time the wind noisily blows leaves across the forest floor. Clearly the towhee must learn to distinguish between harmful and innocuous stimuli.
Birds learn to ignore harmless, repeated stimuli by responding less and less to the stimulus each time it is presented. Eventually they do not respond at all. A diminished response to a harmless, or innocuous, stimulus is termed habituation. In the above example we would say that the towhee habituated to the repeated rustling of leaves caused by the wind, or, in everyday language, that it tuned out the rustling sounds.
Birds are constantly habituating to the ocean of innocuous stimuli they encounter each day. For example, early in the morning, House Sparrows feeding near a highway take flight when the first automobile passes by, but as traffic continues to flow with no harmful effects, their uneasiness diminishes, and soon they are virtually unresponsive.
Dishabituation is a disruption of the state of habituation. An animal that habituates to a noise or other stimulus will again become aware of that stimulus if a different stimulus is presented to it. Sparrows that are habituated to a constant sound like an idling automobile could once again respond to this sound and flee if there is a sudden gust of wind.
Sensitization refers to a process whereby an animal, immediately after responding to one stimulus (for example, food), now responds to a neutral stimulus, one to which it is normally unresponsive. One of the first sensitization experiments dealt with octopuses. Octopuses normally ignore a glass rod (a neutral stimulus) that is inserted into their aquarium. However, if the glass rod is placed into the aquarium immediately after the octopuses have been fed, they will attack the rod. Thus, they become sensitized by the process of feeding and as a consequence respond to a neutral stimulus (in this case, a glass rod).
Generally the consequences of sensitization in the wild can only be inferred, but surely sensitization must be a common occurrence. Perhaps birds that have just been frightened are more readily disturbed by a neutral stimulus, such as an airplane flying overhead, than they would be otherwise.
Classical Conditioning and Operant Conditioning
Conditioning was first studied in detail by the eminent Russian physiologist Ivan Pavlov. Pavlov took advantage of the fact that dogs naturally salivate when presented with a piece of meat. Just before presenting the meat to the dog he presented a second stimulus (for example, the sound of a buzzer) to which dogs normally do not respond. Thus, he paired the buzzer (called the conditioned stimulus) with meat (called the unconditioned stimulus). Soon the dog was salivating every time he heard the buzzer, which was never more than a second or two before the meat appeared. The innate response (salivation) to the unconditioned stimulus (meat) is called the unconditioned response, and the learned response (also salivation) to the conditioned stimulus (the buzzer) is called the conditioned response.
We rarely witness classical conditioning while it is occurring in the wild; more often we only infer that it has occurred. For example, the first time a person drives up to a lake and throws grain to a duck, the duck responds by becoming excited. The excitement, which is a response to the grain (not to the appearance of the person), can be compared to salivation in dogs that are presented meat.
After a few days, the duck associates the approach of the person with the appearance of the grain, in the same way that the dog associates the buzzer with the appearance of the meat. Predictably, the duck becomes excited, indicating that it has become classically conditioned to the approach of the person. The stimuli and responses can be compared to those in the classically conditioned dog. The unconditioned stimulus is the grain, which corresponds to the meat. Likewise, the unconditioned response (corresponding to salivation in the dog) is the duck's excitement upon first being presented with the grain. The conditioned stimulus (the buzzer in the experiment with the dog) is the approach of the person. Finally, the conditioned response (salivation in the dog), like the unconditioned response, is the duck's excitement.
Like classical conditioning, operant conditioning (also called instrumental conditioning or trial-and-error learning) is perhaps more easily understood by example than by definition. When a pigeon is placed in a cage containing a lever that dispenses food, it pecks randomly because it is unaware of the significance of the lever. When it accidentally pecks the lever and food drops into the cage, a contingency (a connection or relationship) is set up between pecking the lever and receiving food. At this point we say that the pigeon is operantly conditioned to peck the lever whenever it wants food. In a variant of this experiment, the pigeon is not rewarded with food when it presses the lever; instead, it is punished with a shock from the electrified floor of the cage. In this case the pigeon has been conditioned to avoid pressing the lever.
In both experiments the act of pecking the lever is called the operant. It is important to emphasize that an operant—usually a body movement—is required for operant conditioning to take place: the animal must do something to initiate conditioning. In classical conditioning it is the conditioned stimulus rather than the animal's behavior that initiates the conditioning process; to be conditioned, the animal need do nothing but respond.
Numerous behaviors in birds are no doubt the products of operant conditioning. Turkey Vultures that wander randomly over the countryside looking for carrion can be compared to pigeons that peck randomly in their cage. By chance the vultures may fly over a highway, just as by chance the pigeon pecks at the lever. Since flying to the highway and pecking the lever are rewarded (in both cases with food), then both behaviors are operants. After a few trials, both the pigeon and the vulture are conditioned to repeat their respective behaviors when they want food.
Innate behaviors, like FAPs, are commonly shaped or modified by operant conditioning. Consider a chick a few minutes after hatching. Almost immediately it engages in an innate feeding behavior, specifically, the FAP for pecking indiscriminately at any small object that is in front of it. The releaser for this FAP could be a tiny pebble, seed, or other small object. Soon the chick picks up only the seeds and ignores the pebbles. Thus, feeding behavior has been modified by operant conditioning: the chick was obviously rewarded for picking up seeds.
Body movements in virtually all behaviors are modified by operant conditioning. For example, young birds inherit the ability to make basic flying movements with their wings, but those movements alone do not enable the bird to fly. Young birds must learn, through operant conditioning, how to modify their movements so as to achieve flight.
Animals conditioned to a particular stimulus also respond (though less intensely) to a second stimulus, as long as the second stimulus is more or less like the first one. This phenomenon is termed generalization because the animals seem to be generalizing that similar stimuli produce the same rewards and punishments. For example, pigeons conditioned to peck round levers will also peck oval levers, and vultures that have been rewarded for foraging along highways will also forage along smaller farm roads.
Extinction of a learned response occurs when the reinforcer is withdrawn. For example, pigeons trained to peck a lever for food eventually cease pecking the lever if food is not delivered. Extinction differs from forgetting. An Acorn Woodpecker might forget the location of a nut it has hidden, but this would have nothing to do with whether a reinforcer is withdrawn.
In imprinting, an animal, usually a very young one, establishes a bond with an animal or inanimate object that it faithfully follows for the next few weeks or months. Young animals normally imprint on their parents, but in the absence of their real parent, they will imprint on a surrogate mother, which can be another animal, including humans, or an inanimate object. Birds imprint during a brief critical period after hatching, a window that normally lasts a few hours or days.
Imprinting is characteristic of precocial birds, those that move about and feed almost immediately after birth or hatching. Waterfowl and quail are common examples of precocial birds. The close bond established by imprinting helps insure that young birds follow their parents during this vulnerable period of life. Imprinting is essentially absent in altricial species (in particular, songbirds), which hatch in a helpless condition. Because altricial birds have no opportunities to stray from parental care, they have no need to bind so closely with their parents.
Learning and the Development of Songs
The enormous literature dealing with song development in birds can hardly be summarized here, but generally, passerines acquire their songs by learning how to sequence innately produced sounds, termed the subsong, correctly. This process has been compared with language acquisition in humans, during which a baby correctly sequences innate sounds (babbling) into speech (although learned sounds are of course also incorporated into speech). Thus, song development, like human language acquisition, can be considered the result of both innate and learned processes.
Usually a young bird learns the correct sequence by listening to an adult bird. When a juvenile male hears an adult male sing the notes in the correct sequence, he learns this sequence and correctly arranges the elements of his subsong to produce the song that we hear. Learned songs also may vary geographically, giving rise to dialects. Dialects could be adaptive because when a female chooses a male with a familiar dialect, she might be selecting a male adapted for surviving in the region where her offspring will live. That dialects can influence mating is demonstrated by the observation that female White-crowned Sparrows often assume the precopulatory position when they hear the dialect of their own region, but rarely do so when they hear other dialects.
Learning is not important for song development in all species. In general, nonpasserines, such as ducks and quail, do not require exposure to the adult song in order to vocalize correctly.
Another type of learning, latent learning, is so designated because the knowledge that is acquired presumably remains latent until it is needed at a later time. Latent learning can be demonstrated in a well-known experiment with mice. A satiated mouse is allowed to explore a maze containing food pellets that are concealed in a particular part of the maze. Eventually the mouse discovers the pellets but ignores them because it is not hungry. Later, after being deprived of food, the mouse is allowed to reenter the maze. At this time it locates the food quickly, much more quickly than a mouse that has never been in the maze. Evidently the mouse learned the location of the food when it first entered the maze but did not use the knowledge until later.
Latent learning is probably impossible to demonstrate in the field. However, satiated birds may occasionally learn the location of a food source that they do not exploit at the moment, only to return to it later when they are hungry.
Insight learning is the most difficult type of learning to demonstrate in animals. Humans employ insight learning when they solve a problem—a mathematical problem or logical puzzle—in a novel way, when the solution comes suddenly as an insight. This type of learning is sometimes referred to as "Aha!" learning.
Decades ago, insight learning was proposed to explain how chimpanzees managed to rearrange tables and put together sticks to reach a banana suspended from the ceiling. It was argued that the chimps had a sudden insight as to how to reach the banana, in effect, that they figured out a novel solution to a problem they had not previously encountered. Neither classical nor operant conditioning seemed adequate for explaining this feat.
On the other hand, it has also been argued that chimps, being playful animals, accidentally solve the problem through their normal playful antics and never really require an insight to arrange the tables and sticks appropriately. Indeed, chimps experienced in play solve the problem more quickly than inexperienced chimps.
Whether birds experience such insights is debatable, but an experiment using Common Ravens certainly provokes thought along these lines. A bird standing on the top of a table was shown a string that was attached to the tabletop and hung over the side. Tied to the other end of the string and suspended about halfway down was a peanut. Most of the birds that were tested looked down at the peanut and seemed incapable of figuring out how to retrieve it. A Common Raven, however, stood on one foot, reached down with its bill, grabbed the string, pulled it up part of the way, held that part of the string with its foot, then repeated the process until the peanut was within reach.
Some Topics in Behavioral Ecology
Commonly called pecking orders, dominance hierarchies are social systems in which certain members of a group dominate others. Everything else being equal, animals that maintain a dominance hierarchy are more likely to survive than those that do not. Their interactions with each other are relatively stable and predictable, making it unnecessary to waste energy fighting among themselves for food or other resources.
The simplest dominance hierarchy is a linear hierarchy, in which one male (designated here as A) dominates for food resources over all other males. The next most dominant male (B) dominates over all males except A, and so forth, so that the order of dominance can be expressed as ABCD, etc.
Dominance hierarchies can be complex and sometimes puzzling. For example, some take the form of ABCA, indicating that whereas A dominates over B and B dominates over C, C actually dominates over A. Also, ranks sometimes shift within the hierarchy depending on the location of the individuals, their state of health, their reproductive state, or any other factor that might influence the outcome of a conflict.
What seems more remarkable than shifting ranks is that sometimes dominant individuals retain their dominance even when disease, malnutrition, injury, or age makes them incapable of winning a contest should their dominance be challenged. In these cases, contests for dominance do not occur, and physically superior subordinates continue acting as subordinates. They could be doing this because they learned a subordinate role as a young bird and retained this bias throughout life.
Dominance is usually less stable and less frequently linear in females. Females engage in far fewer confrontations than males, and when they do confront each other, their mates usually step in and terminate the contest before a hierarchy has been established.
Advantages of Flocking in Birds
Several reasonable hypotheses have been proposed to explain why feeding, roosting, and moving about in flocks could increase a species' chances of surviving in nature. Unfortunately, too few experiments (like the one with pigeons described below) have been conducted to test those hypotheses. Furthermore, it is not always obvious why, or even if, natural selection should favor flocking in extreme cases, such as the hundreds of thousands of blackbirds that roost together in winter.
The following are possible ways flocking might benefit birds:
- Improved detection of predators (Many eyes spot predators better than fewer eyes.)
- Improved defense of food resources
- Improved care of the young through communal feeding and protection
- More efficient predation for larger prey
The hypothesis that flocking improves detection of predators was tested by releasing a trained Goshawk a set distance from different-sized flocks of European Wood Pigeons (Columba palumbus). Increased vigilance was inferred because pigeons in large flocks took flight sooner than those in small flocks. Consistent with those results is the observation that Rock Doves feeding alone feed at a slower rate and thereby have more time to look around for predators.
Altruistic behavior may be defined as behavior that benefits another animal at the expense of the altruist. It is generally assumed that birds are compelled to act a certain way—altruistically or not—but as long as an act benefits another individual and is detrimental in some way to the altruist, then the definition of altruism as used here is satisfied. Therefore, we need not consider questions of motivation or how to distinguish truly altruistic acts from those that are self-serving and disguised (martyrs possibly acting to satisfy a selfish need).
Acts that are detrimental to the individual can be beneficial to the species, as when a bird in a flock sees a predator and risks its life by giving an alarm call. Also, altruistic behavior directed to a family member helps insure that some of the family genes are passed along to future generations. Natural selection would be expected to favor families that have genes for altruistic parental behavior over families that do not.
The same reasoning applies to the extended family, as when sisters, brothers, or older offspring help parents raise their young. These relatives pass along some of their own genes as well as those of the parents, since they share a certain number of genes with the parents. For example, female Mexican Jays that have lost their mates are sometimes assisted by their sisters in raising their offspring. In addition to learning and practicing parental skills before they nest the following year, these young females also pass along genes they share with their sisters. Moreover, in cases where nesting birds retain their territories for life, young birds might remain with their parents a year or two because they have no choice—the habitat is saturated. With no space to establish a new territory, it is best to wait in a territory of proven quality until a death provides room for a new territory.
Finally, altruistic behavior may be a consequence of living in groups, a by-product of parental care. In this case, natural selection favors individuals that feed any individual.
Mobbing, such as when screaming jays swarm around a hawk or owl, has been interpreted as benefiting some birds at the expense of others. Because flying close to a predator would seem to increase the mobbing bird's risk of being captured, natural selection should favor individuals that do not mob. However, some Australian owls capture almost nine times as many nonmobbing species as mobbing species. Moreover, to avoid being mobbed, these Australian owls tend to roost in dense forests, away from the area where most mobbing species occur. Thus, in spite of its risks, the function of mobbing could simply be to drive away predators from a preferred foraging area.
Optimal Foraging Strategies
We can reasonably assume that efficient foragers reproduce more successfully (leave more viable offspring) than inefficient foragers, everything else being equal. How efficiently an animal obtains food depends to a large extent on its foraging strategy. A common foraging strategy in flycatchers is to capture insects by darting out from a perch, whereas swallows fly continuously in search of insects. A bird's foraging strategy also includes variables such as the time of day it feeds, the habitat it utilizes, and the type and size of food it selects, all of which bear on the manner in which it obtains nourishment.
A particular foraging strategy cannot be equally efficient under all circumstances. For example, a hawk's best strategy in a habitat that is densely populated with rodents may be to wait and pounce, whereas where rodents are less common, perhaps it is best to soar and stoop. Although vitamins, minerals, and perhaps taste preferences are undoubtedly important in food selection, it is easier to visualize the main points of optimal foraging theory if we consider a simple case that has only calories as its currency.
Consider a hypothetical situation in which a quail walks through a meadow where two kinds of seeds, A and B, are randomly distributed on the ground. Seed A is relatively tender and takes only 5 seconds to peel and ingest. We say that this seed has a handling time of 5 seconds. Seed B, in contrast, has a tough, acornlike covering that requires a handling time of 25 seconds. Both A and B have the same number of calories.
Three possible foraging strategies are possible. In Strategy 1, the quail selects A. Since A has the shortest handling time, the quail will spend less time peeling and ingesting these seeds, but travel time will increase because in selecting only A the quail passes up all seeds of type B. In Strategy 2, the quail selects both seeds A and B. Travel time is shortened because the quail does not pass up any seeds, but total handling time necessarily increases because of the 25 seconds of handling time that seed B requires. In Strategy 3, A is passed up and only the tough seed B is selected.
Common sense suggests that Strategy 3 is the worst of the three strategies if calories are all we consider, since this strategy requires a longer handling time as well as a longer travel time. But how about the other two strategies? Strategy 1 turns out to be the optimal foraging strategy under some circumstances, and Strategy 2 is optimal in others.
For example, if seeds are spaced closely together, as in a bird feeder, travel time from one seed to the next is essentially zero and the bird would do best to pass up seed B, which takes 25 seconds to husk, and select only seed A (Strategy 1).
On the other hand, in a desert habitat where the two seed types are so far apart that it takes an hour just to get from one seed to the next, it would be foolish to pass up B just because it requires 25 seconds to husk. Under these circumstances, the best strategy is strategy 2 (that is, select both seed types A and B).
In reality, of course, birds usually select from several food items instead of two; moreover, those food items are rarely distributed randomly. Nonetheless, an optimal feeding strategy can be calculated based on a specific bird's caloric requirements, the caloric content of the food items it eats, and the travel and handling times required for the food items. In many cases, these calculated values approximate those observed in the field.
Predatory behavior would not seem to be especially puzzling or complex. A hungry hawk locates a prey species and attempts to capture it while the pursued animal tries to escape. Although variations of this cat-and-mouse game are almost limitless and depend to a large extent on the morphology of the predator and prey species (osprey versus fish, flycatcher versus insect, etc.), the basic pattern in all predator-prey interactions appears, at least superficially, to be more or less identical.
On the other hand, predator-prey interactions become more complex when we consider how much predator populations depend on prey availability. When prey is abundant, predators such as owls enjoy greater reproductive success and leave larger numbers of offspring for the following year. The increased hunting pressure exerted by more predators, however, reduces prey numbers, and in subsequent years, predator populations decline because prey populations are diminishing. When the population of predators is sufficiently reduced (from lack of prey), the prey population increases and the cycle begins again.
Although that scenario seems logical enough, evidence suggests that factors other than predation pressure can influence population cycles in prey species. For example, in some cases prey populations cyclically increase and decrease in the absence of predators, in response to disease, food supply, or other factors unrelated to predation. When we add predatory pressure to this predator-independent cycle, it is obvious that models describing the dynamics of predator-prey interactions must balance a number of ecological variables, including reproductive rate and energy requirements. Such models usually require relatively complex mathematical functions to describe the interactions between these variables.