Classical Ethology, Foraging and Sensory Ecology

Prey selectivity versus availability

Prey selectivity describes whether a predator takes a prey type more or less often than that type occurs in the environment. Availability is the proportion of each prey type in the habitat, whereas diet composition is the proportion actually eaten. Selection is judged by comparing the two. A prey type that makes up a larger share of the diet than of the environment is preferred, one taken less than its availability is taken less than expected, and one absent from the diet despite being common is avoided. Preference therefore depends on comparison with availability, not on the raw count of prey eaten. When diet and environmental proportions match, there is no preference.

In optimal foraging, the prey choice model explains why a predator may ignore some palatable prey. Prey types are ranked by profitability E_i/h_i, where E_i is energy content and h_i is handling time. A prey type enters the diet only if its profitability exceeds the overall intake rate obtained from the more profitable prey already accepted. A shorter handling time, or a higher energy content, makes a prey type more profitable. A high encounter rate with the preferred prey raises the threshold, so a less profitable prey type is excluded and the predator specialises.

For a predator choosing between specialising on prey 1 and generalising, the decision rule is \frac{E_1}{S_1+h_1} > \frac{E_2}{h_2}, where S_1 is the search time for prey 1. Search time is inversely proportional to abundance, so S_1 \propto 1/N_1. The decision therefore depends on E_1, h_1, E_2, h_2 and the abundance of prey 1. The abundance of prey 2 does not enter the condition, because prey 2 appears only through its own profitability term.

Proximate vs ultimate causation

A proximate cause explains how a behaviour is produced within an individual's lifetime. Proximate mechanisms include sensory stimuli, hormones, neural circuits, genes and developmental processes. An ultimate cause explains why the behaviour evolved, in terms of its fitness consequences such as survival, mating success or food gain. Tinbergen's four questions formalise this distinction into mechanism, development, function and evolutionary history. A practical test is to ask whether an explanation names a mechanism operating inside the individual, which is proximate, or a benefit to fitness, which is ultimate. Attraction between individuals is a mechanism that brings animals together, whereas reduced predation or more efficient foraging explains why being together is advantageous.

The same trait can be explained at both levels. A population difference in a display may be ultimately explained by female mate choice that favours a particular feature because it increases mating success. A proximate explanation of the same difference would name the genetic variant involved, the hormone that switches the display on, or the developmental effect of environment on the neural circuitry that controls it. A useful exam habit is to decide whether the answer addresses how the trait is produced or why it is favoured by selection.

Acoustic signal propagation

The transmission of sound depends strongly on frequency. In water and air, absorption and scattering generally increase with frequency, so low-frequency sound attenuates less and travels over long distances with less distortion. This makes low frequencies suitable for long-range communication such as attracting mates. High frequencies have short wavelengths, which give fine spatial resolution and strong reflection from small objects. These properties suit echolocation of prey at close range, but high frequencies are attenuated rapidly and their range is limited. Long-range communication therefore favours low frequencies, whereas high-resolution sensing favours high frequencies. Any statement that reverses these relations, claiming that high frequencies travel further or that low frequencies give high resolution, is incorrect.

The choice of signal channel depends on the environment. In the deep sea sunlight is absent, so visual displays and body colouration are ineffective, and electric fields fall off over short distances. Sound propagates very efficiently through seawater and can carry signals across very large distances. Acoustic signalling is therefore the most likely channel for long-range contact wherever light is absent or very weak.

Marginal value theorem

The marginal value theorem, formulated by Charnov, applies to a forager that exploits a patch whose food yield declines as the patch is depleted. The optimal policy is to leave the patch when the instantaneous rate of gain falls to the average rate that could be obtained across the habitat, which must include the travel time between patches. The theorem therefore predicts the optimal residence time within a patch. It does not address diet choice or the allocation of time between different activities such as foraging and reproduction, which are the concerns of other models. Its predictions include longer residence in patches that are richer and longer residence when travel between patches is costly.

• patch residence time — the time a forager spends in one patch before leaving it. Example: if the gain rate in a patch drops to the habitat average after five minutes, the forager should leave at that point.

Colour signals and aggression

A relationship between a visual trait and aggression must be tested within each morph and after controlling for confounding variables such as body size. Larger animals are often more aggressive, so a difference between colour morphs may simply reflect differences in size. Comparing morphs at equal body size, and examining the slope of aggression against size within each morph, separates the effect of colour from that of size. If aggression rises with body size in one morph but shows no relation to size in another, the relationship is morph-specific and cannot be generalised to all males. A colour signal is supported as a predictor of aggression only if it explains variation that body size does not.

Pigeon navigation via polarized light

Homing pigeons use a sun compass. The bird perceives the sun's azimuth from the polarisation pattern of skylight, and corrects this azimuth for the time of day using an internal clock, because the sun moves across the sky at a predictable rate. The clock is set by the light-dark cycle, so a bird whose clock is shifted reads the time incorrectly. Because the sun moves about 15° per hour, a 12-hour clock error produces a 180° error in the compass correction. A clock-shifted bird therefore orients in the opposite direction to the correct heading. The general principle is that errors in time-based compass correction cause systematic angular errors equal to the clock error multiplied by 15° per hour.

Imprinting

Imprinting is a rapid form of learning in which young animals form a lasting attachment to a particular object, usually a parent or conspecific, during a restricted sensitive period. Konrad Lorenz described it in birds such as geese, which follow the first suitable moving object they see after hatching. Imprinting is distinguished from other learning processes by three features: a sensitive period, rapid acquisition from a limited exposure, and in many cases persistence that is hard to reverse. Habituation, sensitisation and reinforcement are general learning processes that are not confined to a sensitive window. Example: a chick that follows a model of a hen during its first day after hatching is showing imprinting.

Sign stimuli

A sign stimulus, also called a key stimulus or releaser, is a specific feature of a stimulus configuration that reliably triggers a fixed action pattern. Tinbergen showed in stickleback fish that a single feature, the red colour on the underside, released aggressive display in territorial males even when attached to models of very different shape. This demonstrated that the specific feature, rather than the overall form, provides the releasing cue. The concept contrasts with supernormal stimuli, which are exaggerated versions of natural releasers that evoke stronger responses than the natural stimulus, and with gestalt responses, which depend on the overall configuration. Example: a red patch on a crude model can be enough to release an attack response.

Giant fibre escape circuits

Escape circuits must transmit signals very quickly so that an animal can respond before a predator completes its attack. Conduction velocity increases with axon diameter, because a larger axon has lower internal resistance to the flow of local current, so large-diameter axons conduct action potentials faster. Electrical synapses, which are gap junctions allowing direct current flow between cells, transmit signals with almost no synaptic delay. Chemical synapses, in contrast, involve transmitter release, diffusion and receptor binding, which introduces a delay of about a millisecond. A circuit with large axons and electrical synapses therefore gives the fastest escape response, whereas small axons and chemical synapses both slow transmission.

• gap junction — a direct intercellular channel that allows current to pass between two cells, producing negligible synaptic delay.

Ideal free distribution

The ideal free distribution predicts how mobile, identical individuals with perfect information settle among habitats of different resource quality. Individuals move to whichever habitat offers the highest per-capita intake, and at equilibrium no individual can improve its intake by moving. Consequently, a richer habitat holds more individuals than a poorer one, and the per-capita profitability is equalised across all occupied habitats. The total number of individuals in each habitat is not required to be equal. The model assumes free movement and perfect information, and it predicts that crowding in a rich habitat reduces its intake until it matches the poorer habitat.

Producer–scrounger foraging tactics

Producers search for food themselves, whereas scroungers exploit food discovered by others. The relative success of the two tactics depends on how common each is in the population. Scrounging shows negative frequency dependence, in which a tactic does better when it is rare and worse when it is common, because scroungers depend on producers to find food. Producer fitness therefore rises as scroungers become more frequent, while scrounger fitness falls. The two fitness curves cross at an equilibrium frequency where both tactics have equal payoff, and the population tends to stabilise there. Negative frequency dependence is therefore recognised by fitness curves with opposite slopes that cross at an equilibrium, not by constant fitness or by curves that change in the same direction.