Greg Detre
Sunday, 28 April, 2002
the evolution of morphological/physiological traits normally requires inherited genetic variations to arise by mutation/recombination upon which natural selection can act
behavioural evolution has the extra dimension of learning from parents/others
human languages provide the best example of a continuous cultural tradition which maintains different types of behaviour in different populations
cultural evolution presumably requires the ability to modify behaviour by copying + practice
e.g. primates, birds, rodents
rats:
have been shown to acquire the tendency to dig for hidden food from watching, and then become models themselves (Laland & Plotkin 1990)
Japanese macaques:
long-term observation
some of the differences in behaviour of different monkey troops are indeed of cultural origin
e.g. a washing (potatoes in a stream) sub-culture emerged from just one young female
chimpanzees:
distinctive + stereotyped pattern of grooming passed by copying
�fishing� for termites with a twig in a termite mound
different chimpanzees behaviour between Ivory Coast (more extensive monkey hunting for food, wider range of tools for foraging) and Tanzania � these differences appear to be cultural in origin (Boesch 1994)
birds:
song dialects passed down generations � very reliably for the short song of the black-capped chickadees (Ficken and Popp 1995)
differ regionally
opening of milk bottle tops by blue tits (Hinde & Fisher 1952)
appeared to be invented in three areas of London, and spread outwards apparently by imitation
but Sherry & Galef (1984) showed that such cultural transmission could occur without imitation (i.e. without one bird actually seeing another open a bottle)
they showed that chickadees (closely related to tits) learnt to open containers with lids, even if they just saw the opened container (i.e. just from the results of the behaviour)
recognition of danger:
young animals often learn what is dangerous by the responses of adults, particularly their alarm calls, in the presence of predators
European blackbirds give alarm calls to a stuffed owl or even a plastic bucket if they had heard their parents� alarm calls while looking at them (Curio et al. 1978)
ingenious experiment where adults and young were presented with different objects (of which they could only see one) simultaneously
Addo elephants � culturally transmitted fear can persist indefinitely (see early in chapter)
cultural evolution can affect genetic evolution
e.g. the Japanese macaques who learnt to wash their food ventured towards the sea, then started swimming and eating sweaweed etc.
see Bonner (1980) or Wilson (1985)
points out that when animals �know� something, they know it in the same way that computers know, as the result of unconscious mechanism (whether innate or learned), e.g.:
partridges + small rodents flattening themselves on the ground motionlessly to avoid a hawk (very sensitive to movement)
heat seeking missiles
also, that sometimes animals have too much information, and their brain can't process it all to figure out which bits are salient
stimulus = �little goad� (Latin stimulare)
e.g. alarm calls goading/prodding a mass evacuation
or a gradual accumuluation, e.g. the courtship of a male dove leading to hormonal changes in the female making her more ready to take part in nest-building
basic physical qualities (light, gravity, air, water currents etc.) orient behaviour in particular directions, e.g.:
blowfly maggots crawl directly away from light when they move out of their food source to pupate
fish rest with their heads facing into a water current
even young rats in total darkness show a �righting response� which keeps them upright with respect to gravity
transitory cues, e.g.:
Tinbergen showed that a female digger wasp orientated herself when flying back to her nest by the ring of cones placed round it, so she flew to their centre not to the nest when they were moved while she was away
and Emlen showed that indigo buntings use the pole star to orientate themselves during night flight migrations from eastern US to the Bahamas and Central America by placing them in a planetarium where he could control the image of the stars
also, whether they used the same star patterns to orient as north or south depended on their own hormonal state, which depended in turn on changing day length
or (Hoffman 1958???) that honey-bees and birds use an internal clock (that can be reset (�clock-shifted�) in a room with artificial light, confusing their navigation) in order to orient themselves relative to the sun�s changing position throughout the day
Papi (1992) showed that homing pigeons are off by about 15� per hour of clock-shift, but they are able to reorient themselves eventually by recalibrating their internal clock, using geographical landmarks, smells, sounds, magnetism etc.
the character of the world an animal inhabits is shaped by its sensory information
e.g. bees use ultra-violet light as nectar guide patterns
birds may have very conspicuous ultra-violet markers (Burkhardt 1989), which they use as a private channel of communication
or snakes see infrared radiation from warmblooded prey (Newman & Hartline 1982)
Von Frisch (1967) showed that bees can use the pattern of polarisation of light to locate the sun�s position even when it�s obscured (with a small dorsal region of the retina with a row of analysers each maximally sensitive to a different e-vector direction)
also, owl sound localisation
fish (that live in the dark caves/deep ocean) have a �lateral line organ� that picks up mechanical vibrations (little pressure sensors) (Dijkgraaf 1962)
spiders attack a tuning fork touching their web because the vibrations seem like struggling prey (even movements of as little as 1nm (Master & Markl 1981))
smell, e.g. mice � a pregnant female mouse will abort her entire litter if she smells a strange male mouse near (because they eat newly born mice that aren't theirs) (the �Bruce effect�)
they can also use smell to distinguish mice that are genetically similar/different from themselves in one particular genetic region:
known as the Major Histocompatibility Complex � a group of genes originally thought to be entirely concerned with recognition of �foreign� bodies/tissues
odour cues associated with the MHC affect behavioural (i.e. who a mouse nests with, chooses as a mate (Lenington, 1994)) as well as cellular recognition
by smelling out genetically-different mates, mice avoid in-breeding
may be the case with human females rating the smell of t-shirts � reported that pleasanter �from men most different from them in the MHC (Wedekind et al. 1995), and that the disimilar ones reminded them most of their former mates
duck-billed platypi(???)use electroreceptors to detect the electrical activity from muscle activity in crayfish + shrimps (Scheich et al. 1986)
electric fish produce their own electrical environment, identifying objects by the distortions they produce, and even as part of their social life by communicating with them (Hopking & Bass 1981)
two sub-problems (with solutions that are often mutually incompatible): detecting similarities and discriminating differences
animals focus on the key features of sensory information
e.g. dragonflies attempting to lay eggs on a car (picking out �shiny surface� as the key feature of the water they normally lay their eggs in)
in the case of important stimuli (e.g. predators), a few false alarms are a small price to pay for rapid identification
but e.g. deaf turkey hens kill most of their chicks because they never receive the auditory sign-stimulus for parental behaviour (Schledit et al. 1960)
false positives:
e.g. male mosquitos repond to the sound of their females� wings beating at a characteristic frequency, and can be attracted to a tuning fork at the same frequency
Tinbergen (1951) pointed to false positives as one of the most conspicuous characteristics of innate behaviour
e.g. male sticklebacks (that normally respond aggressively to the red colouration of rival male sticklebacks) also display to a red mail van visible through the windows of the aquarium (despite having good enough eyesite to discriminate between a fish and a mail van, but here neglecting other characteristics except the sign stimulus of red)
sometimes this selectivity is related to filtering at a low level (e.g. Narins and Capranica 1976: tree frogs� neurons of the inner ear tuned differently for each gender)
Lettvin (1959): six different sorts of frog ganglion cells:
class 1 + 2: respond to small, moving objects (i.e. prey)
small receptive fields, excitatory centre, inhibitory surround
class 4: large (square???) excitatory fields (i.e. �enemy� stimuli)
supernormal stimuli = it is often possible to build a model that produces a greater response from an animal than the natural object, e.g.:
with the herring gull, the greylag goose and the oystercatcher: the larger an egg, the more it stimulates incubation
lipstick
Arak and Enquist (1993): supernormal stimuli emerge spontaneously in computational models that rely on a few key sign stimuli
would animals that did not respond to sign + supernormal stimuli be more adaptive? three answers:
1. mistakes are so rare they don't matter
e.g. oystercatchers don't normally come across ostrich eggs, same for sticklebacks and mail vans etc.
rules of thumb are adequate and evolve easily � false positives are rare (except for ethology experiments)
2. mistakes occur, but they are not too costly
e.g. the fattest female stickleback may be full of eggs (though she might be full of worms)
when mistakes are relatively rare, they may persist � �the rare enemy� effect (Dawkins 1982)
when the animal consistently and frequently behaves in maladaptive wasy, we expect the evolution of pattern recognition systems that are more discriminating
3. they are not mistakes at all
cycle of exaggerated plumage + other supernormal stimuli being selected for:
�If responding to the biggest, brightest or most colourful male around means that a female mates with a strong healthy male and as a result has strong healthy offspring, then females with a tendency to respond to supernormal males nay be strongly selected for. Not only that, but males sporting super-normal plumage may be favoured, leading to an evolutionary exaggeration of both male plumage and female preference� (see Fisher (1958))
supernormal stimuli may be the key to the evolution of the large and exaggerated signals that are typical of much animal communication
brood parasites, e.g. young cuckoo pushes out reed warbler�s entire brood so that they only feed it
each individual female cuckoo specialises in just one species of host (different strains = �gentes�)
swap their one egg for one real egg, and the young cuckoo hatches quicker than the others
only ten host species are parasitised (very small proportion of total species) � evolutionary arms race, especially at the egg-recognition stage (Davies and Brooke 1989 showed that spotted flycatchers and reed buntings (which currently don't get parasitised) were very good discriminators of model cuckoo eggs, whereas meadow pipits were much less discriminating but don't get parasitised because their diet/nest site aren't appropriate for cuckoos)
sensory systems are adapted to pick out �information-rich� parts of the environment (Barlow 1972)
e.g. representing a complex scene with a quick line drawing (because lines + edges carry a great deal of information)
generalised feature detectors (e.g. line detectors) are very useful when put together to identify more complex objects at a a later stage of analysis
considers the compound eye of the horseshoe crab (limulus) and how the ommatidia (i.e. clusters of 10+ retinal cells) use lateral inhibition to accentuate contrast (Hartline et al. 1956)
Barlow (1972) postulated whole hierarchy of processes, leading up to cingle cells that responded to very specific sets of features (a �grandmother cell� that only responds when you look at your grandmother)
e.g. neurons in the temporal lobe of rhesus monkeys that fire up to 20 times more rapidly when looking at a face (monkey or human) than for lines, gratings or non-face complex objects
they can be referred to as �face-detecting� neurons (Baylis et al. 1985), and even for identifying individuals
but this identification information is distributed (Rolls 1994)
moved away from this idea of serial processing up separate pathways leading to individual end-point cells
now we think in terms of multiple parallel pathways
e.g. separate pathways in mammals for movement + form (Zeki 1993)
signals = conspicuous behaviour patterns, often combined with structures like plumes/crests that have been specially evolved to affect the behaviour of another animal
e.g. the single large claw of the male fiddler crab � larger than needed for feeding or other non-signal purposes � it�s brightly coloured and waved rhythmically in the air at females + rival males as the crab stands by its burrow, suggesting that it has been evolved to be a conspicuous signal to other crabs
communication = when one animal responds to the signal sent out by another animal
this can be a problematic definition though
e.g. a foraging ant lays a scent trail when its find food which its nest mates follow
but so does a small snake, which uses the trail to find the ants� nest and devour them
we would say that the scent trail is a signal used for communication by the ants, but although the snake undoubtedly obtains information from it, we wouldn't want to say that the ants communicated with it (Burkhardt, 1970)
it�s problematic not because it�s interspecific communication
e.g. flowers� nectar guide patterns for bees, or warning/startle colouration of prey animals (e.g. eye-spot patterns of moths)
but because the ants derive no benefit from the response of the snake, whereas the above flowers + moths both benefit from their communication
different definitions of communication then:
Altmann (1962): social communication = a �process by which the behaviour of an individual affects the behaviour of others�
Hinde + Rowell (1962): classified as visual signals only those that were likely to have evolved for this purpose
it can be difficult though to decide whether a signal is �specially evolved�
various factors can be involved in whether a putative communication signal affects behaviour in the usual way (e.g. the size of the black bibs in sparrows� pg 156):
where the encounter takes place (Wilson 1992)
how hungry the two animals are (Moller 1987)
already-established social hierarchy/the history of the interactions
e.g. van Rhijn (1980) saw that younger jays tend to give way to more powerful older members, even with only a minimal �signal� (e.g. displacing younger ones from food just by looking at them) once the rank order had been established (although they would escalate into a recognisable threat display if the minimal signals were ineffective)
meta-communication � does not communicate information itself, so much as qualify other signals that follow it
e.g. play situations with carnivores and monkeys
an adult male lion invites a cub to play by making a unique bow with forequarters lowered
same with dogs + coyotes, bowing + wagging their tails (Darwin 1872, �The expression of the emotions in man and animals�)
monkeys adopt a �play face�
this signal for playful aggressiveness must be necessary to avoid real fights for species where games are an important part of behavioural development
different animals use different types of signal (e.g. songs, visual displays, scent trails, posture changes)
there are at least three different selection pressures operating on animal signals:
1. what travels best in the environment between signaller and receiver
2. what best stimulates the sense organs the brain of the receiver
3. how far the interests of the sender and receiver coincide
there is also the issue of conspicuous signals alerting predators
much of the diversity of signals results from their diff ways of responding to external stimuli, e.g.:
pond skaters rest on the water using feet (tarsi) that repel the water, and that they also use to sense ripples from struggling insects (potential prey), and to communicate with one another (genders use different frequencies) (Wilcox 1979)
blind termites that work in subterranean tunnels rely on tactile communication and scents detected by chemoreceptors
chimpanzees touch and kiss each other�s hands in gestures of reconciliation (de Waal 1996)
visual signals:
fireflies manufacture their own light (since even bright moonlight is only 1/10 as strong as daylight)
blue travels better in seawater, so many reef fish are blue or yellow (Lythgoe 1979)
long distance visual signals (e.g. the white tails of rabbits + some deer) � white (broad range of frequencies) is very conspicuous at all distances
sound signals:
high frequencies attenuate and become scattered by obstacles more easily than low frequencies
e.g. 100 kHz bat echolocation drops to 1/50,000 of its intensity after 4-5cm
so male tree frogs use two different frequencies
getting higher up also helps
e.g. crickets that sing from trees/shrubs can spread their signal over 14 times the area and better attract mates (Paul & Walker 1979)
sounds travels better in water
e.g. Payne & McVay (1971): humpback whales may communicate over distances of hundreds+ miles
chemical signals:
especially in insects + mammals
short-lived/localised signals, e.g.:
ant signals are deliberately volatile (< 1 minute) and disperse quickly (over distances >5m)
moths release sex-attractant pheromones at 1-second intervals (Dusenberry 1992)
persistent signals:
territory markers (Wilson 1965), high molecular weight
e.g. hyenas smear grass stems with past from their sub-caudal scent glands and deposit faeces at latrines, either all round, or just in strategic places if the borders are too big (Mills & Gorman 1987)
in order to be maximally effective at stimulating the brain of the receiver, it�s worth looking at sensory processing in receivers
e.g. female jumping spiders choose their mate initially by whether he moves in a way that resembles food � the males appear to mimick prey in order to catch their attention (Clark & Uetz 1992)
this + other behaviours exploit mechanisms already present in the receiver for non-signal functions (e.g. general sensitivity to motion)
why should female long-tailed widowbirds prefer supernormal male ornaments (Andersson 1982)?
(Darwin, Sexual selection and the descent of man 1871)
using as a sign stimulus
but we still need to know why it has resulted in a self-fuelling circle and persists now � pg 170 ???
e.g.:
two males fighting over a female, both of whom will benefit if the other fled without a fight, so potentially each would gain if it could convince the other that it was stronger
potential cheating vs resistance to being cheated
contact calls between memebers of a group of birds or primates, where the interests of signaller and receiver coincide, so there�s little scope for cheating
the extent to which signaller and receiver�s interests coincide is an important factor in how conspicuous the signals are (Dawkins & Krebs 1978), e.g.:
red deer (see below)
gazelles stotting (see below)
on the other hand, cooperating animals communicate with �conspiratorial whispers� (Krebs & Dawkins 1984)
conflict between signaller + receiver leads to selection pressure for signals to be �honest�, i.e. proof against cheating/bluffing (Zahavi 1987)
how can signals be made honest/guarantee to the receiver that the opponent is not cheating?
e.g.:
red deer (Clutton-Brock & Albon 1979)
males (stags) spend several weeks a year defending/fighting to capture groups of females
loud roars/bellows before/instead of fighting
if the challenger is �out-roared�, he may give up
if both roar at the same rate, fighting is likely
the roars appear to be a reliable (i.e. honest) indicator of fighting ability, since roaring + fighting are both exhausting and use the same set of thoracic muscles
it�s reliable because it�s a loud signal that only good fighters can manage
gazelles (Fitzgibbon & Fanshawe 1988)
run away from wild dogs signal with conspicuous jumps or �stotts�
stotting is done most by animals in good physical condition and so are likely to be able to outrun wild dogs in a long chase
consequently, the wild dogs ignore the ones that stott at the highest rate
skylarks sing when being chased by merlins (Cresswell 1994)
singing loudly and continuously (territorial song) seems a risky strategy when being chased by a falcon
young canaries� begging signal (wide open mouth)
their mouths� red lining is due to the suffusion of blood � parents feed the reddest mouths, as their means of deciding whose need is greatest (Kilner & Johnstone 1997) � after being fed, the blood is diverted to the gut, and the mouth gets paler
e.g. plovers lure predators away from their nests by feigning a broken wing
if the receiver is only rarely deceived, on average it will be beneficial to respond to the signal
the predator that avoids black + yellow wasps will sometimes lose out by being deceived into avoiding a perfectly palatable hoverfly, but on average it gains from avoding black + yellow prey because so many of them are dangerous/distasteful
mimics gain an advantage only as long as they�re rare � once they become common, predators that ignore that signal and attempt to eat them will be at an advantage
how deliberate is such deceit?
Ristau (1991) argues that plovers giving a broken wing display monitor the behaviour of the predator (e.g. fox) and adjust their behaviour according to whether the fox is following them or still heading for the nest
Munn (1986) describes two species of insect-eating birds that appeared to deceive their flockmates by falsely giving alarm calls and so gaining unrestricted access to food
Byrne & Whiten (1988) consider possibly deliberate baboon behaviour � although anecdotal, they argue there are enough examples to support the idea that some animals can be deliberately dishonest
e.g. a subordinate being attacked giving the appearance of spotting a predator in the distance, and then using the momentary distraction to escape
von Frisch, �The dance language and orientation of bees�, 1967
once a single bee found a sugar solution (after perhaps a few hours), lots would follow within minutes � the information had been somehow passed on
he marked the foragers who found the sugar, then watched their behaviour when they returned to the hive, using glass observation hives
when the sugar is <50m from the hive (round dance):
rapid dance for up to 30 secs
roughly circular path just over her body�s length in diameter
moves in circles alternative to the left and to the right
stays in approximately the same place on the comb
other foragers face the dancer, follow her movements closely
they then leave the hive and search nearby
possibly also olfactor cues
either from the food source itself on the forager�s body
or if it�s odourless, from her marking it by opening the Nasanoff scent gland on her abdomen as she drinks
ants + termites have similar alerting displays + pheromones for organising foraging activity towards nearby new food sources
but the honeybee dance becomes more exceptional for further food sources
when the food source is discovered >100m (up to 5km) from the hive (waggle dance):
a short straight run becomes incorporated between the turns, and on this run the dancer wagged its abdomen rapidly from side to side
bursts of high-pitched sound (Esch et al. 1965)
the dance followers read back information, following every move
contains information about the distance + direction of the food source
distance:
tempo falls off with distance
duration of dance (and less importantly (Michelsen et al. 1992), number of waggles and duration of sound pulses all increase with distance)
bees fly at a constant speed in still air, so distance is measured as outwards flight time (i.e. further for upwind)
direction:
the little south Asian bee builds a single vertical comb on which they perform the waggle dance � they point towards the direction of the food source
sometimes the honey bee will similarly dance on the flat landing board provided at the entrance to most types of hive
usually though, honey bees use the vertical face of the comb (as von Frisch saw them) � he noticed that:
the average direction of the waggle run stayed consistent within the dance
and was the same for all the foragers who danced after feeding at the same food source
its mean direction shifted by about 15�/hour because it relates to the apparent movement of the sun
they honey bee �convention� is to take vertically upwards to represent the present position of the sun
thus, like us, a humble insect can convey information in a symbolic fashion
(early) controversy over von Frisch�s results:
he used scent plates (only one with food) placed in a semi-circle and a line to test that direction + distance information were being conveyed
because scent + wind direction were ignored in the early tests, it could be that the dance merely alerted and random search did the rest
which means that information about distance + direction existed in the dance but was not communicated � an intuitive but not a secure argument
e.g. flies who�ve found + eaten a food source � their dance conveys information to a human observer, but not to other flies (Dethier 1957)
Gould (1975) summarised + proved von Frisch�s conclusions
Michelson et al.�s (1989) artificial model bee has ended the controversy
it follows the waggle dance path, and crucially has a vibrating artificial wing to produce an acoustic field like a real dancer
seeing how sound + waggling relate is complicated
the model is problematic because: it doesn't attract other bees as well as a real dancer, requiring the experimented to be repeated a lot
the duration + frequency of dancing by a returning forager is directly related to the concentration/abundance of the nectar/pollen she has found, and attracts a proportionate number of recruits
�tremble dance�:
lasts up to 30 mins
jerky, shaking bnodies, frequent brief pauses, abrupt changes in direction
a sort of �anti-recruitment� signal (Seeley 1992), for when there�s too much food and the forager can't find a house bee to accept her regurgitated food or a pollen storage cell with enough space
leasd to a sharp reduction in the amount of dancing by foragers who encounter the tremble dance, and maybe a switch to brood care, thus reducing the incoming surplus of food
vervet monkeys (Africa)
elegant + group-living
Cheyney & Seyfarth (1980, 1990):
got very close to the monkeys, documented interactions between different members, and were able to experimentally manipulate their communication system by playing sound recordings back from hidden loudspeakers
different alarm calls when they saw different predators
this is not unique, e.g. chickens + ground squirrels have different alarm calls for aerial vs ground predators
but vervets� predator-specific calls are particularly well-developed
when a vervet sees:
a leopard:
loud barking alarm call
others run up into trees, where they can escape to branches which the leopard is too heavy for
an eagle (martial or crowned):
the call is a sort of double-syllable cough
eagles can attack monkeys on the ground + trees
so they look up into the air (or come down from the tree) and run into the thickest bush available
a snake (python, mamba or cobra)
the call is a sort of �chutter�
snakes hunt for monkeys hiding in the grass
the monkeys stand up on their hind legs and look down into the grass
when several have spotted it, they often approach + mob it
because the monkeys respond in kind to a loudspeaker playback, it must be the signal that carries the information
young vervets are less specific than adults in their signals (Cheney & Seyfarth 1980, 1986)
e.g. giving alarm calls to warthogs, pigeons and other non-dangerous animals
but the calls aren't random, e.g. they give the eagle call for something overhead (e.g. pigeon, falling leaf) etc.
no evidence that the adults actively teach the younger ones to be more accurate in their alarm calls
rather, if a young monkey gives a call the adults look for the predator and either carry on regardless or give the alarm call themselves (Seyfarth & Cheney 1986)
vervets also have more subtle ways of communicating
they grunt (harsh raspy noise) in social situations (usually when alert but relaxed
e.g. when aproaching a social dominant individual, a social inferior, when crossing an open space, and when they�ve seen another group of monkeys
they found a consistent difference for two of them when these were played back on loudspeakers
the subtle (indistinguishable to humans) spectral differences between the grunts are the opposite to ritualisation (where signals become large + distinctive to be understood by the receiver)
vervet grunts are a good example of cooperating animals� �conspiratorial whispers�
they have a common interest in staying together as a group, partly because they may be related, and partly out of mutual benefits (e.g. protection from predators)
habituation is a simple form of learnin because it involves the waning of a response that is already there
associative learning � acquiring new responses/capacities
a previously neutral stimulus/action has sufficiently important consequences to be singled out from other such events
after some repetitions followed by the same consequences, a long-term association is built up between the event and its result and the animal�s response changes accordingly
e.g.:
having found sugar solution there previously, a honey-bee picks out the blue dish from an array of dishes laid out by an experimenter
rodents exploring new territories quickly learn the shortest routes to shelter for use when hawk/owl swoops
Pavlov�s dog:
had learned to respond to a new stimulus (the metronome/bell), previously neutral, which Pavlov called the conditioned stimulus
the salivation response to the CS is the conditioned response
prior to learning, only the meat powder or unconditioned stimulus had produced salivation as an unconditioned response
the conditioned response is generalised to similar stimuli, but to a lesser extent � however, the discrimination becomes more fine-grained after repeated trials
use the conditioned discrimination method to measure the sensory capacity of animals
train the animal to a stimulus, then present it with another similar stimulus (which perhaps gets associated with a slight punishment) � the two stimuli are made increasingly similar until the animal can no longer learn to discriminate between them, e.g.:
the colour sensitivity of bees (von Frisch 1967)
touch sensitivity of the octopus (Wells 1962)
chemical senses of fish (Bull 1957)
because predators generalise, e.g. birds avoid the evil-tasting black and yellow caterpillars of the cinnebar moth after a few trials, and it is advantageous for different distasteful insects to resemble one another (M�an mimicry) � each derives a certain degree of protection from bad experiences predators may have had with other similarly-coloured prey
trial and error = learning to eliminate behaviour which led to no reward and increase the frequency of behaviour that is rewarded = instrumental conditioning = operant conditioning (Skinner)
classical conditioning: associates a novel stimulus with a response (the UCR) which was there from the outset
vs
instrumental conditioning/trial and error: it is a novel response (not a novel stimulus) which is learnt
however, animals do not simply associate response and reinforcement � instinctive behaviours + biases restrict and shape learning
� pg 269
Skinner found that, in general, a delay of more than about 8 seconds between a response like bar-pressing and its reinforcement greatly slowed learning
secondary reinforcement, e.g. a rat learns that a reward is delivered when a light comes on in the Skinnerian box, then it will learn to press the bar to switch the light on
discusses rats� inbuilt readiness to associate sickness with a new food � pg 272
extinction, spontaneous recovery, disinhibition (if a new stimulus is presented with the CS after extinction)
Skinnerian box experiments appear to take much longer than classical conditioning to become extinct
drive-reduction
has to posit anxiety-reduction to explain avoidance conditioning (e.g. because once a rat has learnt the shuttle-box problem, it gets no more conventional reinforcement (i.e. foot-shock)
also has problems with exploratory/latent learning
Menzel & Erber (1978): honey-bees learn rapidly and retain very well the effects of even a single association between a colour and food reward
some animals store food from a rich food source for later, either in a single larder, or scattered in caches around their territory (Sherry 1985), e.g. the marsh tit or Clark�s nutcracker
food-storers appear to have more neurons in the hippocampus (involved with memory formation and spatial memory)
� pg 279
Thorpe (1963) problematically considered insight learning in animals: �the sudden production of a new adaptive response not arrived at by trial behaviour� or �the solution of a problem by the sudden adaptive reorganisation of experience�, i.e. a flash of insight
rat maze-learning:
Hull: the rat chains together a series of S-R associations, leading from one to the other through the maze
Tolman (1932): rats build up a mental picture/cognitive map of the whole maze during exploration + learning
but the fact that some rats� are able to change their path (�detour studies�) very rapidly (i.e. take short cuts, I think), supports Tolman
pigeons presented with a number of photographs above buttons (Herrnstein et al. 1976)
some pictures shared a feature in common (tree, person, water etc.), but these pairs of features could be very different (e.g. a single person + a crowd), but the pigeons could still connect them
they were also very good at manipulating/comparing abstract shapes (as we use in IQ tests)
but both of these could be explained in terms of advanced sensory processing (because birds use a lot of visual stimuli, viewed from various perspectives), rather than �intellect� or high-level concepts
K�r (1927) on chimpanzees:
showed that they can overcome increasing hurdles to get their bananas
e.g. pile up boxes to stand on, or fit two sticks together to pull them down
they seemed to benefit from previous experience of playing with sticks and boxes, and then apply this knowledge
correlation between brain development and learning ability:
birds have been underestimated because although they have relatively large brains, those parts homologous with the mammalian cerebral cortex are small
e.g. Pepperberg (1990, 1991) on the African gray parrot (Alex)
nine words for objects (e.g. wood, cork, paper) � only nine???
counts to six
dientify the quality (i.e. shape, colour) common to a group of diverse objects
difficulties of comparative study � pg 288
learning set = learning not just a problem, but something about the principle behind it, and then can steadily increase its learning speed when given a series of similar problems
e.g. Harlow (1949) � a monkey is faced with a pair of objects, one with a reward, one without � it learns which is which � then a new pair of objects, same rules � after 100 or so tests, the monkey simply lifts up one, and then selects that one ever after if it had the reward
Warren (1965): all vertebrates except fish have have the ability to form learning sets
e.g.:
Wallace, orangutang in a cage, spread birdseed nearby then was able to reach out and grab a chicken
a sheepdog faced with a stubborn ewe isolated the rest, brought over a few sheep nearby, then it joined that group and was happy to be herded towards the main pack
training honey-bees to forage at a dish distant from the hive, by moving the dish a few centimetres away from the hive gradually, then bigger and bigger steps, and then eventually the bees are waiting for it in its next position
could be a manifestation of an inborn ability, e.g. as the shadows of a hill move out with the sun, a series of flowers are warmed step by step and open + secrete nectar
Humphrey (1976): the complexities of social life, especially in the primates, provided one of the most important selective forces behind the emergence of thinking ability
Povinelli et al. (1992):
experiment that suggested that chimpanzees have the ability to understand the outcome of their own actions, the way that these affect another individual, and also to put themselves effectively into that individual�s place
two chimps on either side of the apparatus � two food trays inside � the experimenter shows the informant chimp which handle to pull (I think) to get the food, who has to then gesture the information to the operator chimp
then, when their roles were swapped over, they found that the second chimp knew exactly what to do immediately
arguably, this isn't simply a case of mere copying of a motor pattern from one to another because the viewpoint is so different
the operator chimp has to appreciate the significance of the informant�s gestures and, after the swap, take on this gestural role � it does so spontaneously, and using its own gestures which are not necessarily the same as those it had seen used before by its partner
Premack & Woodruff (1978):
chimp was shown a video of a human being using objects well-known to the chimp, and in need of something (e.g. shivering with cold with an unlit stove, or trying to open a locked door)
then the chimp was shown photographs of objects, one of which offered a solution (e.g. the burning wick of the stove, a key)
one chimp chose the �correct� picture 7/8 times
fascinatingly, it chose �neutral� photographs much more commonly of people it disliked
chimp language:
see Kennedy (1992) for skepticism about attempts to teach chimps symbolic language
there is no disagreement regarding chimps� ability to learn large numbers of associations between signs + symbols representing objects/conditions (e.g. �more�, �open�)
the dispute is about whether they can combine them in �grammatical� ways or in novel combinations
critics point to unintentional cuing (e.g. Clever Hans)
bonobos:
Kanzi could follow verbal instructions through earphones or via loudspeaker (so eliminating all verbal cues) (Savage-Rumbaugh et al. 1996)
e.g. �Kanzi, put the hat in the refrigerator�
bonobo groups in the wild (Congo basin): social behaviour patterns to keep together moving through dense vegetation in their forest habitat
they often break into smaller parties which gather again during the night
foraging parties seem to deliberating mark vegetation along trails which can indicate their path to others (e.g. sticks stuck in the ground, leaves pressed down far more than necessary for passage)
elephants:
large brains, complex social life, longevity comparable with ours
perhaps significant infrasonic (8-10Hz) communication between them (Poole et al. 1988)
optimality models often focus on the behaviour of individual animals without taking into account the effect of what other animals are doing
e.g. deciding where to forage, whether to act aggressively
evolutionarily stable strategy (Maynard Smith 1982):
a strategy = a specification of what an animal does (e.g. always attack if challenged)
an ESS is a strategy that, if adopted by most members of the population, cannot be bettered by any other strategy
e.g. aggressive behaviour (Maynard Smith & Price 1973) � 2 strategies:
1. Hawk = always attack
2. Dove = always retreat
in a population of Hawks and Doves, there will be four kinds of encounters
whether to follow a Hawk or Dove strategy will depend on how many other Hawks and Doves there are in the population
in a Dove only population, Hawks will be at a huge advantage, but their numbers will grow, and so the Hawk strategy will become less advantageous because they will get injured a lot, then the Dove strategy becomes the better option
the ESS is a Mixed Evolutionarily Stable State, comprised of a mix of both Hawks and Doves
sometimes though, the ESS can be a single one, e.g. �attack if opponent is smaller, retreat if opponent is bigger� (�conditional�, (Maynard Smith & Parker 1976))
unless a strategy can hold its own in interactions with other strategies, it will not persist in evolutionary time
success in evolutionary terms = leaving offspring that themeselves reproduce
as far as a gene is concerned, the body it happens to be in at a given moment is a useful, if temporary, vehicle for getting it passed on into the next generation (R Dawkins 1989)
e.g. parental care as the result of strategies genes have for perpetuating themselves
similarly, a genetic tendency to help a sister reproduce could be favoured by natural selection because the sister, being so closely related, has a high chance of having the same genetic tendency (W. D. Hamilton 1964)
kin selection (Maynard Smith 1964) = selection which takes account of other relatives as well as immediate descendants
helping a relative to reproduce will only be favoured (genes for it (i.e. genes which, given a suitable environment in which to operate, can bias behavioural development along particular lines) will only be spread) if the benefit (the increase in reproductive chances as a result of the help) more than makes up for the cost (the decrease in reproduction the helper incurs as a result of its action)
Hamilton: rb � c > 0
r = coefficient of relatedness
b = benefit
c = cost of the relative-helping genotype
now DNA fingerprinting allows us to calculate r easily
(this is possible because throughout the genomes of many species, there are regions of DNA in which the patterns of base pairs are repeated over and over again but the repeat patterns are slightly different in different individuals (Queller et al. 1993)
e.g. starlings, which have always thought (like most birds) to be monogamous � one brood of chicks was fathered by three different males (Pinxthen et al. 1993)
helping a large number of distant relatives could be as beneficial �as helping a direct offspring
Haldane: �I am prepared to lay down my life on behalf of four grandchildren or eight first cousins�
it is more difficult to calculate
b and c
examples of different animals� eusociality + kin selection:
Hymenoptera:
haplo-diploidy
Hamilton considered the Isoptera (termites) and Hymenoptera (ants, bees and wasps) � show extreme altruistic/helping behaviour
usually just one reproductive female (the queen) and large numbers of sterile workers
workers:
solely females in ants, bees and wasps
perform all the tasks of the society, e.g. foraging, rearing young, nest construction + defence, and do not reproduce themselves
this form of self-sacrifice (sterile workers)
appears to have arisen independently at least 11 times in the Hymenoptera
(which must be somehow more predisposed towards it than other species
males are haploid, and develop from unfertilised eggs
females are diploid, and develop in the normal fashion from fertilised eggs
�All the sperm from one male are therefore identical � a simple copy of the male�s own haploid chromosal set. When the queen bee fertilises eggs with this sperm, all the resulting daughters receive the same paternal chromosomes and so have half their genes in common (the half donated by their common father). In addition, they share, on average, half the genes inherited from their common mother and so their degree of relatedness r = 0.75�
�This means that although the workers are sterile themselves, many of the genes that they share with the young queens (which are also their sisters) will be passed on to the next generation� in gene terms, it is more advantageous for a female Hymenopteran to stay and help to rear her closely related reproductive sisters than to leave and attempt to rear less closely related daughters of her own
termites:
are also eusocial but have an ordinary diploid mating sys
sterile workers of both sexes have a degree of relatedness to the young reproductives of only 0.5
king and queen termites are long-lived and monogamous
queen may lay up to 36,000 eggs/day, and live up to 70 years
presumably, even a relatively low r combined with high benefits and low costs would lead to helping behaviour
the cost to each worker of being sterile is the loss of those offspring it would have had it if had not been helping the colony � but a pair of termites alone would probably not survive (outside their termite-built microenvironment in the desert), so the costs are low
the monogamous breeding means that the workers are guaranteed a long series of full brothers + sisters to take care of (who would die without their care), i.e. a large benefit to pretty close relatives
naked mole rats (Jarvis 1981):
termite-like social system
� pg 342
average relatedness 0.81 (monogamy plus in-breeding)
wouldn't survive on their own or in pairs
tips the cost-benefit-relatedness analysis in favour of self-sacrificing, life-long worker sterility
Damaraland mole rat (Jarvis et al. 1994):
eusocial, non-breeding workers
much smaller colonies (perhaps because the soils are softer, and so don't require so many individuals to dig them)
their cost-benefit equation is still tipped in favour of some worker sterility, but less strongly � the workers sometimes leave the colony and found their own where at least some of them become reproductive
birds:
in > 200 species of bird, the parents are helped in some way by other individuals, often their own young from previous years
though none that we know of are totally sterile
most birds not only nest in the same territory year after year but are mainly monogamous (so the relatedness is fairly high)
furthermore, helpers can significantly benefit their siblings (nest-building, territorial defence and feeding), e.g.:
white-fronted bee-eaters (Emlen & Wrege 1989) � on average, each helper enables the parents to raise half an extra chick
Florida scrub jays (Woolfendon & Fitzpatrick 1984):
helpers gain by having more siblings
they tend to take over their parents� territory in the future (at the cost of deferring reproduction for a year or so, but nest sites are highly in demand)
kin selection builds upon natural selection:
natural selection: measure success/fitness in terms of the number of offspring reared to reproductive age
with kin selection: success/fitness takes into account (as well as the number of offspring produced by an individual)
also the effects of its behaviour on how many offspring its relatives have
and how many it does not have itself as a result of its helping
Hamilton (1964) �inclusive fitness� = calculate the conditions under which a gene might spread, taking into account the effects that bearers of that gene might have on different sorts of relative
inclusive fitness is not just as weighted sum of all the offspring and other relatives an individual has, taking into account their relatedness
only if the animal has in some way helped its sister to survive/reproduce should the sister be included in this way
phenotype matching = using its experiences of its siblings/parents/self to help recognise unfamiliar animals that look/smell/sound similar as related � pg 347
(�armpit effect� = self-referent phenotype matching, refers to the process animals are said to use to compare other animals' odors to their own in order to distinguish strangers from unfamiliar family)
Belding�s ground squirrels (Holmes & Sherman 1982)
see essay/notes, also �Squirrel My Family Stinks.htm�
bees can discriminate full + half-sisters fairly accurately
in the sweat bee, the willingness of the guard bee to let in other bees was directly correlated with their relatedness (Greenberg 1979)
this appear to be based on chemical odour cues
behavioural ecology (an offshoot of ethology) �has tended to concentrate on adaptive questions about why animals are social and, at least until recently, has neglected those to do with the causation + development of their social behaviour�
female elephants may live in the same family unit for 40 or 50 years � �societies�
group living is often so beneficial that animals on their own are usually not around to make the comparison possible
use experimental evidence (artificially creating groups or placing individuals on their own)
look at naturally occurring variation (within a species, some animals may be more/less likely tog roup than others or different types of grouping)
compare between species that have adopted solitary/social lifestyles
Allee et al. (1938) showed that water fleas cannot survive in alkaline water, but the respired CO2 of a large group could be sufficient to bring down the alkalinity
protection against predators
more animals on the alert, so predators are less likely to remain undetected
Lazarus (1979) showed that red-billed weaverbirds were much more likely to see + respond to a goshawk overhead in groups of two or more
Elgar (1989) reviewed 50 studies, which showed that birds + mammals spend less time in vigilance and more time in feeding in bigger groups (but this may be because big groups clump around rich food sources where feeding is fastest)
Saino (1994) showed this effect with carrion crows, independently of the amount of food present
is cheating a problem in such flocks?
there are advantages to spotting the predator directly rather than relying on the response of other birds
adult meerkats (socially living mongooses) take turns �baby-sitting� the nest-hold (watchign for predators from a high look-out point) while the others are away foraging (McDonald 1986)
meerkats live in close-knit communities and know each other as individuals, so presumably there�s not much cheating
Hamilton (1971) showed that if each animal in the group attempted to put at least one animal between itself and the predators then tight formations would be the inevitable result
also, the group can attack the predator
Krebs et al. (1972) showed that when one member of a tit flock finds a food item the others rapidly alter their searching strategies (concentrate their attention on the same general ara and the type of niche in the trees where the food was found)
sometimes hunting in groups gives each individual a better chance, e.g. black-headed gull group fishing (G�ark et al. (1986))
hyenas seem to set out in different sized packs, and then go after specific prey (Kruuk 1972)
disadvantages of social advantages, e.g.:
increased competition for food
increased risk of disease transmission
greater conspicuousness to predators
greater risks of cuckoldry, mixing and cannibalism of young
size, complexity, types of interaction, sex ratio, degree of differentiation into roles, relationships with other specific individuals in the group, kinship etc.
eusociality = characterised by a reproductive division of labour, where some members of the group lose their reproductive capacity altogether and become members of a worker �caste�
caste = rigid, limited role in society largely determined by upbringing
e.g. �based on what (almost) equipotential bee larvae are fed, or controlled in ants by pheremones secreted by king + queen
though sterile themselves, they help the colony as a whole
associated with a high degree of relatedness between members of different castes and overlap of generations
e.g. termites, ants, bees and wasps
(�replete� workers, hanging permanently, protected and immobile, storing liquid food as living honey casks (H�obler & Wilson 1990))
also (Jarvis 1981) in the naked mole rat
and the snapping shrimp (Duffy 1996)
lives in sponges on tropical reefs, in colonies of <300, but only one reproductive female (the �queen�)
like termites and mole rats, the shrimps are diploid and the �workers� are full sibs
food seems to be abundant, so the workers are mainly for defence, because sponge �homes� are sought by various (larger) species on the crowded reef
to a lesser extent, various young birds + mammals stay in their parents� nest for a year or two
e.g. black-backed jackals (Moehlman 1979) and dwarf mongooses (Creel et al. 1993)
two essential elements of eusociality (in insects):
1. overlap of generations
2. cooperative brood care
e.g. the aggregations formed by cockroaches and earwigs
for the first � of a cockroach�s year-ish life, while developing through a series of nymphal stages, they all live together in a loose aggregation near sources of food + shelter
Wilson (1971) surveys the literature on termites
building complex nests, growing their own food, taking slaves from other nests and communal defence of the nest
discusses individual honey bee lifestyle in detail� pg 384
insects show remarkably consistent social organisation
the social organisation of vertebrates is not rigidly species-specific (in contrast to insects)
some (e.g. the great tit) are gregarious at some parts of the year, and strongly territorial at other parts (e.g. breeding season)
territorial boundaries can be defined by song, visual display, scent posts marked with urine, special glandular secretions or faeces
�economic defendability� (Brown 1969): animals should only go the time/troubel of defending a terriotory if the resource is worth defending and defensible
the energy balance can be very finely tuned
in different species, the male�s part in raising offspring varies
it used to be thought that there was a very close correlation between the amount of care given by the two parents, e.g.:
monogamy existing where both parents cooperated to care for the young
polygamy was associated with one parent giving much more parental care than the other
also polygyny and polyandry
DNA fingerprinting has demonstrated there to be much more multiple mating than previously believed
e.g. starlings � both parents care for the young, but various males may be the fathers of one brood
females often actively solicit copulations from males (Birkhead & Moller 1992)
especially from higher-quality mates than their own (Kempenaers et al. 1992)
multiple mating is not in the interests of males, e.g. male dragonflies� penises have elaborate hooks to try and scoop out the sperm from other mates stored by the female when mating (Waage 1979)
in a parasitic worm, the male seals up the female after he has copulated with her (� pg 393)
in mammals, the retention of the young inside the female�s body means that the male can often do little to improve the survival chances of his offspring, so he�s best off being polygynous
(except that presumably he could protect the female, and more complex animals need longer to grow up in the world, during which time they�re helpless, surely???)
on the other hand, young birds need lots of labour-intensive feeding, which can be done by males or females, perhaps hence 90% of birds species being social monogamous
it�s a lot to do with defence as well � the male may be able to defend more than one female
huge male elephant seals dominate stretches of beach and fight over the females � in some seasons, 4% of the males are responsible for 85% of the matings (Le Boeuf & Peterson 1969)
few males are successful for more than one or two seasons
it may be that the reliability of signals (in place of fighting) can only come through the use of signals that are themselves costly or at least difficult for any but the most vigorous and physically strong animals to produce (Zahavi)
besides very costly signals like roaring, there are other ways to deide whether or not to fight:
1. if the same animals interact frequently enough, they can learn each other�s true fighting ability (dominance hierarchy/pecking order)
e.g. domestic fowl (Schjederup-Ebbe 1935)
linear hierarchy � lots of fighting to begin, but once established, subordinates almost always defer to a more dominant bird
in primates, the hierarchies are often tangled rather than linear, and involve alliances
dominance hierarches can become unstable over time as the relative fighting abilities of different members shifts
this may be why stags need a costly assessment signal, because a stag defending females gets very tired and hungry, and so the relative abilities between stags is constantly changing
2. if individual recognition isn't possible (e.g. there are too many animals), the dominance hierarchy is based on �badges of status�
e.g. the black bibs of male house sparrows
badges of status do not appear to be costly, so why aren't they vulnerable to cheating?
it seems that if a weak animal has a dominant badge, it gets constantly challenged by other dominant individuals (Moller 1987), and so pays a cost for its deception (which may be enough to keep the number of cheats pretty small)
wolves� behaviour doesn't change much throughout the year � they hunt in packs and maintain a very stable social structure based on an extended family unit, usually with one dominant male leader and perhaps some other adult males
from very early in their history, the majority of primates moved round in stable groups
lemurs (Jolly 1996):
in small mixed groups (12-20) have several adult males + females
dominance hierarchy, permanent, cohesive unit
group territories with stable boundaries (marked by scent, defended by calling at neighbouring groups which usually threatens them away) in a mixed woodland habitat
friendly most of the year, except during the two week breeding season
close contact between mother and infant � later, other adults play with the infant, and infants with each other
adults groom each other
gibbons:
single family groups
entirely monogamous (unlike most of the other primates) lifelong pair bond, strict territoriality (maintained by elaborate �singing�, especially at dawn) � a lot like some birds
howler monkeys, vervets, macaques and most baboons:
multi-male, multi-female groups
gelada baboons and mandrills:
herds of several hundred
orangutangs:
almost solitary, associate only for mating and while offspring are dependent on their mothers
entirely arboreal
both sexes stay in ahome range over which they travel throughout the year following the fruiting of trees
marmosets:
small groups, with 1-3 adults of each sex plus their young
always have twins (whereas most other primates give birth to single offspring)
males help with the care of infants, carrying them for much of the time
hamadryas:
harem groups travel together
chimpanzees:
communities � form + disperse, sometimes coming together at night, perhaps foraging in widely separated groups for days on end (similar to bonobos)
the usual approach to understanding social organisation is to compare across various species for a wide range of ecological, morphological and behavioural factors
e.g. nature of food supply, pressure from predators, body size, sexual dimorphism, number of mature males in the group, group size etc.
bear in mind that the reproductive success of females is usually limited by access to resources, whereas that of males is by access to females
living in groups may compensate individuals for having to share food by increasing the efficiency of searching and improving defence
no complete agreement on why primate organisation is so diverse (e.g. see Ridley 1986)
individuals in a group are highly attentive to each other
need to be careful, because it�s easier/tempting to notice the complexity of primate groups over (e.g.) ungulates or carnivores, but there are still reasons for thinking this:
extended period of infancy
greater longevity (so they spend longer in each other�s company)
high intelligence (learning and flexibility of behavioural response (to changing social situation(???))
Humphrey (1976): the complex demands of social life may have been the one of the main selective pressures on the growth of brain size in primates
various signals are used within the dominance hierarchy, e.g. grooming (friendly (Seyfarth (1984))) or sexual presentation as an appeasement gesture
rank may determine access to food, preferred resting places and females
Rowell (1974) critised the way the notion of dominance hierarchies has been used:
dominance may be more important in captivity than in the wild
stressful + overcrowded, also fixed food-source points around which the more dominant animals cluster
may be based more on submission than aggression
however, dominance hierarchies can be observed in wild primates
e.g. pretty linear hierachy in Barbary macaques (Deag 1977)
predictability about social interactions will be advantageous for all concerned
often alliances between two to displace the one above (especially at the top end)
matriarchal groupings (mother + daughters) are often very enduring
the mother usually supports her youngest daughter most persistently (after all, she will have the longest reproductive age if she survives infancy)
young males become independent of their mother, and are ranked mainly by size
often join other groups
hamadryas baboons:
long hairy capes
move around in harems (which stay close together)
unattached females are rapidly acquired by males
yet there is scarcely any poaching between males, even if one is completely dominant over another in terms of access to food or resting places (Kummer 1980, experiments in captivity introducing a paired subordinate male + female into a cage with more dominant males)
Kummer (1969) transplanted a few yellow baboon females into a hamadryas troop, which were quickly herded into harems
they didn't respond to the male�s stare, and fled, but were chased back into the harem, and learnt their lesson � thus, at least some of the variation in social organisation between primate species is cultural rather than genetic
none
why don't moths and other bat prey emit false beeps to throw off the bats� echolocation, like we did in WW2???
�distributed grandmother cells� in Rolls vision revision tute
are they arguing for a definition of communication that involves mutual benefit, or just the benefit of the communicator???
reference for the plover broken wing behaviour (pg 177)???
evidence for the gestural origin of language??? well, bees are quite different from humans�
I don't understand the role males play in Hymenoptera � I don't understand where they get their genes from, or how many or born, or how queens get chosen
presumably though, the Hymenopteran haplo-diploidy slows the rate of evolution/variation
if the MHC outweighs more or less the rest of the genome when mice are deciding mates (Yamazaki et al., in essay), then isn't it a genetic outlaw???
��workers� are full sibs� � siblings???
it may be that the reliability of signals (in place of fighting) can only come through the use of signals that are themselves costly or at least difficult for any but the most vigorous and physically strong animals to produce � what�s the evidence against this (Zahavi�s) thesis???
ungulate /"V<ng>gjUl<schwa>t, -leIt/ a. & n.E19. [Late L ungulatus, f. UNGULA: see -ATE2.] Chiefly Zool.A adj. 1 Having the form of a hoof; hoof-shaped. E19. 2 Of a mammal: having hoofs; belonging to a group of mammals (formerly the order Ungulata) characterized by hoofs. Cf. UNGUICULATE a. 2c. M19.B n. An ungulate mammal. M19.ungulated a. (now rare) = UNGULATE a. 2 E19.
carnivore /"kA:nIv<revc>:/ n.M19. [Fr., f. L carnivorus: see next.] A carnivorous animal; a member of the order Carnivora of mainly carnivorous mammals (including dogs, cats, bears, seals, etc.). Also, a carnivorous plant.