Ethology. 1h 40m Robert Sapolsky video.

Another way to understand the biology of behavior is through ethology whose approach is characterized by 1) study behavior in lots of different environments 2) study behavior in the natural environment (get out of the lab), and 3) assume you are going to have to translate a language. Nobel laureate Nikolaas Tinbergen says "ethology is the process of interviewing an animal but in its own language."

Sapolsky explains the principles of Ethology and shows how ethology has enriched our understanding of behavior and refuted many early ideas from psychology in particular refuting the radical environmentalist program of behaviorists like B.F. Skinner. Many wonderful examples of collecting behaviors in the wild and in the lab are cited. He goes into some depth about what ethology has taught us about learning.

9. Ethology


Sapolsky starts with a review of the history of psychology to provide a contrast with ethology. Around 1900 pioneer William James pursued psychology as an introspective philosophical project and envisioned it as a branch of philosophy. John B. Watson and others moved the field in a new direction by measuring behavior with the idea that with reward and punishment (i.e., reinforcement theory) we can control anything about an organism. So genes don't matter. This early form of behaviorism, advanced by B. F. Skinner and others, is an extreme form of environmentalism yet it became prominent and highly influential. Behaviorists also think a universality applies to all species: rats and pigeons in the lab work just as well as humans or apes in the jungle.

Meanwhile another tradition developed in England by folks with butterfly nets. Ethology works to collect behaviors in the field and wallows in the variability observed there. There are three luminaries in the field who shared the 1973 Nobel Prize in Physiology or Medicine: Nikolaas Tinbergen (the saint), Karl von Frisch (who was old) and Konrad Lorenz (the Nazi defender who discovered imprinting). This field examines the gene-environment interaction where every species solves its environmental challenges in a unique way.

The principles of Ethology. 1) What is the behavior? In the lingo of the field, what are the "fixed action patterns". 2) What in the outside world triggered the behavior? What environmental "releasing stimulus" ("releasers") caused the fixed action pattern? 3) What are the intervening mechanisms? What are the "innate releasing mechanisms" mediating the behavioral response? 4) What is the adaptive value of the behavior?

One way in which ethology started questioning the radical environmentalism of the behaviorists was to point out that you don't get normal behavior in the lab. In the 1960s, experiments showed that rats with "enriched" environments had thicker cortexes (which may have inspired the 1965 Head Start Program). But ethologists discovered that wild rats had thicker cortexes than the most enriched lab rats out there.

Then Sapolsky gives several examples showing that animals do not need to learn many behaviors, but they do need to learn how to do the behavior better and they need to learn the right context for their innate behaviors. Visual cliff response behavior: in the lab put babies of various species on a plate of glass that is "very high" to give them the visual sensation that they just fell off a cliff. As you might expect the animals experience terror (a panic response behavior). Excepting baby sloths who show no fear of the situation: aha, a species-specific behavior refuting the behaviorists claim that you can experiment on any animal with the same results. Once the babies learn that the glass is safe, the panic behavior gradually wanes.

Monkeys raised in isolation show a fixed action pattern of subordination (crouch down, no eye contact) when presented with a movie of a male aggressive display. It is innate!

Vervent monkeys are afraid of their predators: leopards, snakes, and eagles. Their fixed action patterns are distinct alarm calls for each predator. The use of which is sculpted by experience. Adults only pay attention to alarm calls of the young after an adult agrees it's a threat.

Humans have fixed action patterns: infant smiling and suckling (as with all mammals). We need to learn when to smile and how to suckle more efficiently. We also have the fixed action pattern of emotional signaling: we can recognize anger, fear, disgust, and contempt but we need to learn the social context, to learn appropriate use of these signals.

Ethologists also wonder about the adaptiveness of behavior. Tinbergen was studying gulls and wondered why they turn egg shells over so that the speckled side (not the inside/white) faces up. So after Mom leaves the nest Tinbergen turned all the egg shells white side up and then waited. An eagle spots the nest and eats the chicks. Aha, now we know why they did it. Ethologists are crafty experimenters!

Von Frisch studied bees with their figure 8 waggle dance. Through clever experiments he learned that the axis of the 8 tells the direction with respect to sunlight in which the food source is, the length of the dance codes the distance, and a more frantic dance means the food source is more exciting. Using hives on barstools and bees captured and released from different places, he was able to listen to the bees in their own language and decipher it!

Can we experimentally determine the releasing stimulus? When gulls peck at mom to get regurgitated food. Could it be the red spot on her beak that releases their pecking fixed action pattern? Let's put a red circle on paper (replacement) and see if they peck at that. Or blot out Mom's red circle (subtraction) to see if the chicks stop pecking for food. Is it specific to color, size or shape? Will a super-stimulated moving big red circle (exaggeration) trigger the fixed action pattern?

Using subtraction and substitution with styrofoam "females", ethologists can determine what triggers the mating fixed action pattern in male turkeys. Now with robotic bees and cockroaches, ethologists have even better experimental tools to interview animals in their own language.

Ethologists have discovered that some species (and even humans!) are functioning in sensory modalities we did not expect. For example, male deer bellow which turns out to be a releasing stimulus for female deer to ovulate: auditory induced ovulation. Rats ultrasonically giggle when tickled on their rib cage which is a releasing stimulus to come check this out. Electric fish songs encode relatedness and can "jam" each other's signals during male competition. Arachnids (spiders) can communicate through vibration of their web. Caitlin O'Connell-Rodwell (http://news.stanford.edu/pr/01/elephants37.html) finds that elephants can communicate via vibrations detected by receptors in their feet.

When mammals ovulate (including human females), their voice rises in pitch. Males subliminally respond. Brain scanning of individuals smelling sweat from terrified skydivers vs. normal exercise shows increased activation of the amygdala, the brain region associated with fear responses, and the subjects interpreted representations of faces as being more fearful. Human fear is communicated by smell?

Harry Harlow found that rhesus monkeys preferred terry-cloth moms to wire moms with sources of milk. Evidently, comfort and love is a factor in child-rearing? It isn't all about the economics of food delivery? Should we add human touch back into the nursery and premature-birth incubators?

Stephen Jay Gould's classic paper "A Biological Homage To Mickey Mouse" looked at 50 years of evolution of Mickey Mouse in which he became more juvenile: shorter muzzle, bigger rounder eyes, big shiny forehead, with distinctive coloration.

Innate releasing mechanisms are the wiring inside that turns the releasing stimulus into a fixed action pattern. Classical ethology didn't consider the subject, but for about 20 years now neuroethology tries to figure out how the brain mediates releasing stimuli and fixed action patterns. The neuroethology of bird song investigates how seasonal breeding birds learn their unique song each year. If you raise a bird in isolation, what kind of song does it do? It is close, but evidently experience is needed to shape the song. Myna birds imitate sounds from other species (it is not innate!).

In female hamsters, the lordosis reflex (when pushed arching her back to expose her genitalia) only occurs during ovulation when estrogen levels are higher. Neuroethology has reverse engineered this behavior down to the neuronal level (see "Reverse engineering the lordosis behavior circuit" http://www.med.upenn.edu/idom/Newsroom/documents/flanagan_reverseengineering.pdf).

John Wingfield is doing neuroethology studies in the field. Sapolsky does research on baboons and anxiety in the field in Africa. He observes that we need the ethological context in order to make sense of the brain and what sensory systems it pays attention to. In this context, the fixed action patterns are the behavior, the releasing stimulus is the trigger, and the innate releasing mechanism are the neuro-hormonal circuitries.

What is the role of learning?

Maternal competence in monkeys is not instinctual, not a fixed action pattern: it must be learned. Rhesus monkey offspring survive better if their mothers have an older sister. More aunting behavior and chimp/monkey children survive better. Similarly, if mom was a Big Sister. Mothering is not instinctual? Does it work that way in humans? Should parenting school be a requirement for pregnant women and their prospective fathers?

Meerkats teach their pups to hunt scorpions using the progressive difficulty process we are all familiar with from school: first how to eat dead scorpions, then live scorpions whose stinger has been removed, followed by fully intact scorpions.

Ethologists find even more unexpected learning. For example, animals make tools! Jane Goodall showed chimps using sticks to hunt for termites and using hammer and anvil to crack nuts. All apes have tool-making skills. But baboons can't figure it out even after watching chimps.

Learning is enhanced in those who pay attention better (since female chimps learn tool-making better, were females the first tool makers?). Those who live in larger social groups tend to learn more younger. Learning by experience is of course very important. Male chimps have been observed fashioning weapons from large pieces of wood.

Ethologists discovered important counterexamples to the behaviorists beloved reinforcement theory of learning. One trial learning: the duckies of Konrad Lorenz: imprinting. During a (not very precise) "critical period" ducklings will imprint on whatever big thing is moving around them which statistically is likely to be mom (unless an ethologist is at work).

Prepared learning occurs when certain associations are prepared by instinct, but learning is still required to acquire the skill. Bees learn smell associations much faster than shape or color: olfactory prepared learning. All sorts of species come wired to make some associations more readily than others.

Humans and other primates have prepared learning for fear of snakes and spiders, but there are many cultures with no fear of spiders or snakes so it is not innate. We also have prepared learning to notice animate objects over inanimate ones (also snake shaped objects). Martin Seligman provided the interesting example of Sauce-Béarnaise Syndrome: he developed a taste aversion to Béarnaise sauce months after getting an upset stomach even though a performance was closer in time than his eating of the sauce. We have prepared learning to associate upset stomach with food violating a key principle in reinforcement theory.

Cognitive ethology attempts to discover the mechanisms in the head of the animal, the actual mechanics of the innate releasing mechanisms. In the book "Question of Animal Awareness" (1976) Donald Griffin (who discovered echolocation in bats) outlined an initially controversial research program to discover the cognitive and emotional awareness of animals. We now know that animals have strategic awareness and all sorts of other unexpected cognitive capabilities.

Dogs do not have self-awareness: they bark at their own reflection till the end of time. Putting marks on the foreheads of chimps while they are anesthetized leads to them feeling the mark showing that they have self-awareness. Elephants have self-awareness. Then he cites Marc Hauser's possibly disputed research on marmoset monkey self-awareness where marks on the forehead did not invoke self-aware behavior so he moved the marks to another location to discover that marmosets do have self-awareness (unless he fudged the data on that experiment too: Hauser was fired from Harvard for scientific misconduct after this video was made).

Theory of mind: recognition that there are other actors who can have different information than I do. In humans it typically develops between age 4 and 5. The Sally–Anne test (Sapolsky's version is not the simpler one from the Baron-Cohen, Leslie, and Frith study from 1985 http://ruccs.rutgers.edu/~aleslie/Baron-Cohen%20Leslie%20&%20Frith%201985.pdf): in a skit a child goes to school then mom puts their Sally-Anne doll in the washer. If the subject answers correctly that the child in the skit will look for their doll in the bedroom where they left it, we infer they have theory of mind. Chimps have theory of mind: lower ranking individuals will not go for the food unless they know the dominant chimp doesn't know where it is (using clear and opaque windows). Chimps do not have theory of mind in cooperative settings, only in highly motivating competitive scenarios. Corvids (ravens) have theory of mind for hiding food: they behave differently if they know another bird is watching.

Animals can distinguish between intentional and unintentional behavior. Chimps know if their food was accidentally disturbed or intentionally flung and dogs know if you tripped over or kicked them.

Animals have flexible cognitive strategies. In bees if you release a bee from a great food source on a boat in the middle of the lake, when they go back to the hive and dance to tell where the food is no one pays attention. They know that there cannot be a food source in the middle of the lake! They demonstrate a flexible cognitive strategy. Animals can plan for the future. Corvids (ravens and crows) will move food from surplus areas to deficit areas.

Numerosity is a sense of number. Chimps can distinguish order in pattern sequences. They can also determine if the number of males in their group is larger or smaller than the number of male voices bellowing territorial calls from an ethologist's loudspeaker.

Russell Fernald found transitive inference in cichlids (a territorial African fish): if C observes that A defeats B and B defeats C, then C will give a subordinate gesture to A (http://news.stanford.edu/news/2007/january31/fishsr-013007.html). This also appears to only work in highly motivating competitive circumstances.

Sapolsky ends with a description of ethology: "We have a completely different way of thinking about behavior and emphasizing here natural setting, experimentation rather than coming up with those just-so stories, the notions of all sorts of species including us functioning in sensory domains that are unheard of, and finally all sorts of types of learning in ways that organisms are not supposed to learn that all of us do."