Crane brains and behavior 2 - The wiring plan

"To represent the world is to have a special kind of wiring inside your head and special physical connections between that wiring and the world."
Peter Godfrey-Smith¹

Evolution has placed the executive brain (blue in the crane's head to the right) linked by short transmission lines to the forward scanners (eyes, nose, and ears). Within that tiny efficient brain, data inputs are integrated, compared with memories, and matched to coordinated adaptive responses.

In this blogpost, we introduce some recent research about the structural layout within bird brains. From that anatomical context, we can reason upstream, starting with observations of behavior and then making informed guesses as to how the bird brain works.

Within its mind, the crane must have some "representation" of the surrounding world:

• That representation could be just a long list of cues (stimuli) that are linked to behaviors (responses) so that each stimulus from the environment automatically triggers a fixed response. 
• Or that "representation of the world" could much more complicated -- organized in space and time into a coherent worldview with specific historical memories, maps of different environments, and primitive reasoning skills to produce flexible behavioral responses. 

In either case, the representation is based in genetics, incessantly informed by data from the surroundings, and adaptively modified during development and through learning.

Even casual observations show us that a bird's brain can discriminate. Roy and Millie, our crane pair, recognize our cars in our driveway and our voices in the house. When strange cars appear or strange people visit, they withdraw deeper into the marsh.
 
Birds are capable of much more impressive feats:

• Birds are wizards at distinguishing shapes. In experimental laboratory situations, pigeons can memorize up to 725 different visual patterns². Pigeons can even learn to distinguish paintings of Monet from those of Picasso³.
• Birds find their way while migrating enormous distances. Using radiotelemetry, John Wright tracked a population of Sandhill Cranes migrating from Fairbanks, Alaska to west Texas, over 3000 miles. In the spring, these same individual cranes return to interior Alaska, often homing in to nest on exactly the same pond, year after year.
  Three hundred miles south of Fairbanks in Homer Alaska, Ed Bailey and Nina Faust tracked another Sandhill population that nest in south-central Alaska to their wintering grounds in central California⁴.
  The capacity to navigate is surely innate - a talent encoded in crane DNA. The broad regional destination for winter feeding or summer nesting may be partly specified by genetics. But the exact targets and precise migration routes for the various populations are not genetically programmed. Apparently they are socially learned and then stored in the memories of experienced cranes for replay year after year.
  
• Birds make and use tools, and teach toolcraft to others. New Caledonian crows make tools from sticks or leaves, use these tools to fish for insect larvae in burrows, select the right tool for individual holes⁵, and pass the knowledge on to other crows via social learning⁶.
  

Armed with the knowledge that bird behavior is sophisticated, let's ask about the brain wiring. How are the nerve cells clustered together in centers and how are the centers connected in the brain? We don't have detailed brain data for cranes, but we know a lot about other birds and even more about mammals.

So we'll use a comparative approach. Cranes are end-products of lineages; their brains and their behaviors reflect their evolutionary history. If we know which parts of bird and mammal brains are similar by descent, then research on mammal brains might offer insights into crane behavior. And vice versa.

How much do mammal and bird brains resemble each other? They share the same broad architectural plan, but the lineages have been diverging and the brains differ in appearance⁷. In people, the cerebrum with its wrinkled cortex is linked to higher mental functions. The songbird cerebrum is smooth: what does this imply about intelligence? For a time, scientists thought that such differences in structure showed that birds can have only hard-wired instinctive behavior and lack adaptive intelligence. Then behavioral evidence, such as that we cited above, proved that birds are clever. To help us understand bird brains, we need to compare their structure and early development with mammal brains.

The nervous system in fish, birds and mammals arises in the early embryo as a thick-walled tube with the Forebrain, Midbrain and Hindbrain in the head⁸.

The top of the embryonic telencephalon (neuro-anatomists love Greek words) is the Pallium (Greek for "cloak"). The pallium will develop into the executive brain centers for relating input from sense organs, like eyes and ears, to output (muscle contractions). It is also the site for cognition and consciousness. Beneath the pallium are the basal ganglia that are concerned with instinctive behavior, movement, and motor learning.

Since the brain contains a world view, we could ask: Is it is possible to for us to put ourselves at the viewpoint and see a map of the world somewhere in the brain? In the early 20th century, a brilliant Montreal neurosurgeon named Wilder Penfield⁹ was able to find maps on the surface of the human brain - maps that depict sensory input from each patient's world and also direct the motor output to each patient's muscles.

Penfield opened the skulls of anaesthetized but fully conscious patients who suffered from epileptic seizures. . Since our brains have no pain receptors, Penfield could use tiny electrical probes to stimulate various regions of the brain without discomfort to the person. He explored here and there to find the sites in his patients that triggered their epileptic seizures. Then he could try to treat disease by surgically removing the local diseased region of the brain.

In response to Penfield's exploratory probing stimuli, patients reported smells or sounds or flashing lights or even twitched their muscles. From all those results, Penfield and others constructed maps of human cerebral cortex. The figure below shows such a map for touch (left) and motor control (right). Not every part of the body is equally represented. Motor control for the hand occupies a lot of space (on the right) as does sensory input from the face and lips (left).
As embyos develop, the forebrains fold, bulge, and become contorted so that the brains of birds and mammals look quite different on the surfaces. Look again at the songbird brain in the photograph above. At first glance, it is hard to tell exactly how the various parts of smooth-surfaced bird brains correspond to particular regions of the mammal cerebrum. Using the best information available in the early 20th century, brilliant anatomists tried to relate brain structure to the evolutionary histories of birds and mammals. They concluded that the higher centers in the adult bird brain developed from embryonic basal ganglia whereas the mammalian cerebral cortex came from the embryonic pallium. Since bird did not evolve the wrinkled mammalian pallial neocortex, it followed that birds are feeble-minded.

The conclusion was logical but based on incomplete data. Over the last 50 years, results converging from many scientific fields including brain imaging, cell biology, electrophysiology, genetics, molecular biology, ethology, comparative psychology, developmental biology, and even philosophical biology, all led to a single common conclusion - the early 20th century depiction of the bird brain was wrong in many respects.

In 2002, a group of eminent scientists gathered the new data together and produced a coherent modern terminology^6,10. The new nomenclature clarifies evolutionary relationships and identifies the anatomical sites that are probably responsible for higher cognition in birds. We need to understand this model and its implications in order to relate crane behavior to the crane brain.

Look below at cross-sections through the embryonic brain of a quail and of a rat¹¹. The orange pallium overlies purple basal ganglia. Already at this stage, the differences in shape are obvious. The rounded pallium of the bird is divisible into the hippocampus (Hp) and other regions (mesopallium etc.) where higher nerve centers will form. The flattened pallium of the rat is mostly the neocortex which overhangs the the hippocampus.

A cartoon using the modern consensus terminology for bird (left) and mammal (right) brains is shown below⁷. This is a section through the middle of the adult brains. The large pallial components (green in the cartoon) in the bird reflect the maturation of the orange regions in the embryo above. In mammals, the large pallial cerebral cortex overlaps the thalamus and cerebellum (blue) and buries hippocampus (red arrow) which is visible as a small green area just to the left of the blue thalamus in the human brain. The hippocampus is a brain region important for spatial learning.

Crows and jays use their hippocampi (red arrows on the left) to store over 10,000 individual locations where food items have been hidden. The differences in the hippocampi for birds and mammals shows one of many ways in which brain wiring for high mental functions has diverged.

As Penfield showed, the flattened mammalian cortex can be mapped in two dimensions across its surface. But evolution has yielded a different pallial architecture for birds and mapping must be in three dimensions. The higher centers (called nuclei) in the bird pallium are not conveniently arranged across the surface.

Terminology note: In cell biology, the word "nucleus" refers to the cellular organelle that encloses the genetic material. In neuroanatomy, the word "nucleus" refers to a compact cluster of nerve cells.

In birds, brain nuclei are jammed tightly together, piled on top of one another so that many are deep in the pallium. As an example, look at the robust nucleus of a towhee (the darker ovals in the figures below) that lie in the middle of the brain¹². This robust nucleus builds more cells in the spring when the male needs to sing (left) but the cells die off in the fall when the male is silent (right). Speaking personally, I would like to be able to grow new brain cells in order to think faster, but I'm not able to do so. The ability to grow new brain cells is exciting and the nuclei for birdsong have become models for basic biomedical research on multiplying brain cells.

Why are the bird brain nuclei piled up on top of one another? Not just to tease and confound brain cartographers! Probably because space is at such a premium. The expanded surface area and volume of a mammalian cortex may be one convenient design that birds simply cannot afford. Birds need to fly, and a large heavy brain and skull at the leading edge of the airship would be bad aeronautical engineering. Natural selection for a good flying machine favors a small head.

More about brain size and bird intelligence in a future blogpost.

References:

1. Godfrey-Smith P, 2002. On the evolution of representational and interpretative capacities. Monist 85:50-69.
2. von Fersen L, Delius JC, 1989. Long-term retention of many visual patterns by pigeons. Ethology 82:141-155.
3. Watanabe S, Sakamoto J, Wakita, M, 1995. Pigeons' discrimination of paintings by Monet and Picasso. J. Exp. Anal. Behav. 63:165-174.
4. See www.cranewatch.org.
5. Bluff LA, Troscianko J, Weir AA, Kacelnik A, Rutz C, 2010. Tool use by New Caledonian Crows Corvus moneduloides at natural foraging sites. Proc. Roy. Soc. London Biol. Sci. DOI rspb.2009.1953 [pii] 10.1098/rspb.2009.1953.
6. Hunt GR, Gray RD, 2003. Diversification and cumulative evolution in New Caledonian crow tool manufacture. Proc. Roy. Soc. London B 270:867-874.
7. Photo plate and drawing from Jarvis ED, Gunturkun O (25 colleagues), Reiner A, Butler AB, 2005. Avian brains and a new understanding of vertebrate brain evolution. Nature Reviews Neuroscience 6:151-159. This paper provides an overview of the modern nomencalture.
8. The embryonic brain drawing is adapted from Wikipedia.
9. Figure from http://www.pbs.org/wgbh/aso/databank/entries/bhpenf.html
10. For a meticuous exposition on the nuclei of the bird telecephalon, see: Reiner A, (27 colleagues), Jarvis ED, 2004. Revised nomenclature for avian telencephalon and some related brainstem nuclei. J Comp Neurol 473:377-414.
11. Adapted from from Nomura T, Takahashi M, Hara, Y, Osumi N, 2008. Patterns of Neurogenesis and Amplitude of Reelin Expression Are Essential for Making a Mammalian-Type Cortex. PLos ONE 3:e1454. 2009. We have labelled using modern terminology.
12. Figure adapted from Tramontin AD, Benowitz EA, 2000. Seasonal plasticity in the adult brain. TINS 23:251-258.