Squirrel Acrobatics Amaze Scientists
The other day, walking down a tree-lined street, we were startled by a sudden crash and scramble at our feet. A squirrel had fallen from a tree! It quickly recovered and dashed away, back up the same tree trunk from which it had come. That never — or almost never — happens. Why not? Scientists at UC Berkeley have been investigating just that question.
The desire for a peanut is enough to make a squirrel give an Olympic performance. The scientists practically held up “9” and “10” score signs as they watched their guest squirrels run an obstacle course the researchers set up. The squirrels, which live in eucalyptus trees on the campus, even used creative moves when the course was made more difficult. This story shows that good-old empirical science still has the power to fascinate and build understanding, without a need for Darwinian storytelling.
How many movies show a protagonist running from danger, having to make a split-second decision whether to leap over a gap? The scene from Raiders of the Lost Ark showing Indiana Jones running from a rolling boulder, leaping over a chasm, and hanging by his fingernails comes to mind. For squirrels leaping from branch to branch in the trees, this is par for the course. Speaking of that, they also know parkour — the ability to bounce off a wall for extra oomph in a dangerous leap. Parkour is a popular urban sport highlighted in many YouTube videos, where a runner tries to leap from rooftop to rooftop through a rapid series of daring leaps without assistance. It is as dangerous as it looks, but some get really good at it, performing flips and twists in some of the riskier sequences.
UC Berkeley biologists Nathaniel Hunt and Robert Full, with help from two colleagues in the psychology department, set up the outdoor experiments on campus. A quick video shows the setup and some of the squirrel performances in slow motion.
The results made the cover story in Science on August 6 with the title, “Acrobatic squirrels learn to leap and land on tree branches without falling.” It begins with alacrity:
Every day, there are acrobatic extravaganzas going on above our heads. Squirrels navigate remarkably complex and unpredictable environments as they leap from branch to branch, and mistakes can be fatal. These feats require a complex combination of evolved biomechanical adaptations and learned behaviors. Hunt et al. characterized the integration of these features in a series of experiments with free-living fox squirrels (see the Perspective by Adolph and Young). They found that the squirrels’ remarkable and consistent success was due to a combination of learned impulse generation when assessing the balance between distance and branch flexibility and the addition of innovative leaps and landings in the face of increasingly difficult challenges. [Emphasis added.]
Parkour Engineering Specifications
A novice is not going to nail a new parkour move on the first try. Squirrels, however, seem to be born with ability to do it. A combination of traits is necessary: a flexible body, good senses, strength, agility, rapid decision-making ability, instinct, and ability to learn. The UCB research team seems vague about the ratios of these specifications. Which traits are the most critical for success? Lack of any one of them could prove fatal. It looks like an irreducibly complex set of specs is required for parkour, no matter which animal does it. This is the case for monkeys, apes, lizards, cursorial birds, and the deer bouncing over the truck hood and trotting off.
In their Perspective article about the research, “Learning to Move in the Real World,” Adolph and Young bring up another specification: the ability to adjust quickly to internal changes. Animals gain weight; females become pregnant. The body during growth and development is continuously changing mass. This happens to human infants as well, as every parent knows:
During development, new affordances emerge as animals’ bodies, skills, and effective environments change. Human infants can grow up to 2 cm in a single day. One week, babies are crawlers; the next, they are walkers — yesterday, objects on the coffee table were out of sight and beyond reach; today, they are accessible. Thus, learning occurs in the context of development, and the flux of body growth and motor-skill acquisition ensures that infants do not learn fixed solutions. Indeed, static solutions would be maladaptive in a continually changing ecosystem. Instead, infants “learn to learn.” They learn to detect information for affordances at each moment to determine which actions are possible with their current body and skills in a given environment.
In the experiments, the team gave the squirrels challenges that required calculating risks and rewards. To get to the nut, a squirrel had to negotiate a narrow flexible strip and then leap to a post. Squirrels instinctively knew whether the strip could support their weight and was stiff enough for the launch. If the strip was sufficiently stable, they would crawl out to the end and jump. If not, they would start their jump farther back with more energy. Sometimes they would “parkour” off the vertical wall to get to the post. If they overshot or undershot, they had a backup plan: they knew their claws could save them. They would grab the bar and flip over or under it and land on top like a star gymnast. Most of the time, they nailed the landing with all four feet fitting on the tight platform provided. The scientists were amazed at their quick calculations.
“They’re not always going to have their best performance — they just have to be good enough,” he said. “They have redundancy. So, if they miss, they don’t hit their center of mass right on the landing perch, they’re amazing at being able to grab onto it. They’ll swing underneath, they’ll swing over the top. They just don’t fall.”
Well, almost never. And that adds another specification: an excellent kinesthetic sense. The animal must know its body’s strengths and weaknesses, its position, and its current weight. If a mother squirrel is leaping while carrying a baby, it must calculate its resources without error each time.
An Engineering Perspective
The “acrobatic extravaganzas” going on all around us in the living world are easy to take for granted. An engineering perspective helps unpack the requirements. What must be true for this phenomenon to occur? Evolutionists defame the achievements of squirrels and humans with their dismissive statements that phenomena just evolved. The UCB scientists seem to recognize the requirements, but attribute them to evolution anyway:
Gap traversibility depends on the complement of environmental properties with an animal’s locomotive capacities. The synergy between biomechanical energy management and learned information for launching and landing likely determines arboreal leaping and ultimately the path through the canopy. The role of fast and accurate leaping in driving the evolution of biomechanical capabilities, learning-based decision-making, and innovation promises to reveal the mechanisms and origins of arboreal agility
In that last sentence, they admit they cannot account for the origin of these mechanisms. So how can they speak of “the evolution of biomechanical capabilities”? Their worldview forces them to imagine an unguided origin, because at some earlier time in their view animals did not have these biomechanical capabilities. How and when did they emerge? What good is one specification when the others are not yet present?
It’s the engineering perspective that intelligent design offers that brings light to these questions. A complex ability like leaping over a gap to a reward presupposes a set of specifications. Each spec is measurable: for a squirrel mass m, needing to clear a gap of distance d, launching from a platform with springiness x, an engineer can calculate the force needed and test it with robotics. More variables can be specified for in-flight correction (parkour moves) and claw strength for grasping and swinging. The specs are likely to get increasingly complicated when requirements for sensing and balance are considered. But this is science that is precise, measurable, and testable. It also has explanatory value: once the requirements are known, the set of causes necessary and sufficient to achieve them can be evaluated.
Lead author Nathaniel Hunt, who also works in the Department of Biomechanics at the University of Nebraska, should welcome mechanical engineers to his team. It’s not a long leap to a Department of Biomechanical Engineering. That, not evolution, is what would promise “to reveal the mechanisms and origins of arboreal agility.”