The Carney Landis Experiment

Suppose you’re making your way through a jungle, and in pulling aside a bush you find yourself before a huge snake, ready to attack you. All of a sudden adrenaline rushes through your body, your eyes open wide, and you instantly begin to sweat as your heartbeat skyrockets: in a word, you feel afraid.
But is your fear triggering all these physical reactions, or is it the other way around?
To make a less disquieting example, let’s say you fall in love at first sight with someone. Are the endorphines to be accounted for your excitation, or is your excitation causing their discharge through your body?
What comes first, physiological change or emotion? Which is the cause and which is the effect?

This dilemma was a main concern in the first studies on emotion (and it still is, in the field of affective neurosciences). Among the first and most influential hypothesis was the James-Lange theory, which maintained the primacy of physiological changes over feelings: the brain detects a modification in the stimuli coming from the nervous system, and it “interprets” them by giving birth to an emotion.

One of the problems with this theory was the impossibility of obtaining clear evidence. The skeptics argued that if every emotion arises mechanically within the body, then there should be a gland or an organ which, when conveniently stimulated, will invariably trigger the same emotion in every person. Today we know a little bit more of how emotions work, in regard to the amygdala and the different areas of cerebral cortex, but at the beginning of the Twentieth Century the objection against the James-Lange theory was basically this — “come on, find me the muscle of sadness!

In 1924, Carney Landis, a Minnesota University graduate student, set out to understand experimentally whether these physiological changes are the same for everybody. He focused on those modifications that are the most evident and easy to study: the movement of facial muscles when emotion arises. His study was meant to find repetitive patterns in facial expressions.

To understand if all subjects reacted in the same way to emotions, Landis recruited a good number of his fellow graduate students, and began by painting their faces with standard marks, in order to highlight their grimaces and the related movement of facial muscles.
The experiment consisted in subjecting them to different stimuli, while taking pictures of their faces.

At first volunteers were asked to complete some rather harmless tasks: they had to listen to jazz music, smell ammonia, read a passage from the Bible, tell a lie. But the results were quite discouraging, so Landis decided it was time to raise the stakes.

He began to show his subjects pornographic images. Then some medical photos of people with horrendous skin conditions. Then he tried firing a gunshot to capture on film the exact moment of their fright. Still, Landis was having a hard time getting the expressions he wanted, and in all probability he began to feel frustrated. And here his experiment took a dark turn.

He invited his subjects to stick their hand in a bucket, without looking. The bucket was full of live frogs. Click, went his camera.
Landis encouraged them to search around the bottom of the mysterious bucket. Overcoming their revulsion, the unfortunate volunteers had to rummage through the slimy frogs until they found the real surprise: electrical wires, ready to deliver a good shock. Click. Click.
But the worst was yet to come.

The experiment reached its climax when Landis put a live mouse in the subject’s left hand, and a knife in the other. He flatly ordered to decapitate the mouse.
Most of his incredulous and stunned subjects asked Landis if he was joking. He wasn’t, they actually had to cut off the little animal’s head, or he himself would do it in front of their eyes.
At this point, as Landis had hoped, the reactions really became obvious — but unfortunately they also turned out to be more complex than he expected. Confronted with this high-stress situation, some persons started crying, others hysterically laughed; some completely froze, others burst out into swearing.

Two thirds of the paricipants ended up complying with the researcher’s order, and carried out the macabre execution. In any case, the remaining third had to witness the beheading, performed by Landis himself.
As we said, the subjects were mainly other students, but one notable exception was a 13 years-old boy who happened to be at the department as a patient, on the account of psychological issues and high blood pressure. His reaction was documented by Landis’ ruthless snapshots.

Perhaps the most embarassing aspect of the whole story was that the final results for this cruel test — which no ethical board would today authorize — were not even particularly noteworthy.
Landis, in his Studies of Emotional Reactions, II., General Behavior and Facial Expression (published on the Journal of Comparative Psychology, 4 [5], 447-509) came to these conclusions:

1) there is no typical facial expression accompanying any emotion aroused in the experiment;
2) emotions are not characterized by a typical expression or recurring pattern of muscular behavior;
3) smiling was the most common reaction, even during unpleasant experiences;
4) asymmetrical bodily reactions almost never occurred;
5) men were more expressive than women.

Hardly anything that could justify a mouse massacre, and the trauma inflicted upon the paritcipants.

After obtaining his degree, Carney Landis devoted himself to sexual psychopatology. He went on to have a brillant carreer at the New York State Psychiatric Institute. And he never harmed a rodent again, despite the fact that he is now mostly remembered for this ill-considered juvenile experiment rather than for his subsequent fourty years of honorable research.

There is, however, one last detail worth mentioning.
Alex Boese in his Elephants On Acid, underlines how the most interesting figure of all this bizarre experiment went unnoticed: the fact that two thirds of the subjects, although protesting and suffering, obeyed the terrible order.
And this percentage is in fact similar to the one recorded during the infamous Milgram experiment, in which a scientist commanded the subjects to inflict an electric shock to a third individual (in reality, an actor who pretended to receive the painful discharge). In that case as well, despite the ethical conflict, the simple fact that the order came from an authority figure was enough to push the subjects into carrying out an action they perceived as aberrant.

The Milgram experiment took place in 1961, almost forty years after the Landis experiment. “It is often this way with experiments — says Boese — A scientis sets out to prove one thing, but stumbles upon something completely different, something far more intriguing. For this reason, good researchers know they should always pay close attention to strange events that occur during their experiments. A great discovery might be lurking right beneath their eyes – or beneath te blade of their knife.

On facial expressions related to emotions, see also my former post on Guillaume Duchenne (sorry, Italian language only).

Mouse serenade

Even mice sing.
We have known that for 50 years, but we are only recently beginning to understand the complexity of their songs. Part of the difficulty of studying mice songs lies in their ultrasonic vocalizations, frequencies the human ear cannot perceive: in the wild, this kind of calls happen for example when a mouse pup calls for his mother.

In April, in Frontiers of Behavioral Neuroscience, a new Duke University research appeared, showing how mice songs are really much more intricate than expected.
Researchers Jonathan Chabout, Abhra Sarkar, David B. Dunson and Erich D. Jarvis have exposed the mice to different social contexts and, using new specifically elaborated software, they have analysed the frequency modulation and duration of these ultrasonic calls. Researchers have been able to break down the songs into “syllables” and clusters of sound repeated to a certain rythm, and to discover how they vary according to the situation.

If a male mouse is exposed to female urine, and therefore gets convinced that she is somewhere nearby, his singing becomes louder and more powerful, if somewhat less accurate; to awake a sleeping female, he utilizes the same song, but the “syllables” are now pronounced much more clearly.
Female mice seem to prefer songs that are complex and rich in variations; even so, when a male finds himself near an available female, his elaborate courting song switches to a simpler tune. Once the potential mate has been attracted, in fact, our little mouse needs to save energy to chase her around and try to mate.

The mouse’s ability to sing is not as articulate as in songbirds; and yet, changes in the syntax according to social context prove that the songs convey some meaning and serve a precise purpose. Researchers are not sure how much mice are able to learn to modify their vocalizations (as birds do) or how much they just choose from fixed patterns. Forthcoming studies will try to answer this question.

It is nice to better understand the world of rodents, but why is it so important?
The goal of these studies is actually also relevant to humans. In the last decade, we understood how mice are extremely similar to us on a genetic level; discovering how and to what extent they are able to learn new “syllables” could play a fundamental part in the study of  autism spectrum disorders, particularly in regard to communication deficits and neural circuits controlling vocal learning.

Male mice song syntax depends on social contexts and influences female preferences, Jonathan Chabout, Abhra Sarkar, David B. Dunson, Erich D. Jarvis. Frontiers in Behavioral Neuroscience, April 1, 2015.