Weizmann Institute and Ichilov Institute study reveals, using electrodes deep in the brain, how loss avoidance mechanisms affect learning, memory, and anxiety and post-traumatic stress disorders
People are willing to take risks to gain, but when there is a possibility of losing, they will look for any way to avoid it altogether—even if their action goes against logic. The understanding that people give more weight to losses than gains in decision-making is at the heart of the “value theory” that won Daniel Kahneman the Nobel Prize in Economics in 2002. Yet, very little is known about how loss aversion affects behaviors such as learning and memory and the brain mechanisms that underlie these behaviors. In a new study in humans, the findings of which are published in the journal Scientific Reports, Nature and-Current BiologyResearchers and doctors, led by Prof. Rony Paz and Dr. Tamar Reitisch-Stolero from the Weizmann Institute of Science and brain surgeon Prof. Ido Strauss and neurologist Dr. Piras Pahom from the Sourasky Tel Aviv Medical Center (Ichilov) – brain mechanisms that are responsible for humans being more sensitive to loss and relating to it differently during learning and decision-making. Unnecessarily and excessively activating loss avoidance mechanisms can explain the behavior of people dealing with anxiety and post-traumatic stress disorder (PTSD).
The amygdala, an almond-shaped structure in the temporal lobe of the brain, is a center for emotional processing of feelings such as fear and stress and regulates the response to danger. "In recent years, the use of electrodes inserted into areas deep in the brain - including the amygdala - has been increasing to find the focus of an epileptic seizure in patients who do not respond to medication," explains Prof. Strauss. "Unlike electrodes placed on the scalp (EEG) and measuring average electrical activity over many large areas and millions of neurons, deep electrodes measure electrical activity at the resolution of a single neuron and therefore allow us to study and understand the computation performed in the brain by local populations of cells." In the new study, the research team harnessed the deep electrodes to study the neural mechanism of loss avoidance during learning.
During the learning experiment, the subjects connected to the electrodes experienced two types of tasks – tasks in which points could be earned and those in which points could be lost. Each time a tone was played to signal whether it was a gain or loss task, and then two shapes appeared on the screen – one with a high probability of gain (in the gain task) or loss (in the loss task) and another with a low probability. Throughout the experiment, the subjects learned which choice led to more gain and which choice led to less loss. In both conditions, they learned which choice consistently provided better performance.
""Fear of loss causes people to rely less on the knowledge they have accumulated and to insist on finding new ways of doing things, even when it leads to more loss and less profit in the short term""
"Yet, performance in the two tasks was different. Even after learning which choice reduced the chances of losing in the loss task, the subjects continued to choose the other way from time to time. They were desperately trying to find some strategy that would completely prevent a loss and not just reduce its likelihood," Dr. Reitisch-Stollero describes. The findings published inNature. "On the other hand, in the profit task, subjects consistently made the choice with the preferred odds and sought fewer creative courses of action."
Every animal is always in a dilemma between utilizing the knowledge and experience it has accumulated and searching for and exploring new and better ways of doing things. The findings in the study show that in humans this dilemma varies between situations of gain and loss, and that fear of loss causes people to rely less on the knowledge they have accumulated and to insist on finding new ways of doing things, even when this leads to more loss and less gain in the short term.
To unravel the neural mechanisms behind this behavior, the scientists monitored the electrical activity of hundreds of neurons in different areas of the brain at the single-neuron level. They discovered that there is a subset of neurons in the amygdala and temporal cortex that are active and fire at a higher rate before the decision is made to explore new options. In other words, these cells signal whether the next choice will explore a new option or draw on previous knowledge.
"This mechanism operated in both the loss and gain tasks and with equal intensity, so we initially had difficulty understanding why subjects were more curious in the loss tasks," notes Dr. Reitisch-Stolro. "We had to think outside the box and raised the possibility that neural noise shapes coping with the risk of loss."
Neural noise is an inconsistency in the activity of nerve cells, meaning a state in which cells have moments of fast firing and moments of slow firing. The research team identified an increase in the level of neural noise in the amygdala under conditions of risk of loss. "Through computational modeling, we identified that neural noise in the amygdala is associated with the feeling of uncertainty. We also identified that people are more sensitive to uncertainty regarding losses than gains, and this triggers more exploratory behavior in them," describes Dr. Reitisch-Stolro, expanding the scope: "When exploratory behavior loses its brakes and gets out of control, people become preoccupied with thinking that there is a better course of action than their own, and this is, among other things, a clear hallmark of anxiety disorders."
What does the following sound tell you?
In the second part of the same experiment, whichHis findings are published inCurrent Biology, The scientists focused on another aspect of decision-making – the ability to generalize. The scientists played sounds that had previously been associated with gain or loss to the subjects, alongside new sounds that were close or far from them. They found that the subjects tended to recognize a wider range of new sounds as “familiar” when they were close to the sound associated with loss. Moreover, their brains tended to interpret sounds that were similar to the loss sound as a warning signal.
"Learning rules or generalization is one of the main hallmarks of intelligence," says Prof. Paz. "The broad generalization of danger signals has developed throughout evolution and allows us to formulate sweeping precautionary rules from our experience that will protect us from new dangers. However, when generalization gets out of control, it can be harmful. This is why, after hearing a warning about rocket fire, even the growl of a motorcycle passing by on the street makes us panic. It is an excellent defense mechanism, but when it is applied excessively, as in post-traumatic stress disorder (PTSD), it leads to stress reactions, depression, and anxiety even in everyday situations."
Recordings from electrodes deep in the subjects' brains revealed a neural mechanism that explains why we are more likely to generalize when there is a risk of loss. The scientists identified that neurons in the amygdala are activated more by sounds that resemble the sound of loss, and that this activity precedes generalization, that is, it precedes the decision that it is a familiar sound.
"We were able to predict, based on the level of cell activity, whether a person would make a generalization and think that they heard a familiar sound or an unfamiliar sound," says Prof. Paz. "Negative conditioning can make us think that we heard a different sound than what we actually heard, in other words, change our sensory processing. Although the connection between the amygdala and fear and anxiety has been known for many years, until recently we were unable to examine the neural mechanisms of decision-making in humans at high resolution, and therefore it was also not possible to understand what exactly goes wrong in disease states. The new findings pave the way for better treatments for post-traumatic stress disorder and mood disorders."
Dr. Kristofer Aberg, Dean Halperin and Carmel Ariel from the Department of Neuroscience at the Weizmann Institute; Dr. Ganela Morris from Tel Aviv University and Dr. Lilach Goldstein and Dr. Lotem Bergman from Ichilov also participated in the studies.
More of the topic in Hayadan:
