Who hasn’t been tempted to throw a long allowance in the trash or just drive on instead of asking for directions? After all, following instructions is often boring, and we can figure it out on their own … Or maybe? A study published today (May 19) in a scientific journal Nature Human Behavior challenges common theories about our ability to solve complex problems and how certain mental disorders affect it.
“Patients with obsessive-compulsive disorder (OCD) are thought to have difficulty developing sophisticated problem-solving strategies,” said Albina Oliveira-Maya, senior author of the study and head of the neuropsychiatry department at the Champalimo Foundation in Portugal. “However, our new experimental approach provides strong evidence against this theory.”
Two ways to solve one problem
The Oliveira-Maya team did this by exploring how healthy subjects and patients with OCD differ in the way they solve problems. “In general, people use a combination of two complementary strategies known as model-free approaches and model-based approaches,” Oliveira-Maya explained. “While healthy people flexibly use both strategies, patients with OCD tend to use a no-model approach.”
The strategy without the model is relatively simple and works well in stable conditions. For example, imagine the following scenario: you have breakfast on the street every morning on the way to work. There are two cafes on your route: The Bean and Aroma. Since you have to be at work early, over time you’ll learn that Aroma usually delivers your breakfast – fresh croissants – earlier than other stores. So, following a model-free approach, you tend to go to Aroma first, and only if it doesn’t have croissants do you go to The Bean.
However, a modelless approach would not work very well if the croissant supplier hired two vendors who took opposite routes. In the weeks when the first carrier was on duty, The Bean had been getting croissants before. But if a second vendor is working this week, Aroma will get them first.
If you could spot a model – that the availability of croissants depends on which vendor is working this week – you would save yourself from unnecessary travel. So even if The Bean had croissants bright and early for a few weeks, on the first Monday, if that’s not the case, you’ll immediately learn that this week’s Aroma is a safer choice.
“Although a model-based strategy is more complex than computing, especially if you’re still aware of what’s going on, it’s more effective at optimizing your actions in complex circumstances such as this example.” said Oliveira-Maya.
Changing the rules of the game
According to Oliveira-Mai, research that evaluates these strategies typically applies a puzzle called a two-step task that is similar to a second, more complex scenario.
“These studies have shown that healthy subjects use a mixture of a simpler strategy without a model with a more sophisticated model-based strategy in solving these types of problems. In contrast, patients with OCD tend to follow a less effective strategy. The suggested reason is that patients with obsessive-compulsive disorder (OCD) are very common and therefore they tend to repeat actions even if they do not serve a useful purpose, ”Oliveira-Maya explained.
While this conclusion seems simple and consistent, there is a catch. Because the tasks used in these studies are usually very complex, subjects always receive a full explanation of the model before they begin. However, no one has ever rigorously tested the impact of these warning instructions – especially their absence – on subjects ’problem-solving strategies.
To find out how people will only do free experiments, the Oliveira-Mai team collaborated with Thomas Akam, a neurologist now at Oxford University who recently developed a two-step task for mice.
“Since you can’t verbally instruct mice, Thomas created a task that was simple enough for animals to decipher the model by trial and error. In his scientific paper published in the journal Neuron a little over a year ago Thomas showed that mice are really capable of solving puzzles. So we decided to set this task up for people and see if subjects would naturally adopt a model-based strategy, as is commonly assumed, ”said former doctoral student Pedro Castro-Rodriguez, the first co-author of the study.
The results of the experiment caught the researchers by surprise. “Even with extensive experience in this task, only a small minority of the 200 subjects have developed a strategy based on the model. This is surprising given the relative simplicity of the task and shows that people are surprisingly poorly assimilated causal models with only experience, ”Castro Rodriguez said.
Patients with OCD coincide with healthy subjects
At the end of the third session, the researchers divided the subjects into two groups. One group received a complete description of how the puzzle works, and the other did not. The researchers then conducted a fourth and final session to test the impact of obtaining instructions on subjects ’approach to problem solving.
The difference between the two groups was obvious: almost all subjects from the “explanations” group — both healthy volunteers and OCD patients — adopted a model-based strategy. On the other hand, most subjects from the other group continued the modelless approach.
“These results were exciting,” said Ana Maya, a doctoral student who participated in the study. “They have not only shown that explanation plays a more important role than previously thought, but also that with the right set of conditions, patients with resective disorders (OCD) are actually able to optimally solve a two-step problem like healthy people.”
What is the reason for the discrepancy between the results of this study and previous ones? According to the authors, there are several possible explanations. First, the task was relatively simple, as were the instructions. “Because classic two-step assignments are usually very complex, the explanations are also very complex. So you can imagine that a person who is acutely ill and suffering will find it harder to process such information,” Oliveira-Maya explained.
Another interesting hypothesis is that starting with a free experiment makes a difference. Is it possible that three classes without guidance effectively prepared patients for explanations?
“We did not test this issue directly in this study, but there are some hints that it could be. If future research confirms this hypothesis, they may even lead to the development of new psychotherapeutic and behavioral treatments for patients with OCD and possibly other mental health disorders, ”Castro Rodriguez suggested.
The next steps
The team continues to research this topic in several ways. “In this project, we also collected visualization data of subjects performing a task inside an MRI scanner. So our most immediate next step will be to search for neural correlates related to the transition from one strategy to another after receiving an explanation,” Castro said. -Radriguez.
“Pedro’s work is partly part of a larger laboratory, the Neurocomp project,” added co-author Bernardo Barahona-Correa, a psychiatrist with the Champalimo Foundation. “This project, which I am managing together with Albina, will explore many aspects of OCD, especially focusing on an area of the brain called the orbitofrontal cortex. We believe that this area is crucial both for the main manifestations of this disorder and for the acquisition of a model-based action control in tasks such as the one we used in this experiment. ”
“Ultimately, these results underscore the importance of explicit explanations in learning,” Oliveira-Maya noted. “It seems that pure free learning may not be the most effective way to gain new knowledge. In fact, I started talking to my children about it,” he playfully added, “and told them to make sure they pay attention to their teachers.”
Albinos Oliveira-Maya, Accurate knowledge of task structure is a major determinant of action based on the human model, Nature Human Behavior (2022). DOI: 10.1038 / s41562-022-01346-2. www.nature.com/articles/s41562-022-01346-2
Provided by the Champagne Center of the Unknown
Citation: To what extent are we able to find optimal solutions to complex problems? People may not be as accurate as is commonly assumed (2022, May 19), obtained May 19, 2022 from https://medicalxpress.com/news/2022-05-good-optimal-solutions-complex-problems. html
This document is subject to copyright. Except for any honest transaction for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.