Scientists at the California National Primate Research Center (CNPRC) have harnessed the power of machine learning, a branch of artificial intelligence (AI), to track anxiety-related behaviors in monkeys. Published in Nature’s Translational Psychiatry, the study demonstrated a significant link between nervous temperament in infant monkeys, as assessed by human observers, and later brain activity and behavior, as measured using machine-learning techniques. These findings suggest a strong connection between early-life nervousness and the eventual development of anxiety and depressive disorders.
Anxiety disorders, the most prevalent mental illnesses in the United States, impact nearly 20% of the population, disproportionately affecting women. Despite the availability of treatments, over half of those suffering do not seek help. Andrew Fox, a core scientist at CNPRC, believes this research could pave the way for preventing these disorders before they fully develop.
The Study
Fox, along with graduate student Dan Holley, utilized cutting-edge technology to monitor anxiety-related behaviors in 18 preadolescent female monkeys. Initially, human observers assessed the monkeys’ temperaments—identifying traits such as confidence, nervousness, and timidness. Two to three years later, the same monkeys underwent brain imaging while participating in behavioral tests designed to assess anxiety-related behaviors.
Traditionally, this type of experiment required multiple trained researchers to review each video meticulously to quantify behavior. Holley, however, developed a machine-learning technique to automate this process, significantly accelerating the research. “The hope is that machine learning will largely liberate researchers to focus on more interesting and appealing aspects of this work,” Holley explained.
Key Findings
The machine-learning approach revealed a strong correlation between infants identified as nervous and a specific anxiety-related behavior known as freezing, where the monkeys remained immobile for at least three seconds. This behavior was linked to increased metabolic activity in the central nucleus of the amygdala, a brain region crucial for threat processing. This finding replicated a relationship identified by researchers at the Wisconsin National Primate Research Center, suggesting the robustness of this connection across different environments.
Fox and Holley were struck by how accurately human observers could predict future behavior based on early temperament assessments. “When humans were making those observations, they were picking up something about the animal that we have not fully characterized,” Fox noted.
Future Implications
The study’s success has already led to funding for further research involving a larger cohort of 159 monkeys, both male and female.
“Ultimately, our hope is that by understanding the biology, we will be able to develop new behavioral or pharmacological treatments that could help alter the developmental trajectory of individuals with increased inhibited temperament during infancy, preventing the development of anxiety disorders that cause suffering later in life,” Fox said.
What This Means
This study marks a significant advance in understanding the long-term impacts of early-life anxiety. By integrating machine learning into behavioral research, CNPRC scientists have opened new avenues for exploring and potentially mitigating anxiety disorders. The team’s innovative approach and promising results highlight the potential for AI-driven tools to revolutionize mental health research and treatment strategies.
