By Beata Onikul, Strategist at Inizio Evoke
We recently attended the UCL Behaviour Change Conference in Lisbon, Portugal, where researchers and practitioners from around the world came together to explore the latest developments shaping the future of behavioral science.
Across the sessions, one message stood out: the field is evolving rapidly. From embracing systems thinking and precision interventions to improving how we measure behavior and navigate the opportunities—and challenges—presented by artificial intelligence (AI), behavioral science is moving beyond traditional approaches toward more dynamic, personalized, and scalable solutions.
Here are our key takeaways from the conference.
Behavior Change Is More Complex Than We Once Thought
Behavior change cannot be understood through simple, linear models. Human behavior exists within complex adaptive systems, where change is shaped by feedback loops, threshold effects, changing contexts, and iterative cycles rather than straightforward cause-and-effect relationships.
This also means context matters just as much as the intervention itself. The same intervention can produce very different outcomes depending on healthcare systems, social environments, policy, and individual circumstances. As a result, understanding heterogeneity of treatment effects—why interventions work differently for different people—is becoming an increasingly important research priority.
Thinking Beyond Individual Behaviors
Rather than targeting one behavior at a time, researchers are increasingly viewing behaviors as part of interconnected systems. Future work will need to better understand clusters of behaviors, model interactions between them, and consider how multiple behaviors influence one another over time.
This shift also supports the emergence of precision behavior change. Instead of designing interventions for the "average" person, researchers are looking to understand individual differences, explain why people respond differently, and develop more personalized approaches.
Better Measurement Will Drive Better Interventions
The conference also highlighted the need for more robust behavioral measurement. Traditional approaches have relied heavily on questionnaires, retrospective self-report, and static snapshots of behavior.
Increasingly, the field is discussing real-time behavioral measurement, digital and physiological data, and longitudinal monitoring. These approaches provide a richer understanding of behavioral dynamics while improving our ability to develop more precise interventions and strengthen predictive capabilities.
Time to Modernize Behavioral Theory
Many of the theories that underpin behavioral science were developed decades ago and may no longer fully reflect how behavior operates today. Several assume static behavior, focus on individual behaviors in isolation, and explain intentions better than actual behavior.
One example discussed was Social Cognitive Theory, which appears less able to explain behavioral variance among adults aged 35 and older. Future theories will need to better account for changing environments, behavior over time, and individual variability.
Healthcare Systems Need to Catch Up
A recurring message throughout the conference was that healthcare systems are not yet designed to routinely absorb behavioral interventions. Limited clinical buy-in, competing organizational priorities, resistance to change, and system design continue to hinder implementation.
Rather than focusing solely on improving interventions, there is an opportunity to better position behavioral science within healthcare and integrate behavior change into routine care. Ultimately, healthcare systems need to evolve to fit the evidence—not the other way around.
Expanding Behavioral Science's Impact
Speakers also emphasized the growing role behavioral science can play beyond healthcare. In the face of today's interconnected global challenges, there is significant opportunity to contribute to policy development, systems change, societal interventions, and population health.
The 2023 World Health Assembly resolution has created important momentum for behavioral science, but there is still considerable work to do before its full potential is realized across policy and public health.
AI Is Both an Opportunity and a Challenge
AI was another major theme. It offers significant opportunities to make behavioral science machine-readable, integrate data across disciplines, improve computational reproducibility, and build cumulative knowledge bases that accelerate research.
At the same time, AI reflects biases present in both training data and algorithms. As discussed during the conference, AI can act as a "macroscope," magnifying existing societal inequalities while making them more visible. The field still lacks systematic methods for identifying and addressing these biases, reinforcing the importance of fairness and responsible implementation.
Looking Ahead
The conference reinforced that the future of behavioral science lies in embracing complexity, improving measurement, modernizing theory, personalizing interventions, and embedding behavior change more effectively within healthcare systems and public policy. Achieving lasting impact will require not only stronger interventions, but also systems that are designed to support them.
Interested in hearing more? Connect with us here.