mental Health Advice Ofen Overlooks a Crucial Step: Subtraction
Table of Contents
London,England – August 21,2025 – Conventional wisdom in mental health often focuses on adding positive activities-like exercise or mindfulness-but groundbreaking research indicates a significant blind spot: the failure to advise individuals to stop harmful behaviors. This “additive advice bias,” as researchers call it, permeates conversations, social media, and even artificial intelligence recommendations, potentially leaving people feeling more burdened than supported.
The Pervasive Bias Towards ‘More’
As mental health challenges escalate globally, with services facing increasing strain, individuals frequently turn to friends, family, online communities, and AI for guidance. Understanding the nuances of how advice is given is critical to improving its effectiveness. A comprehensive study, encompassing eight experiments and involving hundreds of participants, analyzed advice given in various contexts-from strangers seeking help to self-reflection-and assessed responses from ChatGPT. The research, published in Communications Psychology, revealed a consistent pattern: people overwhelmingly suggest adding activities rather than subtracting detrimental ones.
Participants were asked to advise others on scenarios involving both negative habits, such as excessive gambling, and missed opportunities for beneficial activities, like regular exercise. The findings consistently demonstrated a preference for additive solutions.
Did You Know? …
that the World health Organization estimates that nearly one billion people live with a mental disorder?
Key Findings Summarized
| Finding | Description |
|---|---|
| Additive Dominance | Across all scenarios,adding activities was far more common than suggesting the removal of harmful ones. |
| Perceived Feasibility | Participants generally viewed adding activities as easier and more beneficial than eliminating negative behaviors. |
| Relational Variance | Removing harmful habits was perceived as easier to suggest for close friends than for oneself. |
| AI Replication | ChatGPT exhibited the same additive bias, mirroring patterns observed in online social media advice. |
Expert Insights on the Implications
“In theory, effective advice shoudl strike a balance between encouraging positive actions and discouraging negative ones. Though, we consistently found a tendency to simply add more to people’s already full plates, and even AI has learned to do so. While well-intentioned, this approach can inadvertently make mental health feel like an endless to-do list,”
Dr. Tom Barry, Senior Author, Department of Psychology, University of bath, England
Dr. Nadia Adelina, co-author from the Department of Psychology at the University of Hong Kong, emphasized the growing role of AI in mental health support. “As AI chatbots become increasingly prevalent as a source of mental health guidance, they risk amplifying this bias,” she stated. “Incorporating prompts that encourage exploration of what individuals might remove from their lives could lead to more balanced and less overwhelming advice.”
Pro Tip: …
When seeking or offering mental health advice, consciously consider what habits or activities could be reduced or eliminated.
Do you think the rise of AI in mental health will exacerbate existing biases in advice-giving? What strategies can individuals use to ensure they receive truly balanced support?
the Research Behind the Findings
The research, supported by the Research Promotion Fund of the Department of Psychology at the University of Bath, England, analyzed experimental data, real-world advice from Reddit, and ChatGPT responses.The study’s findings highlight a essential flaw in how we approach mental health support, suggesting a need for a more nuanced and subtractive approach. This aligns with broader research on decision-making, which demonstrates that people often struggle to consider the benefits of removing options, even when those options are detrimental [[1]].
Evergreen Context: The Growing Mental Health Crisis
The global mental health crisis is a pressing issue, with rates of anxiety, depression, and other mental health conditions rising steadily in recent years. Factors contributing to this trend include increased stress, social isolation, economic uncertainty, and the pervasive influence of social media.The COVID-19 pandemic further exacerbated these challenges, leading to a surge in mental health needs worldwide.Addressing this crisis requires a multi-faceted approach, including increased access to mental health services, greater awareness and destigmatization, and a shift towards preventative care. The findings of this research underscore the importance of providing balanced and effective support, recognizing that sometimes, the most helpful advice is to simply stop doing something harmful.
Frequently Asked Questions
- What is additive advice bias? It’s the tendency to suggest adding activities to improve mental health, rather than removing harmful ones.
- Why is this bias problematic? It can overwhelm individuals and make mental health feel like an endless list of tasks.
- Does AI exhibit this bias? Yes,ChatGPT was found to give predominantly additive advice.
- How can we overcome this bias? By consciously considering what habits or activities could be reduced or eliminated.
- What are the implications for mental health support? Support systems should focus on both adding positive behaviors and subtracting negative ones.
We hope this article has provided valuable insight into the importance of balanced mental health advice. Please share this data with your friends, family, and community to help raise awareness and promote more effective support. We also encourage you to leave a comment below with your thoughts and experiences. And don’t forget to subscribe to our newsletter for more insightful articles on health, wellness, and current events.