A new study in Neuroscience of Consciousness reveals a surprising finding: simply believing you are working with a machine can lower your confidence in decisions, even when your judgments are correct. The research shows that human–machine interactions shape self-belief in ways that may affect everyday choices at work and in learning environments.
This insight is timely as Thailand expands its tech ecosystem. Thai educators, students, and professionals are increasingly using AI in classrooms, clinics, and offices. Understanding how perceived machine collaboration influences confidence could inform the design of human-centered technologies in education and health services.
Led by a team that includes a senior scientist from a renowned French research center and a major university, the study asked participants to decide whether to change their initial answer after receiving feedback from a supposed partner. They performed a perceptual task involving judging motion direction and were told their partner was either a machine or a human. Across conditions, confidence—not the accuracy of the first choice—predicted whether participants revised their answers. The results suggest internal confidence often drives decisions more than task difficulty, regardless of whether the partner is human or machine.
Researchers found that participants tended to have lower confidence when they suspected their partner was a machine, yet their performance remained steady. This bias may reflect an assumption that machines are inherently more accurate, a belief increasingly common in technology-rich environments.
In Thailand’s context, where digital education resources and virtual assistants are becoming common, these insights point to practical steps. Thai schools and workplaces could integrate confidence-building features into e-learning platforms and AI-assisted tools to preserve student autonomy and critical thinking. Ensuring learners feel capable while using technology is essential for sustaining innovation and resilience.
Physiological data from the study linked confidence to observable markers such as pupil dilation and blink rate. If similar patterns emerge in Thai settings, these signals could guide adaptive teaching or work tasks in future smart classrooms and AI-enabled offices, aligning challenges with learners’ or workers’ confidence while keeping a human-centered approach.
The findings also invite reflection on Thai cultural dynamics. Respect for teachers and expert opinions remains strong in Thai society, and technology is often viewed as a trusted partner in knowledge work. As AI tools become more integrated, policymakers and educators should balance perceived machine accuracy with reinforcement of human expertise to preserve the value of professional judgment.
For decision-makers, the study offers a reminder: as Thailand accelerates digital transformation, design and training should address the subtle effects of machine perception on confidence. This approach helps maintain morale and decision-making effectiveness while embracing AI’s benefits.
Ultimately, Thai education and health policies should weave collaboration training with AI that reinforces both human skills and technology. Building public understanding and trust will be crucial to ensuring AI supports, rather than diminishes, human capability.