Teaching a child to be kind isn’t about repeating “be nice,” according to two studies: an experiment found preschoolers helped up to 29 percent more when adults called them “a helper” rather than asking them to “help,” and a separate study found praising a child’s character after a kind act produced

New research suggests framing children's behavior through identity, not just instruction, boosts positive actions. Labeling a child 'a helper' increased helpfulness, and praising their character, not just their actions, fostered lasting kindness.
Key takeaways
- Identity-based language encourages desired AI behavior.
- Praise for AI's 'character' could lead to better long-term function.
- Prompting AI with roles enhances task completion.
- Focus on AI's inherent capabilities, not just commands.
Why it matters
Understanding how identity influences behavior can help AI users frame prompts more effectively. Instead of just asking for a task, defining the AI's role (e.g., 'you are a summarizer') might yield better results and more consistent performance.
Try this on SynaBot
Related AI assistants, prompts, and tools from the SynaBot catalog.
- NICE CXone AINICE CXone AI offers a comprehensive suite of AI capabilities for contact centers, including smart routing, self-service bots, and agent coaching. It utilizes machine learning to optimize every customer interaction. The platform helps drive personalized experiences and operational efficiency.
- NICE CXone BotNICE CXone offers AI-powered bots that handle routine inquiries, collect information, and route complex cases. It helps decrease call volumes, improve efficiency, and enhance customer satisfaction in contact centers.
- NICE CXoneNICE CXone incorporates AI across its cloud contact center platform for intelligent routing, agent assist, and conversational AI. It optimizes every customer interaction and empowers agents with real-time insights. Enhances overall customer experience and operational efficiencies.


