No, threads work independently, so if they return the same information, it’s likely because they are accessing the same data.
Lowering the temperature will only limit the type of expressions the model uses, making the language more formal and causing it to use similar words in each iteration.
A higher temperature will make the response more complex and fluid, and if you repeat the conversation, you’ll get different comments and words.