We introduce a new type of 3D text-based CAPTCHA, called DotCHA, which relies on human interaction and overcomes the limitations of existing 2D and 3D CAPTCHAs. DotCHA asks users to rotate a 3D text model to identify the correct letters. The 3D text model is a twisted form of sequential 3D letters around a center pivot axis, and it shows different letters depending on the rotation angle. Because each model consists of many small spheres instead of a solid letter model, DotCHA is classified as a scatter-type CAPTCHA and resists character segmentation attacks. Moreover, DotCHA is resistant to machine learning attacks because each letter is only identified in a particular direction. We demonstrate that DotCHA is resistant to existing types of attacks while maintaining usability.