This paper introduces a method to improve the readability of time-frequency and time-scale representations by reassigning the values of these representations. The reassignment method, first applied to the spectrogram by Kodera, Gendrin, and de Villedary 15 years ago, is generalized to any bilinear time-frequency or time-scale distribution. The method moves the values of the representation away from their original locations to separate signal components, improving the concentration of these components and reducing misleading interference terms. The paper presents a new formulation of the reassignment method, discusses its theoretical properties, and demonstrates its practical application to various known time-frequency and time-scale distributions. Experimental results show that the reassignment method effectively enhances the readability of these representations, making them easier to interpret and more suitable for nonstationary signal analysis. The method is particularly useful for improving the spectrogram, Wigner-Ville distribution, Margenau-Hill distribution, and time-scale representations like the scalogram.This paper introduces a method to improve the readability of time-frequency and time-scale representations by reassigning the values of these representations. The reassignment method, first applied to the spectrogram by Kodera, Gendrin, and de Villedary 15 years ago, is generalized to any bilinear time-frequency or time-scale distribution. The method moves the values of the representation away from their original locations to separate signal components, improving the concentration of these components and reducing misleading interference terms. The paper presents a new formulation of the reassignment method, discusses its theoretical properties, and demonstrates its practical application to various known time-frequency and time-scale distributions. Experimental results show that the reassignment method effectively enhances the readability of these representations, making them easier to interpret and more suitable for nonstationary signal analysis. The method is particularly useful for improving the spectrogram, Wigner-Ville distribution, Margenau-Hill distribution, and time-scale representations like the scalogram.