General structure of the sound externalization framework. (a) Processing of the binaural signal x_L, x_R resulting in a cue-based error d_cue, a stage mapping the cue-based error to a normalized externalization rating e_cue, and a decision stage combining the cue-specific ratings to a final externalization rating E. (b) Processing of the binaural signal to calculate a single-cue error based on monaural cue templates. (c) Processing based on interaural cue templates. (d) Decision stage based on a static weighted-sum model (WSM). (e) Dynamic decision stage based on the selection of a cue-specific rating. Minimalist (LTA), conformist (MTA), and perfectionist (WTA) approaches were considered for selection (see text for further explanation).