The Perceptual Anchor Model of Context Coding [Braidaet al., J. Acoust. Soc. Am.76, 722–731 (1984)], developed for auditory intensity, considers resolutions to be determined by both sensory and memory processes, and postulates that tasks differ in the memorial coding they require. The theory was tested with steady‐state synthetic vowels in the range /i/‐/I/‐/ɛ/. Resolution was measured in four discrimination conditions (two‐interval forced‐choice and same‐different, fixed‐, and roving‐level), and in complete identification conditions with and without a standard. In identification and roving‐level discrimination, resolution was poorer than in fixed‐level discrimination, but not uniformly so: sensitivity differences among tasks were smallest near category boundaries. The data suggest that some categorical phenomena can be understood in terms of perceptual anchors. [Work supported by NSF.]