Contextual theories of judgment assume that the psychological impression of any particular stimulus depends upon its relative position in a context of similar stimuli. According to these theories, the impression of a neutral stimulus would be favorable when viewed in the context of very stimuli and unfavorable in the context of very stimuli. The neutralpoint anchoring demonstrated in this study contradicts this generalization. Contextual manipulations did affect ratings of the and negativeness of personality-trait words, but the neutral-point anchor (the descriptively neutral personality trait) was judged to be neutral in all contexts. Contextual manipulations caused stimuli to vary in their positiveness and stimuli to vary in their negativeness, but shifts across the neutral-point did not occur. These findings contradict both the adaptation-level premise that ratings vary as a function of shifts in the stimulus judged to be neutral and the range-theory assumption that ratings are formed in relation to the most extreme stimuli in the immediate context. The findings are consistent with a variation of the range theory that assumes that the neutral-point anchor is always judged to be neutral. Psychological impressions are relativistic. Stimuli are judged to be positive or negative (big or small, pleasant or unpleasant, heavy or light, etc.) in comparison with some frame of reference or standard of comparison. In some instances, as with judgments of socially relevant stimuli, the frame of reference is based largely upon previous experience; in other situations, as with psychophysical tasks in which stimuli vary along a single physical dimension, this basis is the immediate context formed by the other stimuli. For example, subjects asked to judge the goodness or badness of a moral act bring some frame of reference to the task, but subjects asked to judge the largeness or smallness of an isolated square must be exposed to the different sizes of squares in the experiment before a stable frame of