Most people seem to comprehend metaphorical expressions with little effort, and such expressions abound in everyday usage. However, if one were to analyze the components of metaphorical expressions, it becomes clear that metaphor cannot occur because of a one-to-one relationship between the features that constitute it. The characteristics of a billboard do not match those of warts, but the expression "Billboards are warts" is a meaningful metaphor. An experiment was advanced from the belief that people use distinct processing strategies when dealing with verbal information that ranges from literal to metaphorical. Under three different instructional sets, participants verified true or false sentences for literal truth, possible sense, or metaphoricity in a response-time task. The time needed for processing varied as a function of instructional set. When processing in a literal mode, participants verified true sentences quickly and accurately, but false sentences took time to deny as true, and metaphors were likely to be taken as true. When in a possible-sense mode, participants required slightly more decision time and identified standard false sentences as nonsense. When in a metaphoricity mode, participants used the most decision time in an explicit search for metaphorical meaning. People make use of a variety of processing strategies that enable them to deal with a wide range of information. Models of metaphor comprehension that either assume that metaphors are automatically understood or are the product of the reinterpretation of literal meaning are not mutually exclusive. (Monica Gregory and Nancy Mergler)