In a seminal study, Bott and Noveck (2004) found that the scalar inference from ‘some’ to ‘not all’ was associated with increased sentence verifcation times, which they took to show that scalar inferences are associated with a processing cost. Recently, van Tiel, Pankratz, and Sun (2019) hypothesised that the presence of this processing cost critically depends on the polarity of the scalar word. We comprehensively and systematically evaluated this polarity hypothesis on the basis of a sentence-picture verifcation task in which we tested the processing of 16 types of adjectival scalar inferences. We develop a quantitative measure of adjectival polarity which combines insights from psychology and linguistics. In line with the polarity hypothesis, this measure of polarity reliably predicted the presence or absence of a processing cost, i.e., an increase in sentence verifcation times. We conclude that the alleged processing cost for scalar inferencing in verifcation tasks is not due to the process of drawing a scalar inference, but rather to the cognitive difculty of verifying negative information.