ABSTRACTIn a recent paper Berglund et al. (1993) applied the triangle task in conjunction with economic incentives to assess the bias and sensitivity of panelists discriminating between food and beverage products. A re‐examination of the data from that study casts doubt on their interpretation of the results. Detection theory analyses of performance in sensory difference tests are increasingly used in food science and, where appropriate, permit independent bias and sensitivity measures to be extracted from discrimination data without the additional costs associated with the economic incentives tes