Introduction. - A great number of authors have studied the links between cognition and emotion. More particularly, cognitive biases, which constitute erroneous information-processing, are implicated in the onset and maintenance of depression and other disorders. The most frequent identified were the four negative cognitive biases: catastrophizing, overgeneralizing, personalizing, and selective abstraction. The errors have been variously identified among depressed young people, anxious young people and young people experiencing both depression and anxiety. However, these cognitive biases exist in a non clinic population too. Aims of the study. - The objective of our study was to study the cognitive biases which can predict a non-pathologic anxiety in an adult population. Method. - A French version of the Beck Anxiety Inventory was administered to a sample of 50 non-clinic adults. Participants also completed 40 scenarii of daily life for measuring the cognitive biases. Participants were aged between 30 and 46 years and 20 were girls. Results. - Results were treated with regression analyses. The cognitive bias the most correlated to anxiety was the overgeneralizing bias that is the fact to induce a conclusion from only one example. The second cognitive bias links to anxiety was the arbitrary inference that is the fact to make a conclusion without real premises. Conclusion. - We discussed the interest to know the specific links existing between certain cognitive biases and anxiety. In particular, we discussed the fact that the mental health could be characterized by the frequency and the intensity of the cognitive biases. (C) 2015 Elsevier Masson SAS. All rights reserved.