The present study sought to investigate to what extent patients with chronic low back pain and pain-free control subjects selectively attend to pain-related stimuli as measured with 2 dot-probe tasks with word stimuli and pictorial stimuli. Selective attentional processing was measured by means of 3 indices: the bias index, a congruency effect, and an incongruency effect. Pain-related fear as a trait measure (Tampa Scale for Kinesiophobia [TSK]) was expected to be positively associated with all indices of selective attentional processing of pain stimuli. Results were analyzed with repeated-measures analysis of variance. An incongruency effect was found for patients and to a significantly less degree for pain-free control subjects on the dot-probe task with pictorial stimuli, indicating that pain patients have difficulty disengaging from threat pictures. Pain-related fear as a trait measure (TSK) was not associated with selective attentional processing of word and pictorial stimuli in either pain patients or control subjects. Results from the present study are discussed, and directions for future research are provided. PERSPECTIVE: Demonstrating difficulty to disengage from threat might be clinically relevant because patients might pay less attention to fear-disconfirming information and remain engaged in avoidance, which might eventually lead to prolonged anxiety states.