The detection and recognition of two lines

These studies test how well the detection and recognition of two parallel lines may be understood in terms of independent detection of the individual lines. A two-alternative forced choice method was used; on each trial either one or two lines was presented and the subject had to make both a detection response (i.e. one of two intervals in some experiments, one of two orientations in others) and a recognition response (i.e. one or two lines). A basic independent detection model gives a good fit to the probability of detecting two lines separated by 14′ or more; however the observed probability of detection and recognition is slightly too low for a one-line stimulus and slightly too high for a two-line stimulus. A satisfactory overall fit can be obtained by modifying the model to take account of false positives, detectors responding simultaneously to both lines (bidetectors) and general variations of sensitivity from trial to trial. © 1981.