Experiments are reported the aim of which was to elucidate the cause of each of the components of the lambda response, and particularly to evaluate the role of ‘on’ and ‘off’ visual effects which appear at various times during the oculomotor process and also the possible influence of non-visual mechanisms. Eight subjects with normal sight were studied under the following conditions: (i) horizontal eye movements of 12° were guided by fixation points placed on a dimly-lit uniform black field of 20°; a checkerboard of 6° aperture was placed in this field so as to be integrated into the oculomotor process at different times—at the beginning, during and at the end of the eye movement; (ii) successive horizontal eye movements of 3°, 7° and 11° scanned a checkerboard of 20°, each square of which had a 40′ aperture; (iii) the checkerboard was moved with an amplitude and period similar to those of the eye movements in (ii), but this time with gaze fixed. Horizontal and vertical movements of both eyes were recorded with an EOG. An EEG of the parieto-occipital regions was obtained using eight linked bipolar derivations in line on two montages, median longitudinal and right-left transverse. The EEG and EOG data were digitalized and a numerical programme of waveform recognition was used to identify the beginning of the saccade which triggers the averaging out of the EEG before (100 ms) and after (500 ms) the eye movement. A discussion of the results, taking into account the latency of the different components and their reinforcements or inhibition depending on experimental conditions, suggests that the two initial components of lambda response (including the initial portion of the classical lambda wave) might be due to visual effects (‘off effect’) that arise at the start of the movement or slightly before it at the time that the saccadic suppression begins. The later components could be attributed to visual effects brought into play towards the end of the movement (‘on effect’), when perception becomes normal again. It is, however, difficult to explain some of the results related to the amplitude of lambda components without bringing in a mechanism of non-visual origin (corollary discharge).
