Overview
The O'Connell effect is an asymmetry in the photometric light curve of certain close eclipsing binary stars. It was named after the astronomer Daniel Joseph Kelly O'Connell, SJ of Riverview College in New South Wales who in 1951 studied this phenomenon and distinguished it from the so-called periastron effect described by earlier authors, as it does not necessarily appear near the periastron, when tidal effects and an increase in mutual radiation may cause an increase in luminosity.
The effect
The out-of-eclipse brightness maxima of some binary stars are unequally high. This is contrary to expectations that the observed luminosity of an eclipsing binary should be the same when its components switch positions every half period. The maximum following the primary minimum is nearly always brighter than the preceding one. This is called the positive O'Connell effect, the reverse case is referred to as the negative O'Connell effect. The difference increases with the ellipticity of the stars, and the differences in their sizes and densities. Also, spectral differences have been observed between subsequent maxima.
Attempts at explanation
From Wikipedia (CC BY-SA 4.0).