Experimental studies of gaming have dealt with task, situational, and personality variables. Most studies have focused on either dyadic matrix games (with the Prisoner's Dilemma as a prototype) or triadic competitive games (with the Board Game as a prototype). Types of variables in each category are: (1) task variables matrix variations, mode of presentation, length of run, threat, and power relationships; (2) situational variables strategy variations, instructions, feedback and communication, character of opponent, and reward; and (3) personality variables general differences between Ss, family background, psychopathology, attitudes and traits, and motives. In dyadic games, level of cooperation is significantly influenced by conditions in all 3 categories. In triadic games, similarly, all 3 classes of variables influence coalition formation, strategy, and outcomes. The following problems arise: comparison of game-situations, effects of situational constraints, demand characteristics, relevance to social interaction, relevance of variables, interactional effects, and variations in group size. Strictly rational theories emphasize task (and situational) variables, in contrast to theories which emphasize perceptual, cognitive, and motivational processes (personality and situational) variables. Neither kind of theory is sufficient, and a field theory is preferable. (4 p. ref.) (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1969 American Psychological Association.