Three clearly distinguishable lists were given simultaneously for learning. In one condition, subjects recalled all three lists; in a second condition, each list was given a different type of retention test although one of the lists was recalled. All subjects were fully informed of the materials and tests. The critical interest was in the recall of the same list in the two conditions. Several lines of thought led to the expectation that recall would be better when three different retention tests were used than when all lists were recalled. Three experiments gave little support to the expectation. To enhance differential encoding of the lists to match the retention tests, a fourth experiment was conducted that included a condition in which subjects learned each list separately before simultaneous learning. No evidence for differential encoding was found. There was relatively little transfer from single-list to simultaneous learning of three lists represents a marked change in context. (Author)