This study examined the hypothesis that manifestly anxious Ss generalize expectancies (GEs) from recent past experience. The anxious S was seen as more constrained by the pushes and pulls of his most recent successes and failures. 80 Ss (40 males, 40 females) were divided into equal groups on the basis of high and low Manifest Anxiety scale scores. Following an initial estimate of GE, Ss were administered 4 tests of manual dexterity, with the value of positive or negative reinforcement controlled by E. At Test 5, a 2nd GE was requested, which was then followed by a change in the level of reinforcement (from positive to negative and vice versa). A test 6 GE was then requested, and change in GE was assessed through a D-score measure between the 2nd and 3rd GE estimates. The findings support the general hypothesis. There is some question about whether or not anxious Ss lower their GEs more absolutely following negative reinforcement on Test 5 than do nonanxious Ss. Anxious Ss clearly raise their level of GE following a positive reinforcement on Test 5 more absolutely than do the nonanxious Ss. (19 ref.) (PsycINFO Database Record (c) 2016 APA, all rights reserved)