Identifying subgroups that succeed or fail with three levels of physical activity intervention: the Activity Counseling Trial.

The authors used recursive partitioning methods to identify combinations of baseline characteristics that predict 2-year physical activity success in each of 3 physical activity interventions delivered in the multisite Activity Counseling Trial. The sample consisted of 874 initially sedentary primary care patients, ages 35-75 years, who were at risk for cardiovascular disease. Predictors of 2-year success were specific to each intervention and represented a range of domains, including physiological, demographic, psychosocial, health-related, and environmental variables. The results indicate how specific patient subgroups (e.g., obese, unfit individuals; high-income individuals in stable health) may respond differently to varying levels and amounts of professional assistance and support. The methods used provide a practical first step toward identifying clinically meaningful patient subgroups for further systematic investigation.     

Deconstructing the Tower of London: Alternative moves and conflict resolution as predictors of task performance

Despite widespread use the cognitive demands of the five-disc Tower of London (TOL) are unknown. Research suggests that conflict moves (those that are essential to the solution but do not place a disc in its final position) are a key aspect of performance. These were examined in three studies via a verification paradigm, in which normal participants were asked to decide whether a demonstrated move was correct. Experiment 1 showed that individual move latencies increase with the number of intermediate moves until the disc is placed in its goal position (resolution). Post hoc tests suggested that the number of alternative moves and moves to resolve a disc were independent predictors of performance. Experiment 2 successfully manipulated these factors in an experimental design. Experiment 3 showed that they remain determinants of performance as familiarity increased. Overall, errors on the task were significantly correlated with spatial memory. The implications of these findings for the use of the TOLin cognitive psychology and as an assessment tool are discussed.