Psychology Science, Volume 48 (2006), Issue 1

This issue opens with work by Oliver Dickhäuser and Wulf-Uwe Meyer on how boys and girls differ when they explain their own performance in mathematics. Even at a young age, children tend to credit or blame different causes for how well they do with numbers, and those explanations do not fall out evenly by gender. The authors look at what young students point to when a math task goes right or wrong.

Achim Elfering and Victor Sarris turn to memory and the way recall bends toward its surroundings. Using a delayed matching-to-sample procedure, they examine how a stimulus held in mind drifts toward the context it sits in once a delay is introduced. The question is whether what we remember stays faithful to the original or quietly assimilates to whatever frames it.

Margit Gramer studies people high in social anxiety and how their bodies respond when they are asked to cope actively with a demanding situation. Cardiovascular measures give a window into what happens physiologically during these moments, letting the research connect a social fear to measurable changes in the heart and circulation rather than self-report alone.

A paper by José Santacreu, Víctor J. Rubio and José M. Hernández returns to an older idea from Cattell: so-called T-data, personality information gathered from a person's actual behavior in objective tasks rather than from questionnaires or ratings. The authors revisit this objective route to measuring personality and push it a little further than Cattell left it.

Alexander von Eye and Eun-Young Mun take up the problem of how two raters line up, or fail to. Instead of collapsing everything into a single agreement coefficient, they map out the specific patterns where raters agree and where they part ways, treating configurations of agreement and disagreement as information in their own right.

Closing the issue, Martin Kersting and Lutz F. Hornke describe a German approach to standardizing proficiency assessment and raising its quality. Their account deals with how tests of ability can be built and calibrated so that the results mean something consistent across takers and settings.