Fifty years older, fifty percent slower? Meta-analytic regression models and semantic context effects

Empiricist and rationalist approaches to data-quality control in meta-analyses are discussed, and a compromise approach advocated by Strube and Hartmann (1983) for theoretically driven meta-analyses is applied to Laver and Burke's (1993) data sets. When older adults' response times (RTs) and semantic priming effects (SPEs) are regressed on young adults' RTs and SPEs, the slopes of the regression lines are approximately 1.5 as long as the age difference is held constant and performance is reasonably accurate. Both increases and decreases in RTs following unrelated and related primes, relative to RTs following baseline primes, were analyzed. The results, like those regarding RTs and SPEs, support the hypothesis that, compared with young adults who are 50 years younger, older adults take approximately 50% longer to process lexical information.