Dual-process theory hypothesizes that recognition memory depends on two distinguishable memory signals. Recollection reflects conscious recovery of detailed information about the learning episode. Familiarity reflects a memory signal that is not accompanied by a vivid conscious experience but nonetheless enables participants to distinguish recently-experienced probe items from novel ones. This dual-process explanation of recognition memory has gained wide acceptance among cognitive neuroscientists and some cognitive psychologists. Nonetheless, its difficulty in providing a quantitatively satisfactory description of performance in item recognition experiments has precluded a consensus not only about the theoretical structure of recognition memory but also about how to best measure recognition accuracy. In two experiments we show that neither the standard formulation of dual-process signal detection theory (DPSD) nor a widely-used single-process model (UVSD) provides a satisfactory explanation of recognition memory across different types of study materials (words and travel scenes). In the variable recollection dual-process model (VRDP), recollection fails for some old probe items, as in standard formulations of dual process signal detection theory, but gives rise to a continuous distribution of memory strengths when it succeeds. The VRDP can approximate both the DPSD and the UVSD. In both experiments it provides a consistently superior fit across materials to the superset of the DPSD and UVSD. The VRDP offers a simple explanation of the form of conjoint item-source judgments, something neither the DPSD nor the UVSD can accomplish. The success of the VRDP supports the core assumptions of dual-process theory by providing an excellent quantitative description of recognition performance across materials, response criteria and type of response.

A commentary on this article appears in this issue on page 267.

Item difficulty effects in skill learning were examined by giving participants extensive training with repeated alphabet arithmetic problems that varied in addend size (e.g., C 2 D 5 ? is easy; C 2 J 5 ? is harder). Recognition memory for the items, as measured by interpolated recognition tests, was acquired early in training and was unaffected by item difficulty. Memory for the solutions to items, as measured by the participants' strategy reports that they had retrieved, rather than computed, the solution, was acquired later and was affected by item difficulty. Solutions to easier items were learned earlier in training for both young adults (18-24 years) and older adults (60-75 years), superimposed on an overall lower level of solution learning in older participants. The results suggest that the formation of associations between problems and their solutions is effortful and shares limited processing resources with the computational demands of the problem.

No abstract

Path analysis was used to examine the relationship between class start times, sleep, circadian preference, and academic performance in college-aged adults. Consistent with observations in middle and high school students, college students with later class start times slept longer, experienced less daytime sleepiness, and were less likely to miss class. Chronotype was an important moderator of sleep schedules and daytime functioning; those with morning preference went to bed and woke up earlier and functioned better throughout the day. The benefits of taking later classes did not extend to academic performance, however; grades were somewhat lower in students with predominantly late class schedules. Furthermore, students taking later classes were at greater risk for increased alcohol consumption, and among all the factors affecting academic performance, alcohol misuse exerted the strongest effect. Thus, these results indicate that later class start times in college, while allowing for more sleep, also increase the likelihood of alcohol misuse, ultimately impeding academic success.