The properties of the internal clock, temporal memory, and decision processes used to time short durations were investigated. The peak-interval procedure was used to evaluate the timing of 8-, 12-, and 21-s intervals, and analyses were conducted on the mean response functions and on individual trials. A distractor task prevented counting, and visual feedback on accuracy and precision was provided after each trial. Mean response distributions were (a) centered at the appropriate real-time criteria, (b) highly symmetrical, and (c) scalar in their variability. Analysis of individual trials indicated more memory variability relative to response threshold variability. Taken together, these results demonstrate that humans show the same qualitative timing properties that other animals do, but with some quantitative differences.

Fits of retention data were examined from 5 conditions: 3 types of cued recall, an old-new recognition task, and a remember-know recognition task. In each condition, 100 participants had either 18 recall or 27 recognition trials at each of 10 delays between 0 and 99 intervening items, providing the first data obtained in experimental psychology that were precise enough to distinguish clearly among simple functions. None of the 105 2-parameter functions tested produced adequate fits to the data. The function y = a^e~" 115 + a 2 e~I IT2 + a 3 fit each of the 5 retention conditions. The X2 parameter in this equation equaled 28 for the 3 recall conditions and the remember-know recognition condition and 13 for the old-new recognition condition. Individuals' recall data fit the same function with parameters varying with gender and scholastic aptitude scores. Reaction times support the claim that the aye'" 1 15 term describes working memory, and the remaining 2 terms describe long-term memory.The goal of mathematically describing retention is as old as the experimental study of memory (Ebbinghaus, 1885(Ebbinghaus, / 1964), yet no data exist that are precise enough to allow discrimination among the different mathematical functions commonly proposed. Rubin and Wenzel (1996) reviewed the substantial literature on existing retention functions. They found over 200 data sets in the literature and fit them all to 105 two-parameter functions. The data sets included the best available: All had 5 or more retention intervals and were smooth enough to correlate with at least one function .9 or greater. The following could be fit to four functions: recall, recognition, and sensorimotor tasks in people with retention intervals ranging from seconds to decades; delayed matching to sample in birds, rodents, and primates; and all other procedures and species, except autobiographical memory tasks. These functions were the logarithmic, y = bm • ln(t) (favored by Woodworth, 1938, and other early researchers); the power, y = b • t~m (favored by J. R. Anderson & Schooler, 1991;Rubin, 1982;Wixted & Ebbesen, 1991); the exponential in the square root of time, y = b-e'" 1^' (favored by Wickelgren, 1972); and the hyperbola in the square root of time, y = l/(b + m • V t) (previously unconsidered).

Several lines of evidence suggest that nicotine may be useful in treating the symptoms of Attention-Deficit/Hyperactivity Disorder (ADHD). The current study was an acute, placebo-controlled double-blind experiment to determine whether nicotine might be useful as an alternative treatment of adults with ADHD symptomatology. Six smokers and 11 nonsmokers who were outpatient referrals for ADHD were diagnosed by DSM-IV criteria. Measures of treatment effect included the Clinical Global Impressions (CGI) scale, Hopkins' symptom check list (SCL-90-R), the Profile of Mood States (POMS), Conners' computerized Continuous Performance Test (CPT), the Stroop test, and an interval-timing task. The smokers underwent overnight deprivation from smoking and were given a 21 mg/day nicotine skin patch for 4.5 h during a morning session. The nonsmokers were given a 7 mg/day nicotine skin patch for 4.5 h during a morning session. Active and placebo patches were given in a counter-balanced order approximately 1 week apart. Nicotine caused a significant overall nicotine-induced improvement on the CGI. This effect was significant when only the nonsmokers were considered, which indicated that it was not due merely to withdrawal relief. Nicotine caused significantly increased vigor as measured by the POMS test. Nicotine caused an overall significant reduction in reaction time (RT) on the CPT, as well as, with the smokers, a significant reduction in another index of inattention, variability in reaction time over trial blocks. Nicotine improved accuracy of time estimation and lowered variability of time-estimation response curves. Because improvements occurred among nonsmokers, the nicotine effect appears not to be merely a relief of withdrawal symptoms. It is concluded that nicotine deserves further clinical trials with ADHD.