Arthur J. Riopelle

On Confidence Intervals for Within-Subjects Designs.

Confidence intervals (CIs) for means are frequently advocated as alternatives to null hypothesis significance testing (NHST), for which a common theme in the debate is that conclusions from CIs and NHST should be mutually consistent. The authors examined a class of CIs for which the conclusions are said to be inconsistent with NHST in within-subjects designs and a class for which the conclusions are said to be consistent. The difference between them is a difference in models. In particular, the main issue is that the class for which the conclusions are said to be consistent derives from fixed-effects models with subjects fixed, not mixed models with subjects random. Offered is mixed model methodology that has been popularized in the statistical literature and statistical software procedures. Generalizations to different classes of within-subjects designs are explored, and comments on the future direction of the debate on NHST are offered.

Circumferential adjustment of ultrasound probe position to determine the optimal approach to the internal jugular vein: a noninvasive geometric study in adults.

Circumferential adjustment of the position of a two-dimensional ultrasound (US) probe around the neck has been recommended as a strategy for reducing the potential for unintentional common carotid artery puncture during internal jugular venous (IJV) cannulation. We obtained multiple high-resolution US images bilaterally from the necks of 107 adult subjects and analyzed these to determine the degree to which this strategy permits identification of a pathway from the skin to the IJV that minimizes venoarterial overlap while maximizing venous target (angular) width. The method consistently permitted identification of an approach to the IJV superior to that obtainable with any one of four popular surface anatomy-based (“blind”) approaches and was even more powerful if used in concert with a US-guided 1) adjustment of the degree of head rotation, 2) choice between a high and low approach, and 3) choice between the right and left IJV. Use of a high-resolution US imaging device also permitted identification of the precise boundaries of additional cervical anatomic structures (nontarget vessels, lymph nodes, and the thyroid gland) potentially relevant to selection of an optimal approach to the IJV.

Treatment of the cutaneous pain of acute herpes zoster with 9% lidocaine (base) in petrolatum/paraffin ointment

BACKGROUND Treatment of the pain of acute herpes zoster by local anesthetic injections has drawbacks. Topical percutaneous local anesthesia (TPLA) may offer another strategy of providing regional analgesia in affected patients. OBJECTIVE We evaluate the analgesic efficacy and safety of 9% (wt/vol) lidocaine (base) in petrolatum/paraffin ointment in patients with acute herpes zoster. METHODS Ointment was applied to the affected skin of 22 patients. Pain, tenderness, sensitivity to pinprick and cold, and blood lidocaine concentration were measured repeatedly during a 20-hour interval and intermittently thereafter. RESULTS Mean pain, tenderness, and cutaneous sensation scores were reduced at measurements taken from 4 to 20 hours after ointment application (p < 0.05), but not every patient obtained relief. No patient had local skin irritation or systemic toxic effects related to the local anesthetic. CONCLUSIONS TPLA is a promising therapy for control of cutaneous pain of acute herpes zoster. Controlled studies should be performed to prove efficacy, determine optimal TPLA formulation, and define dosage limits.

Are Effect Sizes and Confidence Levels Problems For or Solutions To the Null Hypothesis Test

Abstract Some have proposed that the null hypothesis significance test, as usually conducted using the t test of the difference between means, is an impediment to progress in psychology. To improve its prospects, using Neyman-Pearson confidence intervals and Cohens standardized effect sizes, d, is recommended. The purpose of these approaches is to enable us to understand what can appropriately be said about the distances between the means and their reliability. Others have written extensively that these recommended strategies are highly interrelated and use identical information. This essay was written to remind us that the t test, based on the sample—not the true—standard deviation, does not apply solely to distance between means. The t test pertains to a much more ambiguous specification: the difference between samples, including sampling variations of the standard deviation.

PROTEIN DEPRIVATION IN PRIMATES: VI. FOOD PREFERENCES OF ADULT RHESUS MONKEYS MAINTAINED ON LOW-PROTEIN DIETS

Three groups of adult female rhesus monkeys, maintained on low-protein diets (.5 gm, 1 gm, and 2 gm protein per kg body weight) were compared with a control group (4 gm protein per kg body weight) on a food-preference task. Food responsiveness was assessed by presenting 8 small pieces of a certain food, equally spaced about the perimeter of a turntable attached to the home cage, and recording number of pieces taken, number of pieces eaten, and elapsed time for taking all 8 pieces. 21 different foods were used in sequence, 3 each from the following 7 categories: cheese, meat, vegetable, nut, cereal, fruit, and candy. Scores on all 3 measures were highly correlated, and the order of preference was generally the same for all groups. The 2 lowest-protein groups accepted more foods at the lower end of the palatability spectrum than did either the 4-gm or the 2-gm group. There was a tendency for the foods least preferred by the protein-deprived monkeys to be themselves high in protein. Thus, although protein deprivation appears to increase the catholicity of food preference, there is no corresponding increase in the relationship between palatability and protein content.

PRIMATES IN NUTRITIONAL AND DEVELOPMENTAL RESEARCH

Supported in part by a grant from the National Institute of Child Health and Human Development, NIH (HD-07470).

The Difference Between t and z and the Difference It Makes

Tests of the null hypothesis for comparisons involving sample means use the t test when the conditions of the z test cannot be met. The 2 tests have different rationales and can lead to different conclusions regarding significance. In the present study, the authors compared the properties of t and z in simulation runs. The differences in the results are a result of fluctuations in the t test sample variances that do not exist in the z test, and those differences lead to differences in designating the significance of comparisons.

Probability-reward preferences of rhesus monkeys

The relative preferences of rhesus monkeys for reward probability versus amount were investigated with procedures which contrasted general experience with specific instructions, and evaluated the relationship between probability-amount combinations and preference strength. Four stimulus objects, each signifying a different combination of reward frequency and amount (100% with one unit; 50% with two units; 33% with three units; or 25% with four units), were presented in pairs, one pair per daily session, with trial schedules providing the same amount of reward within each set of 12 trials. In Phase A, 4 monkeys (Group 1) were tested on the six choice-pairs with no preliminary training. In Phase B, Group 1 was joined by an additional 4 monkeys (Group 2), and each of the tasks was preceded by a demonstration of the relevant stimuli, one at a time, together with their associated probabilities and amounts. Group 1 animals developed preferences during Phase A for the more frequently rewarded objects, which persisted into Phase B, whereas Group 2 animals showed no preferences. This result indicates that preliminary instructions concerning the reward combinations associated with stimulus objects can prevent the development of a preference for greater probability over greater amount of reward, but cannot extinguish it once it has been formed or reestablished within the context of a particular task.

Protein deprivation and food-related risk-taking preferences of rhesus monkeys

The relative preferences of four rhesus monkeys for reward probability versus reward amount when they were maintained on a low-protein (3.35%) diet were compared with those demonstrated when they were fed an adequate (13.4%) protein diet. Four stimulus objects, each signifying a different combination of reward frequency and amount (100%-one piece, 50%-2 pieces, 33%-three pieces, or 25%-four pieces), were presented in pairs, one pair per daily session, with trial schedules providing the same amount of reward within each set of 12 trials. Selections of the more frequently rewarded objects, but with lesser amounts per trial, were significantly higher during the low-protein phase than during either the preceding or the following normal-diet phases. Protein deprivation produces a changed motivational state making these animals less tolerant of infrequent or postponed reinforcement.

Functional Anatomy of the Null Hypothesis and of Tests of It

Abstract The author compared simulations of the “true” null hypothesis (z) test, in which ò was known and fixed, with the t test, in which s, an estimate of ò, was calculated from the sample because the t test was used to emulate the “true” test. The true null hypothesis test bears exclusively on calculating the probability that a sample distance (mean) is larger than a specified value. The results showed that the value of t was sensitive to sampling fluctuations in both distance and standard error. Large values of t reflect small standard errors when n is small. The value of t achieves sensitivity primarily to distance only when the sample sizes are large. One cannot make a definitive statement about the probability or “significance” of a distance solely on the basis of the value of t.