Interpretation of significance levels by psychological researchers: the .05 cliff effect may be overstated.

Comments about previous studies indicate that the interpretation of significance levels by psychological researchers is unequivocally dictated by a binary decision-making framework. In particular, confidence in a p level would drop abruptly just beyond the fateful .05 level (cliff effect). A replication of Rosenthal and Gaito's (1963) experiment on the degree of confidence in p levels shows that these claims should be moderated. Detailed analysis of individual curves reveals that the attitude of researchers toward p values is far from being as homogeneous as might be expected. However, most psychological researchers in our study rated graduated confidence judgments, as either exponential or linear. Only a minority of all-or-none respondents exhibited an abrupt drop in confidence.

Bayesian predictive approach for inference about proportions.

This paper investigates the Bayesian procedures for comparing proportions. These procedures are especially suitable for accepting (or rejecting) the equivalence of two population proportions. Furthermore the Bayesian predictive probabilities provide a natural and flexible tool in monitoring trials, especially for choosing a sample size and for conducting interim analyses. These methods are illustrated with two examples where antithrombotic treatments are administrated to prevent further occurrences of thromboses.

Orientation of attention and sensory gating: an evoked potential and reaction time study in cat.

Somesthetic-evoked potentials (SEPs) were recorded in eight cats trained to perform a reaction time task (RT). The preparatory period (PP) preceding the RT began with the cat placing its paw in a box and ended 1.5 s later with a paw-withdrawal signal, an imperative stimulus (IS). In experiment 1 (E1), the IS took the form of a vibration in the box, whereas in experiment 2 (E2) a tone was used. The SEPs were elicited by electrical stimulation delivered to the paw once per trial and at a different moment during the last second of each PP. The SEPs were recorded in the lemniscus medialis (LM), the centrum medianum (CM), and over the anterior suprasylvian gyrus (ASSG). During the PP of E1 (four cats), SEPs recorded in the CM and over the ASSG were found to be significantly depressed. No significant change was noted, however, in those recorded in the LM. No significant SEP depression was found in any of the structures in E2 (four cats). Comparison between the two experiments revealed notable differences at both the thalamic and cortical levels. Moreover, SEP depression diminished as the moment for the IS approached. While, in E1, this occurred in the CM and over the ASSG, in E2 it was found in the CM only. Lemniscal and thalamic SEP amplitude was found to be correlated with performance. These findings are discussed in terms of orientation of attention, discrimination, and sensory gating.