An objective theory of statistical testing

Theories of statistical testing may be seen as attempts to provide systematic means for evaluating scientific conjectures on the basis of incomplete or inaccurate observational data. The Neyman-Pearson Theory of Testing (NPT) has purported to provide an objective means for testing statistical hypotheses corresponding to scientific claims. Despite their widespread use in science, methods of NPT have themselves been accused of failing to be objective; and the purported objectivity of scientific claims based upon NPT has been called into question. The purpose of this paper is first to clarify this question by examining the conceptions of (I) the function served by NPT in science, and (II) the requirements of an objective theory of statistics upon which attacks on NPT's objectivity are based. Our grounds for rejecting these conceptions suggest altered conceptions of (I) and (II) that might avoid such attacks. Second, we propose a reformulation of NPT, denoted by NPT*, based on these altered conceptions, and argue that it provides an objective theory of statistics. The crux of our argument is that by being able to objectively control error frequencies NPT* is able to objectively evaluate what has or has not been learned from the result of a statistical test.     

Positive contrast obtained in reacquisition following interpolation of nonreinforced or partially reinforced trials

Two experiments investigated the effects of shifting from either nonreinforcement or partial reinforcement (PRF) to continuous reinforcement conditions (CRF). In the first experiment, three groups of rats were given food reinforcement under CRF conditions in a runway followed by regular extinction trials (RE), extinction trials where Ss were delayed for 30 sec before entering the empty goal box (DE), or CRF trials where Ss were delayed for 30 sec before entering the baited goal box (DF). Then all Ss were run on the delayed reinforcement condition (DF). In the final delayed reinforcement condition, group DE ran significantly faster than group DF, reflecting positive contrast. In the second experiment, four groups of rats were trained in a runway to receive either 4% or 18% sucrose reinforcers under either PRF or CRF conditions. Then all Ss were transferred to a Skinner box and bar presses were continuously reinforced, with each S continuing to receive the same sucrose concentration as before. The former PRF Ss, regardless of the reinforcer, bar pressed at a significantly higher rate in the Skinner box than the former CRF Ss. The evidence seemed to favor the view that the effectiveness of a reinforcer is not an absolute, unchanging quantity but rather depends on the historical context in which the reinforcer occurs.     

An eye movement analysis of the spatial contiguity effect in multimedia learning

In three studies, eye movements of participants were recorded while they viewed a single-slide multimedia presentation about how car brakes work. Some of the participants saw an integrated presentation in which each segment of words was presented near its corresponding area of the diagram (integrated group, Experiments 1 and 3) or an integrated presentation that also included additional labels identifying each part (integrated-with-labels group, Experiment 2), whereas others saw a separated presentation in which the words were presented as a paragraph below the diagrams (separated group, Experiments 1 and 2) or as a legend below the diagrams (legend group, Experiment 3). On measures of cognitive processing during learning, the integrated groups made significantly more eye-movements from text to diagram and vice versa (integrative transitions; d = 1.65 in Experiment 1, d = 0.85 in Experiment 2, and d = 1.44 in Experiment 3) and significantly more eye-movements from the text to the corresponding part of the diagram (corresponding transitions; d = 2.02 in Experiment 1 and d = 1.35 in Experiment 3) than the separated groups. On measures of learning outcome the integrated groups significantly outperformed the separated groups on transfer test score in Experiment 1(d = .80) and Experiment 2 (d = .73) but not in Experiment 3 (d = .35). Spatial contiguity encourages more attempts to integrate words and pictures and enables more successful integration of words and pictures during learning, which can result in meaningful learning outcomes.     

Encoding variability theory and the spacing effect in associate learning

Encoding variability theory accounts for the spacing effect by assuming that, as the lag between repetitions increases, the memorial representations approach independence. A method for testing this assumption in determining repetition effects in associative learning is suggested. In the independence hypothesis, we simply assume that each presentation of a word pair is represented independently, so that the expected proportion recalled is simple, P=P1+P2 - PIP2. Two studies are reported using a continuous paired-associate task representing a factorial combination of lag and retention interval as well as single-presentation control conditions. The results indicate that for moderately long retention intervals, as the lag between repetitions increases, the observed proportion recalled is first less, then eventually exceeds, and then returns to the level expected by an independence hypothesis.     

Modeling: An analysis in terms of category accessibility

It is proposed that many of the releasing or disinhibiting effects caused by models can be accounted for by information-processing mechanisms, without recourse to the concept of reinforcement. More specifically, it is suggested that the process of viewing a model's behavior involves the activation of an interpretive schema. This renders the information the schema incorporates more accessible for subsequent use. If the schema incorporates (or is closely related to) behavior-specifying information, that information becomes more accessible as well, thus making it more likely to influence overt behavior. Two studies are reported that assessed the plausibility of this reasoning. In Experiment 1 subjects who had viewed an aggressive model perceived greater hostility in the behavior of an ambiguous stimulus person that they subsequently encountered than did control subjects. This finding is consistent with the assumption that the model's behavior activated a conceptual schema for use in interpretation. In Experiment 2 subjects in whom an aggressive schema had been activated under a guise displayed greater aggression in their subsequent behavior in a different context than did control subjects. This finding is consistent with the assumption that activating the conceptual schema also activated behavioral information. The discussion centers on the implications of the findings.     

An examination of the seductive details effect in terms of working memory capacity

Previous work on learning from text has demonstrated that although illustrated text can enhance comprehension, illustrations can also sometimes lead to poor learning outcomes when they are not relevant to understanding the text This phenomenon is known as the seductive details effect. The first experiment was designed to test whether the ability to control one's attention, as measured by working memory span tasks, would influence the processing of a scientific text that contained seductive (irrelevant) images, conceptually relevant images, or no illustrations. Understanding was evaluated using both an essay response and an inference verification task. Results indicated that low working memory capacity readers are especially vulnerable to the seductive details effect. In the second experiment, this issue was explored further, using eye-tracking methodology to evaluate the reading patterns of individuals who differed in working memory capacity as they read the same seductively illustrated scientific text Results indicated that low working memory individuals attend to seductive illustrations more often than not and, also, for a longer duration than do those individuals high in working memory capacity.