Three studies examined mental model generation after preexposure to uncontrollability and in a depressive state. The purpose of the experiments was to test the implications of the cognitive exhaustion model, applying an explicit conceptualization of social mental models and a process-tracing method developed by U. von Hecker (1997). An experimental situation was created for observation of consecutive, rule-based construction steps as a function of input diagnosticity, and for the quality assessment of constructed mental models. The findings show that participants preexposed to uncontrollability, as well as depressed students, were able, as were controls, to identify rule-relevant information needed for model construction. However, they were less able than control participants to engage in a more cognitively demanding and generative step of processing (i.e., in integrating the pieces of input information into a coherent mental model of sentiment relations).

The performance of older adults and depressed people on linear order reasoning is hypothesized to be best explained by different theoretical models. Whereas depressed younger adults are found to be impaired in generative inference making, older adults are well capable of making such inferences but exhibit problems with working memory (Experiments 1 and 2). Restriction of the available study time impairs reasoning by nondepressed control participants and. as such, proves to be a good model of older adults' but not depressed participants' limitations (Experiment 3). These results are replicated comparing depressed and older participants with a control group in the same study, providing increased power and linking the results to additional control measures of processing speed and working memory (Experiment 4).

No abstract

The authors suggest that depressed mood is associated with a defocused mode of attention, allowing irrelevant information to be noticed and processed more than in nondepressed states. Working on a source monitoring task, subclinically depressed college students selected with the Beck Depression Inventory (A. T. Beck, 1967; D. Kammer, 1983) had better memory for irrelevant stimulus aspects than nondepressed control students. However, depressed students' performance on the relevant stimulus aspects was unimpaired. These results are in conflict with a capacity reduction view of depressed mood and support the hypothesized altered, defocused mode, in which attentional resources are more evenly allocated across various aspects of the materials. The results are discussed within the framework of adaptive functions of emotional states.

Memory performance in linear order reasoning tasks (A > B, B > C, C > D, etc.) shows quicker, and more accurate responses to queries on wider (AD) than narrower (AB) pairs on a hypothetical linear mental model (A - B - C - D). While indicative of an analogue representation, research so far did not provide positive evidence for spatial processes in the construction of such models. In a series of 7 experiments we report such evidence. Participants respond quicker when the dominant element in a pair is presented on the left (or top) rather than on the right (or bottom). The left-anchoring tendency reverses in a sample with Farsi background (reading/writing from right to left). Alternative explanations and confounds are tested. A theoretical model is proposed that integrates basic assumptions about acquired reading/writing habits as a scaffold for spatial simulation, and primacy/dominance representation within such spatial simulations. (PsycINFO Database Record