Knowledge that Transforms

The relative loudnesses of tones that differ in sound frequency can depend strongly on the stimulus context, that is, on the set of intensity levels in the stimulus ensemble Using a new paradigm, called matching in scaling, this investigation sought to confirm that context modifies loudness relations per se, and not, for example, only overt responses To this end, two experiments revealed that changes in stimulus context differentially affect direct comparisons of loudness of 500-Hz and 2,500-Hz tones, as well as numerical judgments of individual tones—when loudness matches and scaling judgments alike are obtained in the same experimental sessions These contingent effects vary dynamically over time as a function of the recent stimulus history A third experiment revealed analogous effects in a simple matching paradigm, with no numerical judgments at all These findings support the contention that basic properties of loudness perception—grounded in auditory processes often considered “low level”—nevertheless can be deeply contextual

This experiment evaluated the relationship between availability of ethanol and preference for ethanol in rats. One dipper served a mixture of 10% ethanol and 10% sucrose, and a second dipper served 10% sucrose. In the first condition, access to each dipper was governed by a variable-interval 5-s schedule. In subsequent conditions, the interval requirement for the ethanol mixture was increased to 30 s. However, increases in the interval requirement did not decrease preference for ethanol. Instead, preference increased somewhat or remained about the same. In contrast, when both dippers served sucrose, increases in the interval requirement at one dipper systematically shifted preference to the other dipper. Analysis showed that preference for ethanol was correlated with amount of ethanol consumed, whereas preference for sucrose was correlated with changes in the reinforcement contingencies. Ethanol's food value, as measured by calories, did not predict preference. Rather, the results were most simply explained by the theory that ethanol's pharmacological consequences maintained responding.

The study of emotions is complicated by a number of factors, including the biological bases of emotional experience and expression. Although biochemical changes have long been thought to be part of the evocation of emotions, research has not consistently included this level of measurement, and theories of emotion have not systematically specified the role of these biochemical bases. In this paper, we briefly review relevant biochemical principles and measures that can be integrated into research and theory on emotions. Applications of these principles in extant studies of emotion are also discussed.

Behavior-genetic studies of emotionality in rats and human studies of adult personality and childhood temperament provide evidence of genetic effects on emotion-relevant trails. Twin studies of autonomic reactivity suggest that psychophysiologic responses to stress are heritable. Genetic analyses of fears reveal that the magnitude of genetic influence varies across fears and across development. Finally, recent findings in psychiatric genetics shed light on the role of heritable variation in emotional functioning in major psychiatric disorders. Thus, there is evidence from a variety of sources that characteristics relevant to emotion are influenced by genetic factors. However, behavior-genetic methods generally have not been applied in laboratory-based studies of the psychophysiologic, cognitive, or expressive correlates of emotion. Because recent developments in quantitative genetics permit us to go beyond the simple estimation of heritability, a broader application of behavior-genetic strategies to the study of emotion, one which incorporates elicitation of emotion in the laboratory, promises to elucidate the interrelationships among the components of emotional systems.

Birnbaum and Sotoodeh ( PS 2(4), 1991, pp. 236–240) used psychophysical techniques to generate severity weightings for the stressful life events from Holmes and Rake's (1967) Social Readjustment Rating Scale. The techniques Birnhaum and Sotoodeh applied are designed to uncover the structure of judgments and decisions, but in the context of predicting physical symptoms, they do not improve on the original weightings published in 1967. Neither the original Holmes and Rahe weights nor Birnbaum and Sotoodeh's weights were significantly better than unit weighting (1 if event occurred, 0 if not) for predicting physical symptoms of 115 college students.

One major reason that scientific information is not often effectively applied in the formation of social policy is that available methods for coping with the uncertainty in scientific judgments are overlooked. Application of such methods is as necessary as reducing uncertainty through the acquisition of more knowledge for these methods result in the more effective use of knowledge already at hand. Furthermore, failure to cope with uncertainty through explicit, analytical methods results in failure to separate fact (truth) from value (justice), which in turn results in wasted knowledge and unnecessary dispute. We present four concrete examples of how such circumstances can be avoided.