Lazarus has challenged the view that there are circumstances under which affect precedes cognition and that affective arousal that does not entail prior cognitive appraisal exists. His argument, however, is based entirely on an arbitrary definition of emotion that requires cognitive appraisal as a necessary precondition. To satisfy this concept of emotion, Lazarus has broadened the definition of cognitive appraisal to include even the most primitive forms of sensory excitation, thus obliterating all distinction between cognition, sensation, and perception. No empirical evidence is offered to document the principle of cognitive appraisal as a necessary precondition for emotional arousal. The contrasting view of an affective primacy and independence, however, is derived from a series of findings and phenomena, including the existence of neuroanatomical structures allowing for independent affective process. Only a few years ago I published a rather speculative article entitled \"Feeling and Thinking\" (Zajonc, 1980). The title also included the provocative subtitle \"Preferences Need No Inferences,\" deliberately suggesting an occasional independence of emotion from cognition. In this article I tried to appeal for a more concentrated study of affective phenomena that have been ignored for decades, and at the same time to ease the heavy reliance on cognitive functions for the explanation of affect. The argument began with the general hypothesis that affect and cognition are separate and partially independent systems and that although they ordinarily function conjointly, affect could be generated without a prior cognitive process. It could, therefore, at times precede cognition in a behavioral chain, I based this proposition on a number of diverse findings and phenomena, none of which alone could clinch the argument, but all of which taken together pointed to a clear possibility of an affective independence and primacy. This idea was first advanced by Wundt (1907) and later reiterated by others (e.g., Izard, 1984). Lazarus (1982) takes a very strong issue with all of this and almost categorically rejects the likelihood of the independence of affect of cognition, let alone the possibility of an affective primacy. In this article I will review Lazarus's position and contrast it with mine. Lazarus employs two definitions, one for emotion and one for cognition. All of his inferences are based on these two definitions. Lazarus's definition of emotion (which requires cognition as a necessary precondition) is central to his position. On the basis of this definition alone, therefore, the argument is unassailable. If Lazarus insists on his definition, as he has the right to do, we must agree that affect cannot be independent of cognition because by definition cognition is a necessary precondition for affective

On the primacy of affect.

A monograph communicating the current realities and future possibilities of unifying basic studies on anatomy and cellular physiology with investigations of the behavioral and physiological events of waking and sleep. Steriade established the Laboratory of Neurophysiology at Laval U., Quebec; McCarl

Brainstem control of wakefulness and sleep

Neurophysiology of the suprachiasmatic circadian pacemaker in rodents.

Axonal transport techniques were used to define the organization of projections from the forebrain to the paraventricular (PVH) and supraoptic (SO) nuclei of the hypothalamus in the adult rat. First, injections of a retro-gradely transported fluorescent tracer, true blue, were placed in the region of the PVH, and the distribution of retrogradely labeled cells was charted. Then, the autoradiographic method was used to confirm the results of the retrograde transport experiments, to describe the route taken by inputs to the PVH and SO, and to determine the distribution of each projection within the various subdivisions of the nuclei that have been recognized previously on the basis of cytoarchitectonics, efferent projections, and immunohisto-chemistry.



The results can be summarized as follows. First, direct projections to the PVH arise from all areas of the hypothalamus (except the SO, the medial and lateral mammillary nuclei, and the magnocellular preoptic nucleus), and from the subfornical organ and the bed nucleus of the stria terminalis. Second, every forebrain region projecting directly to the PVH appears to innervate cell groups in parvocellular parts of the nucleus that in turn project to the neurohemal zone of the median eminence, and to autonomic centers in the brainstem and spinal cord. In addition, each input has a unique distribution within the parvocellular division. Third, relatively few regions in the forebrain project directly to magnocellular neurosecretory parts of the PVH, and are thus in a position to influence directly the release of posterior pituitary hormones. These regions (each of which also projects to the parvocellular division) include the dorsomedial nucleus of the hypothalamus, the median preoptic nucleus, the subfornical organ, and the bed nucleus. Therefore, inputs from the hippocampal formation, amygdala, and lateral septum that have been shown to influence the magnocellular neurosecretory system appear to be relayed by way of short projections from other parts of the hypothalamus (including perhaps parvocellular parts of the PVH) and from the bed nucleus. Fourth, most of these cell groups innervate preferentially either oxytocinergic (dorsomedial and bed nuclei) or vasopressinergic (subfornical organ) parts of the magnocellular division of the PVH. Only the projection from the median preoptic nucleus appears to be distributed uniformly over parts of the magnocellular division in which oxytocin- and vasopressin-con-taining cells are concentrated. And fifth, each region that preferentially innervates either oxytocinergic or vasopressinergic neurons in the PVH also preferentially innervates the same cell type in the SO (which does not contain a distinct parvocellular division). The results indicate that the coordination of neuroendocrine and autonomic responses from the PVH is mediated at least in part by a complex series of neural inputs, each of which innervates more than one functional subdivision within the nucleus.

The organization of forebrain afferents to the paraventricular and supraoptic nuclei of the rat.

The efferent projections of the suprachiasmatic nucleus (SCh) in the rat hypothalamus have been reexamined with the anterograde tracer Phaseolus vulgaris leucoagglutinin (PHA-L), which displays labeled axons with the clarity of a Golgi impregnation. Fibers from the SCh can be divided into six pathways for descriptive purposes. By far the densest terminal field arising from cells in the SCh ends in a roughly comma-shaped zone between the SCh and paraventricular nucleus on the one hand and the periventricular nucleus and anterior hypothalamic area on the other. A few axons continue dorsally from this "subparaventricular zone" to pass through parvicellular parts of the paraventricular nucleus and the overlying midline thalamic nuclei to end in midrostrocaudal parts of the paraventricular nucleus of the thalamus, and a larger number continue caudally to end in the dorsomedial nucleus, dorsal parts of the cell-sparse zone surrounding the ventromedial nucleus, and the posterior hypothalamic area. The other five pathways all consist of relatively small numbers of fibers and give rise to relatively sparse terminal fields. The second pathway consists of rostrally directed fibers that end in ventral parts of the medial preoptic area and anteroventral periventricular nucleus. The third consists of anterodorsally oriented fibers that pass through the medial preoptic nucleus and adjacent regions to end ventrally in the intermediate lateral septal nucleus. The fourth consists of fibers just caudal to the third group that end in the preoptic continuation of the bed nucleus of the stria terminalis, as well as in the parataenial nucleus and rostral part of the paraventricular nucleus of the thalamus. The fifth consists of laterally directed fibers that course over the optic tract to end in the ventral lateral geniculate nucleus. And the sixth consists of fibers that course posteriorally through the anterior hypothalamic and retrochiasmatic areas to end in the cell-sparse zone between the arcuate nucleus and ventral parts of the ventromedial nucleus, as well as in adjacent parts of the lateral hypothalamic area. The distribution of projections labeled following PHA-L injections centered in the subparaventricular zone was also examined and was confirmed with retrograde tracer experiments (Watts and Swanson: J. Comp. Neurol. 258:230-252, '87). The results indicate that the subparaventricular zone projects to essentially the same regions as the SCh, only much more densely, and also sends fibers back to the SCh.(ABSTRACT TRUNCATED AT 400 WORDS)

F.K. Stephan

Papers

Efferent connections of the rat suprachiasmatic nucleus