For practical reasons, many of the studies that followed Schachter did not use epinephrine. However, at least two studies did use epinephrine, and these produced results that raised some questions about Schachter's theory.
In these studies (Marshall & Zimbardo, 1979; Rogers & Deckner, 1975), subjects were injected with adrenalin and placed in conditions that should have aroused euphoria or at least happiness. Instead, participants reported increased negative emotions. In company with similar effects with a different, hypnotic manipulation of arousal (Maslach, 1979), these results led Marshall and Zimbardo, as well as Maslach, to propose that arousal had a "negative bias" and was experienced as unpleasant, even when the situation and its attendant cognitions implied a happy experience.
Other studies manipulated arousal by other kinds of drugs, such as ephedrine or caffeine, that have less powerful effects than epinephrine. The results have been mixed at best (Manstead & Wagner, 1981; Reisenzein, 1983, 1994). These studies vary on a multitude of dimensions, so any conclusion about the differences between the studies that do and do not support Schachter's theory must be tentative. However, it does appear that the studies that have reported increases in emotional feelings after arousal manipulations have been those in which the target feelings were anxiety, fear, anger, or romantic love, whereas other feelings, including happiness and sadness, were unaffected by arousal (Laird & Bresler, 1990).
An additional complication in these studies is that the meaning of drug effects is often unclear. All these drugs were used in Schachterian experiments because they produce peripheral changes that mimic autonomic arousal, but they may also have effects in the central nervous system. For example, there is little dispute that caffeine in substantial doses causes peripheral arousal symptoms, such as increased heart rate, sweating, and flushing, and it also produces anxiety feelings (Newman, Stein, Tretlau, Coppola, & Uhde, 1992). The question is whether the anxiety is mediated by the peripheral symptoms or whether the two are produced in parallel, with the caffeine causing the anxiety directly by its effects in the brain. (In the case of epinephrine, this explanation is less probable, because peripheral epineph-rine is not thought to cross the blood-brain barrier, and epinephrine is metabolized very rapidly, so that the effects probably all occur peripherally.)
Another manipulation produces complex physiological effects and elicits changes in feelings that may be due to autonomic arousal. In this technique, called the hypoglycemic clamp, blood glucose is experimentally lowered by a continuous injection of insulin into the veins of normal volunteers. The resulting hypoglycemic state produces increases both in autonomic arousal and in unpleasant emotional feelings belonging to Thayer's (1996) tense arousal factor (Gold, MacLeod, Frier, & Deary, 1995; Hepburn, Deary, Munoz, & Frier, 1995; McCrimmon, Frier, & Deary, 1999). Since the effects of the hypoglycemic state are complex, the mood effects may well be produced by some other mechanism than self-perception. Nonetheless, in the absence of an articulated alternative, we may take these effects as some support for Schachter's model.
In sum, the studies in which arousal was manipulated with drugs are best described as "not inconsistent" with Schachter's theory or the predictions of James and self-perception theory. Some studies support the theories, but others do not. These studies also suggest that, at a minimum, Schachter's original view must be qualified: Increases in arousal intensify only some feelings, such as anger or fear, and not all feelings, as Schachter had proposed.
Was this article helpful?