The relationship between delusions and hallucinations

Brendan A. Maher DrPhil The origins of delusional beliefs have been a question of great interest for centuries. For many years, the widely held view was that there was a fundamental defect in reasoning in the patient that gave rise to the false belief. Much effort was directed at establishing the kind of defect and the circumstances that triggered it. The search for the basic reasoning defect has met with limited success, mainly because the flaws found in the reasoning of deluded persons are also found with substantial frequency in the general population. The co-occurrence of hallucinations and delusions is consistent with findings that repeated anomalous experiences of various kinds are followed by the development of delusions. In this case, it would be reasonable to regard the hallucinations as exercising a causative role in the development of delusions. - Introduction The co-occurrence of hallucinations and delusions has long been noted. This article examines the nature of the possible relationships between them. Several relationships are possible. Hallucinations may play a primary causative role in the genesis of the delusion; they may be a correlate of the delusion, with both rising from some third cause; or the causative factors may be heterogeneous, so that no one hypothesis about etiology will have universal application. Before turning to this question, we need to consider the definitions of delusion and hallucination. Delusion is defined in DSM-IV (page 765) as a false belief based on incorrect inference about external reality that is firmly sustained despite what almost everyone else believes and despite what constitutes incontrovertible and obvious proof or evidence to the contrary [1]. Hallucination is defined on page 767 as a sensory perception that has the compelling sense of reality of a true perception but that occurs without the external stimulation of the relevant sensory organ [1]. Important aspects of these definitions need to be considered. The definition of delusion implies a hypothesis about the pathogenesis of delusions. The delusion arises, it is suggested, from an incorrect inferencethat is, pathologic malfunction of the reasoning process. This hypothesis requires that, on tests of various kinds of reasoning, we can demonstrate that the performance of deluded persons will contain more errors than the performance of healthy controls on similar tests. For later discussion, we will term this the Basic Reasoning Defect Hypothesis. The definition also implies that the diagnosis of delusion requires that the delusional belief be demonstrably false and that the patient still maintains the belief when presented with disconfirming evidence. However, clinical experience shows that many delusions are not readily falsifiable. There is often no practical way in which unambiguous disconfirming evidence can be obtained and presented. In many cases, falsification would require the clinician to establish the truth of a negative. The belief, for example, that one is being spied upon by the disguised agents of a powerful organization cannot be refuted, given that the agents are said to be disguised and the agency is said to be acting in secret. In actual practice, a belief is likely to be judged to be delusional on the basis of estimated improbability. (This circumstance explains why occasional individuals are thought to have delusions when they correctly hold beliefs that are improbable but true.) Finally, it should be noted that delusions have been reported to be associated with more than 70 different psychiatric and general medical conditions [2,3]. A possible implication of the wide range of conditions in which delusions occur is that the origins of delusions are heterogeneous. If this is so, we might expect to find that various hypotheses about the origin explain some, but not all, cases. Two major hypotheses about the origins of delusions have been emphasized in research studies. The first of these is that there is a basic defect in the reasoning of the deluded individual and that without this defect no delusion would have been created. The second hypothesis is that the delusion arises in an attempt to explain anomalous conscious experiences. The Basic Reasoning Defect Hypothesis The belief that the deluded individual suffers from a primary disorder of reasoning dates back to the work of Kraepelin [4] in the late 19th century. He asserted that delusions are a secondary consequence of primary disorders of reasoning. The primary disorder of reasoning was subject to influence by intense emotional factors. Thus, the causative chain ran from disturbed emotion to a disturbance of reasoning to the emergence of the delusion. As we have seen above, this assumption is reflected in the DSM-IV definition of a delusion. The Kraepelinian formulation makes no mention of the possible role of hallucinations or other anomalous experiences, leaving open the possibility that they, too, might be a consequence of emotional disturbance but certainly assigning them no role in the causal chain. The hypothesis of a basic reasoning defect has been tested many times in many ways. One general strategy has been to present tests of formal syllogistic reasoning to deluded and nondeluded individuals with the expectation that the latter would handle the deductive consequences of the syllogistic premises with greater accuracy. The possibility that the selection of an invalid conclusion arises from a perceived ambiguity in the premises or from a failure of inference from acceptable premises is one source of possible variation in the interpretations that are made from these kinds of data. There have been many studies of this question. Most of the recent work has been summarized by MujicaParodi et al. [5]. Many of the studies suffer from serious methodologic weaknesses, including small samples (sometimes no more than two or three patients) and the use of patients with schizophrenia and other participants regardless of whether they were currently expressing delusional beliefs. Most important, it was frequently found that that nonpatient control subjects made the same kinds of reasoning errors as did the patients. Syllogistic reasoning tests typically provide a sequence of premises followed by a conclusion. The participant is required to decide on the logical validity of the conclusion, given the preceding premises. Conclusions combine two features: They may be valid or invalid and they may be believable or unbelievable [6]. In short, successful performance in the test requires that the participants ignore what they know to be empirically true and concentrate only on whatever follows from the premises provided. This situation is, of course, quite remote from the conditions under which everyday reasoning occurs and may well have little relationship to it. Not surprisingly, conclusions that were invalid but empirically believable were more likely to be endorsed than ones that were logi (...truncated)