The exact cause of auditory hallucinations remains unknown but researchers have come up with several competing theories to explain why they occur.
One popular
theory about the causes of auditory hallucinations in schizophrenia suggests
that they are a type of inner speech that is mistakenly attributed to an
external source. A recent discovery that blood flow increases in the Broca’s
area (the language production center of the brain that is also activated during
inner speech) during verbal hallucinations supports this idea. Mc Guire, Shah,
and Murray found that “a predisposition to verbal hallucinations might reflect
aberrant connectivity between the areas concerned with the generation and
monitoring of inner speech.” Thus, failure to monitor inner speech
is, in a sense, a failure of communication between the “mind’s voice” and
the “mind’s ear.” There are currently two theories that attempt to explain
why this miscommunication occurs, the Subvocalization
Theory and the Phonological Loop Hypothesis.
Subvocalization is the process of whispering, or mumbling, or other inaudible engagement of the vocal apparatus that often occur unconsciously. Although the majority of Schizophrenic patients do not vocalize the voices that are speaking in their minds, the majority of them do subvocalize, and these movements can be detected (Frith and Fletcher) In 1966 McGuigan found that the chin electropmyogram measurements (EMG) as well as the breathing of the subject increased two seconds before subjects reported having a hallucination. He also noticed that he could occasionally discern the content of the subvocalizations and that it was semantically related to the content of the auditory hallucination. Inouye and Shimizu (1966) found that hallucinating patients demonstrated significantly more subvocal activity when hallucinating than they did when they were not, a finding later confirmed by Bick and Kinsbourne (cited in Bick and Kinsbourne)
Researchers Bick and Kinsbourne wanted to determine whether the patient was shadowing the hallucinated voice or generating it during subvocalization. They discovered that by obstructing subvocalization by otherwise occupying the vocal apparatus (such as having the subject open his or her mouth) actually suppressed the auditory hallucination. They reasoned that “auditory hallucinations may be the projection of schizophrenic patients’ thoughts, subvocalized due to deficient cerebral cortical inhibition." They concluded that the voices come from subvocalization, which the patient hears and then attributes to an external source. The voices are not shadowed but generated first by the patient and then projected to a source outside of their minds. This process of projection can also explain the grammatical and tonal differences perceived in the verbal hallucinations.
The phonological loop hypothesis is another theory (and a much debated one at that) that attempted to explain the problems schizophrenic patients who experience auditory hallucinations have in monitoring inner speech. Haddock et al. explain in their paper “Functioning of the Phonological Loop in Auditory Hallucinations” that deficits in the phonological loop contribute to auditory hallucinations. The locus of inner speech consists of two distinct sub-components of the working memory system. One is a “phonological input store capable of representing speech for a brief period” of one or two seconds. The other is an articulatory loop by means of which information in the phonological store can be refreshed before it fades." It is believed that while information is held in store, information that is processed auditorily has automatic access to the store. Haddock et al. (1996) note that if auditory hallucinations are indeed misattributed inner speech acts, then it is possible that deficits in the phonological store and loop contribute to the patient’s misattribution of the source of the hallucination. The researchers proposed that “perhaps deficits in these mechanisms lead to degraded auditory representations so that the hallucinators are denied the phenomenological cues which allow normal individuals to identify inner speech as self-generated."
The results, however, did not support this possibility. Haddock and colleagues found that although the psychiatric patients performed more poorly overall than the normal subjects, there was no evidence of abnormal functioning of the phonological loop. These findings led them to conclude that auditory hallucinations are not the result of abnormalities in the phonological loop process. Instead, “the hallucinating patient’s misattribution of inner speech to a source which is alien to the self is likely to be a product of abnormalities in cognitive symptoms other than the phonological loop itself."
But that was not the end of the story. Directly contradicting the conclusion of Haddock et al., a study conducted by Hoffman (1999) based on a narrative speech perception task supports the idea that “hallucinated voices in schizophrenia arise from disrupted speech perception and verbal working memory systems rather than from non-language cognitive or attention deficits." Two other researchers, Aleman and De Hann took issue with these findings and contended that although disrupted speech perception may play a part, this does not therefore eliminate the possibility that cognitive variables such as reality monitoring also contribute to the auditory hallucinations. Hoffman (2000) retorted that "the reality monitoring theory would lead one to believe that hallucinations should mirror the nature and content of the imagery of ordinary thought." But, he argued, hallucinatory experiences are attributed to an external source because they are so different in quality and content from a person’s ordinary inner thoughts.
In a study that seemed to open up the phonological loop hypothesis to even more debate, Evans, McGuire, and David observed that clinical intervention techniques that seem to interfere with inner speech demonstrate that the hallucinations were “competing” for a common cognitive resource, the phonological loop. Yet, in their recent study seeking to clarify this issue, they found that on tasks that relied on “inner voice/inner ear partnership for successful performance,” Schizophrenic patients who experienced auditory hallucinations performed as well as the non-hallucinating ones.
Many Schizophrenic patients who experience auditory hallucinations find it difficult to say whether the “voice” they experience is subjectively located inside or outside their head (Evans et al. (2000). The self monitoring hypothesis attempts to account for the misattribution of inner speech with the claim that auditory hallucinations arise from a “failure of internal feedback mechanisms that distinguish self generated from externally stimulated neural activity” (Feinberg). For people with normal self-monitoring abilities, we know that we are about to speak, and our ears expect to hear exactly what we end up saying. However, in patients with low self-monitoring, “it is the expected sounds of their own voices, rather than the awareness of their intention to speak, that enable them to know who is speaking (Frith and Fletcher). In addition, Bental, Baker, and Havers found that hallucinators are more likely to attribute events to an external source when they are uncertain as to the source (internal or external) of a perceived event.
Heilbrun hypothesized that “given the need to discriminate between one’s own lexical thought and a voice form another source, the person less familiar with the properties of his or her thinking would be more likely to mislabel the source." Hallucinators, he found, were less able to identify lexical, semantic, and syntactic properties of their own thoughts, findings that support the theory of low self-monitoring in hallucinating Schizophrenics.
In another experiment testing this theory, Frith and Fletcher distorted the subject’s voices and fed it back to them simultaneously through headphones. Non-hallucinators were aware that the feedback was their own distorted voice, but hallucinators perceived the distorted voice to be another person speaking while they spoke (“I think its an evil spirit speaking when I speak.")
Other evidence to support this theory derives from evidence from Haddock,
Bentall, and Slate (cited in Davies) that the cognitive-behavioral technique
of focusing helps re-attribute the voices to the self rather than to external
sources. The significance of this finding being that when people are taught
to self-monitor they are more aware of their own speech.