Article Text

Loss of silent reading in frontotemporal dementia: unmasking the inner speech
  1. Neurological Department, Grenoble University Hospital, 38043 Grenoble cedex 9, France
  1. Mr L Vercueil vercueil{at}

Statistics from

Presenting signs of frontotemporal dementia usually include social disinhibition, loss of initiative, compulsive features, cognitive decline, and motor symptoms. Behavioural features have recently been detailed, in an attempt to distinguish patients with frontotemporal dementia from patients with Alzheimer's disease.1 2 We report a new feature inaugurating this syndrome which could be a strong illustration of a prefrontal inhibitory role in the control of inner speech, through an early loss of silent reading as an inaugural sign of frontotemporal dementia.

A 69 year old man, a retired baker, with no family history of neurological disease, was admitted to the neurological department for evaluation and diagnosis of a progressive dementia with movement disorders. Since the beginning of 1998 (when he was 67), his wife had noticed that he had slight head movements with small amplitude, occurring only while watching TV. During 1998 the patient had started to read aloud at bedtime, annoying his wife. His voice was very loud, and it was difficult for him to stop. Moreover, he often spoke aloud to himself during the day or suddenly began singing, all features which were very unusual for him. Later in the same year, he showed memory impairment, loss of motivation, and frequent throat clearing sounds. In 1999, a gait disorder with occasional falls appeared, while stereotypic movements resembling temple rubbing with both hands developed. At this time, the patient presented with severe depressive symptoms and became bedridden. At examination, in March 2000, he was able to stop motor and vocal stereotypies for up to 1 minute. After discontinuing the self control, a rebound effect occurred and the patient showed repetitive preauricular rubbing with both hands, with an associated meaningless vocalisation. When asked why he performed these movements, he answered that it was a habit he had developed before. Severe dysarthria and drooling, with swallowing dysfunction were present, but no motor deficit or sensory loss was evidence. Bilateral grasping phenomena, perserveration, and imitative behaviour were easily elicited. Slight symmetric distal akinesia with minimal rigidity was present, but the patient was taking antidopaminergic agents. There were no cerebellar, pyramidal, or dysautonomy signs, or oculomotor palsy. Extensive investigations were negative (including blood sample analysis, genetic testing for Huntington's disease, EEG, brain MRI, and cutaneous biopsy). Neuropsychological tests confirmed the dysexecutive syndrome, with a Mattis score of 122/144 in July 1999, dropping to 87 in March 2000. In August 2000, a percutaneous endoscopic gastrostomy was performed due to swallowing difficulties and a substantial loss of weight. At this time, he continuously repeated his compulsive motor stereotypies and utterances, mainly cursing. Treatment with tetrabenazine, flupentixol, and hydroxyzine were of little help. He died in October 2000 and there was no postmortem analysis.

The patient reported here presented with FTD, according to the current criteria.3 He progressively developed behavioural abnormalities including motor and verbal stereotypies, coprolalia, with poor self control, although he remained able to discontinue them momentarily. He performed very poorly on the frontal battery test and failed to improve under various drug regimens. Brain MRI remained generally atrophic. We suggest that the pathological process underlying his progressive loss of behavioural control was already at work when the patient became unable to read silently. Inner speech is an understudied phenomenon giving rise to mentation under internal control. The major role of “speech for self” in the developing human has been outlined.4 Silent reading could be an externally triggered inner speech as opposed to an internal production of inner speech linked to the thinking process. Besides the generating mechanisms producing inner speech at the interface of thought and language, an endogenous control must exist to inhibit the utterance of inner speech. A pathological process involving this inhibitory control may unmask inner speech. We speculate that such an inhibitory control of inner speech may be pathologically involved in our patient, underlying both initial loss of silent reading, and the most recent vocal compulsive stereotypies. Prefrontal cortical dysfunction may be the underlying pathophysiological mechanism explaining the inability to inhibit the vocalisation of inner speech in our patient, as assumed by the final clinical diagnosis.


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.