Introduction
Prosody is a characteristic part of speech that includes intonation, tone destribution and rhythm. It carries the emotional aspect of the speech in non verbal way: among other things, it plays a fundamental role in the early months of life for language acquisition and, afterwards, it supports the memorization from the listener. Several studies, carried out before on cerebral damaged patients for cerebro-vascular incidents and more recently, on normal subiects thanks to modern neuro-imaging technics, have proved, not without disagreement, that the areas responsible for prosody would be located in the right cerebral emisphere, so that the frontal, temporal an parietal lobes and the base ganglia become of great importance. A considerable number of studies have investigated the causes of the loss of prosody in it various forms, productive aprosody (incapability of adapting the emotional tone to the speech contents) and sensory one (incapability of understanding the emotional contents of a listened speech).
The former seemed to be due to bilateral lesions or prevalently concerning the left emisphere, the latter was more frequent in patients with right make-up lesions. More recently however it has been proved that the productive aprosody in patients with left emisphere lesions would be secondary to peculiar dysartria of the same people, in further confirmation of a localization of the prosodical function in the cerebral right sections. Also in the case of Alzheimer’s disease prosody looks seriously compromised, much more than the multi-infarct dementia, where aprosody is directly correlated to the deterioration, once again, of the right sections.
Psychiatric making n°4 – October-December 1997
In spite of the constant interest in the alterations of prosody in neurology damaged patients, the literature analysis shows that this aspect has been insufficiently studied in patients with psychiatric pathologies.
Baltax et al. in 1995 examined the verbal production of 47 schizophrenic patients finding out prosody was one of the most altered characteristics both in young and in adult patients. Mcguire et al, in a very recent research, on the contrary, suggest that a reduced activation of the areas appointed to prosody goes with a less predisposition to auditory hallucinations. In our research we intend to examine, searching for obiective parameters thanks to a computered exploitation, the verbal production of schizophrenic and depressed patients comparing it to a controlling group one.
The sample
18 subjects form our specimen, equitably divided into 6 schizophrenic patients (41.5 middle ages) 6 depressed (45.5 middle ages) and 6 controls (28.7 middle ages).The fifth class was the minimum school attendance for the insertion in the specimen.
Materials and Methods
Each subject has been asked to read aloud, after doing it once silently in order to catch the emotional contents, a short passage of the novel “Il piccolo Principe” by A. Saint-Exupery: “It is very simple: you can’t realize it but with your hearts”. A performance like that was registered by digital recording system D.A.T. (Digital Audio Tape) which allows the elimination of the tape hisses generally present in the common analogical recording and therefore a wider precision in the spectrum analysis, which has been carried thanks to applicated software: (Cubase VST 24 recorder). After getting thanks to the last one the performance graphic elaboration of each subject, we went on analysing the following parameters: mean intensity, range of intensity, time of the sentence, interval time (silences) in absolute value and in percentage as to the time of the sentence, peark mean number and peak number per second.
Results
As already mentioned, the verbal production intensity has been, the first of the parameters considered. The average in the three groups, measured in S.U. (Sound Unity – a parameter of intensity used by the researchers) is reproduced in table 1 and in graph 1.
<
Graph 1
It appears higher in the controls than patients, even if not significant, probably also because of a certain degree of inhibition of our patients for being compelled to speak in front of a microphone, sophisticated pieces of electronic equipment, in the presence of the sound engineer.
Table 1 – S.U. intensity
The intensity mean range, as it can be deduced from the previous diagram, does not differ but a few S.U. among the three groups, and, examining the single cases, you can notice that in all three groups there is a considerable heterogeneity of the diagrams: some subjects get a very large range with peaks up to 80 S.U. ; others get a very low production, nearly whispered.
We have then passed to analyse the total time spent to recite the sentence, that, we remind, is not dependent on difficulty of reading or understanding since each subject has already looked over the text before reciting it aloud.
The result points out how, on average, the depressed patients take a double time (5.9 sec. ) in comparison with the normal subjects (3 sec. ). The value of the schizophrenic subiects is placed in the middle (4.2 sec. ) – (Graph 2).
Graph 2
A very remarkable observation, however, comes when the silence length and their percentage of the total time spent are analysed. In fact the percentage is the same in the depressed and schizophrenic subjects, more than triple compared with the control group. So a mentally ill person delays in the pauses, he takes more time between words and syllables, he looks like as if he should strive to put together the verbal components of the sentence in comparison with a normal subject. (Graph 2 – Table 2).
Table 2 – Silences
In the end we thought of great interest the examination of the average number of intensity peaks noticeable in the three groups: depressed: 22.8; schizophrenics 20.2; controls 15.3. Therefore normal subjects are inclined to speak with a more constant intensity, keeping it for more length and so producing fewer oscillations, the rapid succession of which is noticed already at the first observation of the diagrams of the patients.
This reveals a scarce control of the verbal production and certainly the constant variation of intensity, auditively similar to syllabify, corresponds to a impoverishment of the comparable prosody whether in schizophrenic subjects or in depressed (Graph 3).
<
Graph 3
Relating the number of the peaks to the total length of time, in the end, it is pointed out how the depressed patients produce fewer peaks per second (3.4), than schizophrenics (5.5) and controls (5.1), so obiectifying the sensation of the slow and “watered” speaking of a depressed patient, who even though produces the same number of oscillations as the schizophrenic, delays distributing them in an almost double time (Graph 4 e 5).
<
Graph 4
<
Graph 5
So these observations, even if they are related to a not too large sample, confirm the only literature datum above mentioned concerning the prosodic impoverishment of schizophrenic patients, extending this statement, even if with the differences just pointed out, also to the depressed patients.
Anyway they are some first observations in a not much “frequented” field that our Institute intends to study in depth in the future.
Summary
The authors examine the spectrographic analysis of the voice of depressed and schizophrenic patients recorder with DAT (Digital Audio Tape) during the reading aloud of a short sentence, and compare it with a control group underlining the alterations of prosody in the patients.
voice 1 – voice 2 – voice 3 – voice 4 – voice 5
