Emotional facial expressions in individuals with total laryngectomy

Abstracts

Purpose

: to verify whether there are differences in the degree of intelligibility of individuals with total laryngectomy, using different forms of oral communication (speech whispered, electronic larynx, esophageal voice and tracheoesophagealspeech); verify whether there are differences in facial expression (number and duration of facial movements) in total laryngectomized individuals using different forms of oral communication during spontaneous happy and sad speeches and verify if there are differences in facial expression (number and duration of facial movements) of individuals with high and low intelligibility of spontaneous happy and sad speeches.

Methods

: treated an exploratory, descriptive and comparative basic methodology cross, and constituted a group of thirteen individuals with total laryngectomy. Subjects were videotaped during both spontaneous happy and sad speeches and evaluated for speech intelligibility for people without pathology of communication, and facial expression analysis was performed using the Facial Action Coding System. We used Statistical Package for the Social Sciences 19for descriptive and inferential analysis, and held the Mann-Whitney and Kruskall-Wallis statistical tests.

Results

: we found that different types of communication have varying degrees of intelligibility in speech and descriptively differences in facial expression of laryngectomized subjects during the two different type of speeches used in this study. We also verified statistically significant differences in facial expression in groups of low and high intelligibility, the average duration of each unit of sad eyes in spontaneous speech (p-value=0.01).

Conclusion

: it was found that the two emotional contexts, the laryngectomy, regardless of their communicative way, are less facially expressive than normofalantes.

Emotions; Facial Expression; Laryngectomy; Speech Intelligibility

Objetivos

: verificar se existem diferenças no grau de inteligibilidade de indivíduos laringectomizados totais, que utilizam diferentes formas de comunicação oral (fala murmurada, laringe eletrônica, voz esofágica e voz traqueoesofágica); verificar se existem diferenças na expressão facial (número e duração de movimentos faciais) de indivíduos laringectomizados totais, que utilizam diferentes formas de comunicação oral, durante tarefas de discurso espontâneo alegre e triste e, ainda, verificar se existem diferenças estatisticamente significantes na expressão facial (número e duração de movimentos faciais) de indivíduos com elevada e baixa inteligibilidade durante tarefas de discurso espontâneo alegre e triste.

Métodos

: tratou-se de um estudo exploratório, comparativo de base descritiva e de metodologia transversal, sendo que se constituiu um grupo de treze indivíduos laringectomizados totais. Os indivíduos foram filmados durante as duas tarefas de discurso espontâneo alegre e triste e avaliados quanto à inteligibilidade do discurso por pessoas sem patologia da comunicação, sendo que a análise da expressão facial foi realizada com recurso ao Sistema de Codificação de Ação Facial. A análise descritiva e inferencial foi efetuada com recurso ao SPSS19 por meio da aplicação dos testes Mann-Whitney e Kruskall-Wallis.

Resultados

: verificou-se que os diferentes tipos de comunicação apresentam diferentes graus de inteligibilidade no discurso e que descritivamente existem diferenças na expressão facial dos indivíduos laringectomizados totais durante o discurso espontâneo alegre e triste . Verifica-se ainda diferenças estatisticamente significantesna expressão facial nos grupos de baixa e elevada inteligibilidade, na duração média de cada unidade dos olhos no discurso espontâneo triste (p-valor=0,01).

Conclusão

: verificou-se que nos dois contextos emocionais, os laringectomizados, independentemente da sua forma comunicativa, são menos expressivos facialmente do que os normofalantes.

Emoções; Expressão facial; Laringectomia; Inteligibilidade da fala

INTRODUCTION

Cancer has an impact increasingly emergent in societies once it is one of the most feared diseases and it provokes many strong emotions not only in patient, but also in those who surround him/her. ¹

In Portugal larynx cancer corresponds to twenty-five per cent of the malignant tumours in head and neck region ¹ and it could compromise larynx functions, such as breathing, deglutition and phonation, and, thereafter, communication. It is most frequent in men between fifth and sixth decade of life and its etiology is strongly related to smoking, alcohol, professional exposition to metals, textiles and radiations, and also hereditary factors. ²

The location and staging of the tumor are proxies in delineating the treatment and, in most cases, the option is surgical, total or partial laryngectomy associated or not with radiotherapy and/or with chemotherapy, in pre or post surgery.

After a total laryngectomy, the breathing becomes to be held definitively by tracheal stoma and the production of laryngeal voice becomes impossible ³ .

Some authors relate that a total laryngectomy is emotionally more traumatic than another type of surgery due to psychological and functionalimpairment ⁴ . The initial aim is the rehabilitation of the oral communication of the individual, thus allowing a social, professional and a family reintegration of the laryngectomized person ^3,5 .The return of communication becomes the triad’s priority: laryngectomized, phonoaudiologist and physician. However, the prognosis of vocal rehabilitation varies from person to person according to anatomical and physiological, psychological and social factors. Classically, in Portugal, the vocal reeducation may start by usingremaining anatomical segments (esophageal voice), prosthesis (electronic larynxes) or prosthetic-surgicaldevices (tracheoesophageal fistulas).

The voice plays a fundamental role in communication as it permits the transmission of our ideas, desires and emotions. On the other hand, it is known that the voices acquire an affective value and to listen to them is enough to influence our perception and interaction with people ⁶ . Thus, intelligibility is seen as the capacity to use phonetic units of speech, so that the speaker can be well understood by the listener in different situations of communication ² .

The human face is a rich source of information where one can infer about the different characteristics of the individuals, such as, age or sex. The emotional expression, being the most important channel, plays an important role in social communication ⁷ .Emotions can, therefore, be linked either tovocal acoustic parameters orto facial expressions. The facial expressions of emotion are modeled by the movements of the muscles of the face that correspond internally to affective states ⁸ and it is the most basic and common form to express emotions. They are also the most rich and important way in interpersonal relations to express the states of mind, the emotions, the desire to communicate and the level of expressiveness during communication, throughthe changes in the configuration of the eyes, mouth and the position of eyebrows.

Despite some aspects of expressions are culturally determined there are basic expressions that are universally recognized, such as happiness, sadness, aversion, surprise, fear, and anger ⁹ .

Several studies have been carried out in the context of facial expressions in some pathologies, in its great majority in neurological and psychiatric population. Patients with depression showed incapacity to accurately identify several emotional expressions, which lead to changes in interpersonal relations ¹⁰ . In addition to the previously mentioned, they tended to judge neutral emotions (as surprise) as being more negative and they are less facially expressive than healthy individuals. In this context, the studies also indicate a decline in the competence of facial expression of emotions as they become older, as well as a decrease of the intensity of these same emotions ¹² . In short, do not be able to transmit a nonverbal message adequate to the verbal content or to the emotion desired can be of great concern and a source of bad understanding ¹³ . Towards the previously exposed, it will be expected that the laryngectomized person presents not only a difficulty in expressing ideas and concepts, but also in expressing emotions.

After the bibliographic review, it was found that there is a lack of studies on the capacity of transmitting emotions in the population of laryngectomized patients. In Portugal, as far as we could ascertain up to the present, there is only one study on the spontaneous emotional facial expression with the same methodological approach, referring to the vocal pathology ¹⁴ . 

The present study aimed: (1) to describe the emotion perceived and its intensity in spontaneous speech emotionally positive and negative of normospeakers and laryngectomized people; (2) to describe the level of intelligibility in several alternatives to laryngeal voice in the group of laryngectomized; (3) to check if there are statistically significant differences in the level of intelligibility of several alternatives of communication from laryngectomized person; (4) to describe the facial expression (the number and duration of movements) in neutral, happy and sad contexts of normospeakers and laryngectomized during the task of automatic language; (5) to describe the facial expression (the number and duration of movements) in normospeakers and laryngectomized during happy and sad spontaneous speeches ; (6) to verify the existence of differences statistically significant in facial expression of happiness and sadness during the happy and sad spontaneous speech in normospeakers and laryngectomized; (7) to describe the transmission of happiness and sadness (the number and duration of movements) in normospeakers and laryngectomized during happy and sad spontaneous speech; (8) to describe and verify the existence of statistically significant differences between the facial expression of happiness and sadness in laryngectomized patients of low and high intelligibility during  happy and  sad spontaneous speech.

METHODS

It was an exploratory, comparative study, with a descriptive basis and a cross-sectional methodology. The research sample was comprised of all laryngectomized people available in the Otorhinolaryngology Service of the Hospital Garcia de Orta, between February and July of 2011, who met the inclusion criteria: male, total laryngectomy for at least six months, without depressive symptoms, without changes of the orofacialmobility and without neurologic pathology. The present study was approved by the Committee for Ethics in research with human beings of the Hospital Garcia de Orta, with the number 21922. For comparative purposes, it was constituted a homogenous group of normospeakers collected in our circle of knowledge, fairly balanced as to age.

The sample was, then, composed of thirteen total laryngectomized men (two with electronic larynx, three with whispered speech, four with acquired esophageal voice through the method of injection of air, or Dutch method, and four with tracheoesophageal voice), with an average age of 64.31±10.55 years old and thirteen normospeakers men with an average age of 62.08±8.89 years old (Table 1). Although the group of normospeakersis younger, both groups did not show significant differences regarding the dependent variable age (p= 0.56). The classification of the level of intelligibility, performed by thirty people, of both sexes, collected within our circle of knowledge, without contact or hearing training prior to the pathology (laypeople), has determined the existence of eight laryngectomized people with low intelligibility with an average of five years of surgery and five with high speech intelligibility with an average of three years of surgery (Table 2).

Table1
–Sociodemographic characteristics of the sample groups

		Normospeakers N=13	Laryngectomized N=13
Age	Mean±SD [Min-Max]	62,08±8,89 [51-83]	64,31±10,55 [51-86]
Qualifications	Without qualifications	0%	7,7%
	Firstcycle	7,7%	7,7%
	Secondcycle	38,5%	53,8%
	Thirdcycle	23,1%	23,1%
	Secondary	7,7%	0%
	Graduation	23,1%	0%
	Master'sdegree	0%	7,7%
Civil status	Married	100%	84,6%
	Single	0%	0%
	Widower	0%	7,7%
	Divorced	0%	7,7%
Professional situation	Worker	46,2%	15,4%
	Retired/Unemployed	53,8%	84,6%
Household	Alone	0%	15,4%
	Wife	46,2%	38,5%
	Wifeandchildren	53,8%	30,8%
	Wife, childrenandgrandson	0%	7,7%
	Mother	0%	7,7%

SD – standarddeviation; Min – Minimum; Max – Maximum

Table 2
– Characterization of group laryngectomized in low and high intelligibility

		Low intelligibility	High intelligibility
N		8	5
Age	Mean±SD	67,88±11,84	58,60±4,72
	[Min-Máx]	[56-86]	[51-62]
Time surgery	Mean±SD	5,38±6,55	2,80±0,84
	[Mín-Máx]	[1-19]	[2-4]

SD – standard deviation; Min – Minimum; Max – Maximum

The instruments used for data recording were the sociodemographic characteristics record sheet of the sample and the qualifying and quantifying emotional record sheet. On the emotional qualification record sheet the six basic emotions universally recognized are present; this sheet was drawn up with the aim of understanding whether the emotion examined by the authors is in agreement with the emotion perceived by the individual. Nevertheless, the record of emotional quantification relating intensity (emotional quantification) was carried out by a visual analogue scale (VAS). This scale is graphically represented by a vertical line of 100mm, which contains, in the extremes, zero and ten, which correspond to the emotion “Null” and “Extreme”, respectively. In this vertical line the individual was instructed to register the amount of emotion experienced at the time.

Procedures

All participants were informed about the study and authorized its participation by informed and enlightened consent. After the ethical procedures, the screening of depressive states through the Portuguese version of the Scale of Depressive States (CDS-D) ¹⁵ was carried out. All individuals with answers above 16 values were eliminated. 

Subsequently, the record of the connectors of sociodemographic characterization of the sample was carried out and also video and audio recording in two tasks: automatic language (neutral emotional context) and happy and  sad spontaneous speech. Image collection was performed by Sony HDdigital camera, throughout a fixed distance of one meter. Video and audio collection was accomplished inthe phonoaudiologist cabinet, in calm and quiet environment.

The induction of emotion happy was given by verbal stimulus “Imagine that I would say to you that you were the unique Euromillions winner. Please, describe me your next week” and the emotion sad through the stimulus “Imagine that I would say to you that your cancer has returned. Please, describe me your next week”, for the laryngectomized group and “Imagine that I would tellyou that you had a larynx cancer. Please, describe me your next week” for normopeakers group. The choice of the positive stimulus was performed based on its usage in a previous study ¹⁴ with dysphonic population and which proved to be inducer of happiness in perceiving of the studied population. The choice of the inducing stimulus of sadness was performed because it was identified by laryngectomized population as the health condition with greater impact on their quality of life ^4,13 and also because the cancer is one of clinical situations most feared by the population. The order of emotions requestedwas at random, and it was requested before and after the task of spontaneous speech to individuals who appoint the months of the year (automatic language). After each spontaneous speechit was also asked to individuals to qualify and quantify the emotion felt. 

In the videos analysis, toprevent contamination of the facial analysis, video files were visualized at random order and without sound. The analysis of the videos was made by Anvil 5.0 ¹⁶ program, and the annotations were about movements of minimum units of movement, belonging to the Coding System of Facial Action (FACS) ¹⁷ . It was also decided that, with basis on the Portuguese study, referring the vocalpathology ¹⁴ , the units of action concerning happiness were,for the eyes,the A6, and, for the mouth, the A12, A13 and A14; and for the  sadness, for the eyes, the A7, A41, A42, A43 and A44, and,for the mouth, the A10, A11, A15, A16 and A20. 

Finally, in order to check the type of communication with better level of intelligibility, it was asked to thirty individuals (men and women), collected in our circle of knowledge, without connection or hearing training prior to the pathology (laypeople), with ages  between twenty-seven and eighty-three years, to evaluate the level of intelligibility of the communication of the laryngectomized by audio and video visualization (in order to approachas much as possible to the context of social interaction) of the first 30 seconds of happy and sad spontaneous speechand it was requested them to refer what they understood from the recording, and to quantify from zero to ten the level of understanding, as zero corresponded to “nothing” and 10 to “everything”. For the images viewing it was used a TOSHIBA  Satellite  A200-12X computer and for images hearing it was used Apple Headphones. The answers of individuals were recorded witha Olympus  digital VN-7800PCDictaphone. After this quantification of intelligibility, one proceeded to a cut-off point of five to groupthe individuals by low and high intelligibility. 

Data were processed using SPSS 19, using Shapiro-Wilk test to check the existence of normality, and Mann-Whitney test and Kruskal-Wallis test for the comparison between independent groups. It was used for inferential analysis the level of significance of α= 0.05.

RESULTS

The results will be presented in the order of proposed objectives.

Quantification and emotional qualification

We can see in Table 3 that, in both groups, the emotional intensity perceived is positive. Laryngectomized with electronic larynx were those who felt happiness (8.75±1.77) and sadness (7.50±3.54) with greater intensity in relation to laryngectomized with esophageal voice (7,50±2,74) that are also the group with the greatest intensity of sadness. Both normospeakers and laryngectomized realized the positive emotions in a more intense way than the negative. This finding is justified in studies carried out in this area ¹⁸ , due to the fact that happiness is a marginal emotion because it is the only one that involves an excitability/generalized activation before the tendency for action, while all the other emotions are defined in terms of readiness to take specific actions. These findings do not meet the resultsby the group with electronic larynxonce they quantified the excitement in  happy spontaneous speech with high values, but they were not the most expressive in all facial segments. On the other hand, the group with electronic larynx and esophageal voice quantified the emotion, in sad spontaneous speech, with a high intensity, compared to other groups. However, these groups weren’talso the ones who presented more movements associated with the expression of sadness, which goes against the findings of another study ¹⁸ , in which the author mentions that the relation between the duration and the intensity felt in general is low or even negative and the relationship is caused by external stimulus that also influences the emotional duration.

Table 3
– Emotiona lquantification for different types of communication in the evocation

Emotional quantification	Types of communication
	Normospeakers	WS	EL	EV	TV
	Mean±SD	Mean±SD	Mean±SD	Mean±SD	Mean±SD
Happiness	7,59±2,70	6,33±3,18	8,75±1,77	7,13±1,89	6,60±4,32
Sadness	6,72±2,81	6,00±1,80	7,50±3,54	7,50±2,74	6,33±2,45

SD – standard deviation; WS – Whispered Speech EL – Electronic Larynx; EV – Esophageal Voice; TV – TracheoesophagealVoice

Most subjects identified the positive emotions perceived as happy, as expected. From the positive emotion, we only obtained a response of surprise, which fact is justifiable by the attitude/initial response to given news and not with the perceived emotion after the news. In the negative context, it was found several qualifications, in which the identification of surprise and aversion were more frequent. This fact was interpreted by the same reasons of happinessplus, in negative context, the internal emotional self-regulation in response to the stimulus is often distorted by denyinga hypothetical situation. These findings agrees with the results of another study ¹⁸ , in which it was found that the sensation of the body, in a given state, represents the individual consciousness of self-focus, which means that the experienced emotion awareness is not objective and as the emotional transmissionis innate, this can lead to the individual cannot be awareof it and cannot actually referthe emotion he transmitted. Nonetheless, scientific evidence emphasizes that mot healthy individuals are able to identify and facially transmit their emotional states and this is a likeable fact for their social interaction ¹⁰ , while some pathological groups, such as depressive population, cannot do it ¹¹ .

Intelligibility in different types of oral communication

As we can see in Table 4, only the communication by electronic larynx and tracheoesophageal prosthesis are perceived with a positive intelligibility, a fact that may be related with the characteristics of quality and audibility inherent to them, which corroborates with other studies, once the authors refer that these individuals have a functional speech and with high intelligibility ^2.19 despite the flattened prosodic characteristics characterizingthe electronic larynx mentioned by some authors. On the other hand, the group with esophageal voice was one of the groups that presented worse intelligibility, and such inference will also meet the results of a study once performed ² . Regarding the group with whispered speech, no studies were found that describe the intelligibility of these individuals; however, based on what was said by individuals during the that classification, it appears that some people have said that “they don’t hear any sound and they only saw mouth movements”; such facts are corroborated by two authors, when they define this type of communication as the absence of any sound vibration and the surrounding air can only be modulated by articulators and, therefore, it is barely audible/understandable ²⁰ .

Table 4
– Speech intelligibility in different types of oral communication

	Types of communication
	WS	EL	EV	TV
	Mean±SD	Mean±SD	Mean±SD	Mean±SD
Speech intelligibility	3,12 ±0,66	6,75±2,76	2,95±1,02	8,44±1,23

SD – standard deviation; WS – Whispered Speech EL – Electronic Larynx; EV – Esophageal Voice; TV – Tracheoesophageal Voice

By  Kruskall-Wallis test it was found the existence of differences in intelligibility statistically significant among various communicative forms (p≤0.05). Mann-Whitney test allowed us to verify that the aforementioned differences are between whispered speech and tracheoesophageal voice (p= 0.03) and the esophageal voice and the tracheoesophageal voice (p= 0.02), in both tracheoesophageal voice evidences a significantly superior intelligibility.

Automatic Language

In the task of automatic languagewe can check in table 5 that in neutral context normospeakers are facially less expressive compared to laryngectomized individuals, once they just presented facial activity at the level of their eyebrows, while laryngectomized also presented movements at the level of the mouth, although restricted, which agrees with the results of a study where the author mentions that these facial segments are not or are slightly activated in non emotional contexts ²¹ .

Table 5
– Facial expression in normospeakers and laryngectomized during the task of automatic language

	Normospeakers			Laryngectomized
	Neutral	Happiness	Sadness	Neutral	Happiness	Sadness
	M±SD	M±SD	M±SD	M±SD	M±SD	M±SD
Number of eye movements		1,00±0,00	1,00±0,00		1,00±0,00
Total duration of eye movements (sec)		0,16±0,04	0,15±0,00		0,08±0,06
Average duration of each unit eyes (sec)		4,56±2,42	0,41±0,00		2,62±0,74
Number of movements of the mouth		1,00±0,00	1,00±0,00	1,00±0,00	1,00±0,00
Total duration of mouth movements (sec)		0,19±0,00	0,14±0,00	0,05±0,00	0,06±0,01
Average duration of each unit mouth (sec)		0,86±0,00	1,03±0,00	0,05±0,00	0,81±0,66
Number of movements of the eyebrows	1,00±0,00	1,00±0,00	1,00±0,00	2,00±2,00	2,00±2,00	2,00±1,41
Total duration of eyebrows movements (sec)	0,12±0,03	0,12±0,06	0,10±0,01	0,15±0,14	0,11±0,13	0,13±0,10
Average duration of each unit eyebrows (sec)	1,97±1,22	2,97±2,63	3,50±3,54	3,81±2,42	2,98±2,96	2,26±2,50

M - Mean SD – standard deviation

In happy and sad emotional contexts, one can verify that normospeakers present only a few more movements and with greater total and unitary duration in most facial segments, and also, more units of action associated with the expression of happy; this fact can be justified with some findingsin which some authors of the study refer that, after the access to emotions of evocation, the feeling endures more, what it can be translated into longer facial movements to keep the expression produced active and which translates a greater emotional intensity during the proceedure ²² . In the sad context there aren’t actions of mouth and eyes movements by laryngectomized, which can be related to the fact that the evocation in these subjects is a reality already experienced and assume autobiographical characteristics ²³ that leadthem to be more apathic, by a psychological mechanism of compensation ^5.24 . 

The fact that in these three contexts normospeakers and laryngectomized support the speaking in eyebrows segment, leads to the findings of several authors ²⁵ , where it was found that the upper part of the face transmits more emotional information and that, during the various emotional contexts, is approximately 120% more active than in neutral context.

Happy spontaneous speech 

In Figure 1, we can verify the activation of all facial segments in both groups, as we verify that the eyes are the facial segment less active in both groups, being a fact corroborated in several studies ²⁵ , where the authors refer that the role the eyesplay is to intensify the main role of the mouth during the transmission of emotions of happiness and sadness. We can also verify the presence of few units of action activated by individuals, and the only one which is associated withhappiness and which is activated by them meets the facts where it was found that for the characterization of the expression of the emotion of happiness, one of the movements is the elevation of the cheeks (A6) ²⁶ . Among all the facial segments which were analyzed, the mouth is the most active segment in both groups, being slightly higher in laryngectomized. These results can be justified with the fact that the lower part of the face contains higher levels of effective action and isit is relatedtospeech production, once the articulatory processes play a crucial role in the movement of the face ²⁵ . Despite the previously mentioned, we didn’t find statistically significant differences between both groups (p≥0.05).

Figure 1
– Facial expression of happiness in normospeakers and laryngectomized during happy spontaneous speech

Sad spontaneous speech 

Finally, regarding sad spontaneous speech  (figure 2), we can verify that the eyes are the facial segment that presents the smallest number and duration of movements, which coincides with the results of happy spontaneous speech and meets, once again, several studies ²⁵ . However, these authors report that the eyes do not carry emotions of sadness or happiness, but only intensify the expression given by the mouth and such fact is not corroborated by the findings of the present study, oncewe realized the presence of units of simple and complex action associated withboth emotions. We also found in this spontaneous speech that the mouth is the facial segment more active and slightly higher in laryngectomized, and this fact was also observed in population with vocal pathology ¹⁴ . 

Figure 2
– Facial expression of sadness in normospeakers and laryngectomized during sad spontaneous speech

However, no statistically significant differences were found between both groups (p≥0.05).

Transmission of happiness and sadness

From the observed units and those which refer to the emotion of happiness (Figure 3), in happy spontaneous speech, these ones will meet several findings ²⁶ , once the pulled corners of the mouth (A12) were observed, by both groups, but only in combined movements. At this level, werealized the presence of more movements and with greater temporal supportassociated withhappinessby normospeakers, which may highlight the idea advocated by several authors when they say that laryngectomized, as they showed a change in communication, represent difficulties in expression of thoughts, feelings and/or emotions. On one hand, at the level of the eyes, werealized the presence of more movements but with less temporal support, by laryngectomized, which indicates that, in these individuals, the upper part of the face intensifies the emotional expressivity and the speaking. On the other hand, the units present at the level of the mouth, in sad spontaneous speech, existin a greater number and they are more complex compared to happy spontaneous speech; however, their presence leads to several autores ²⁶ when they mention that the characteristics of the expression of sadness comprise the mouth half-opened (A25) with elevation of the upper lip (A10), corners of the mouth in tension (A12) and facing down (A15). The fact that we found the presence of units of action concerning happiness in this spontaneous speech by both groups, but in a greater number by laryngectomized, raises the hypothesis that these individuals exhibit more strategies of coping with the intention of reducing the emotional disorder generated by emotional content, keeping the hope and optimism and denying both the situation and consequences ²⁷ . Comparing the facial expression in bothemotional contexts we can verify that normospeakers are, thus, more expressive facially, which indicates that this difficulty by laryngectomized can be related to the fact that theyare slightly older; there is a decline in the expression of emotions because of the increasing age, and these emotions are less frequent and less intense ¹² ; This fact leads to the conception defended by another autor ² .

Figure 3
– Transmission of happiness and sadness in normo speakers and laryngectomized during happy and sad spontaneous speech

According to face behavior of total laryngectomized individuals that present different types of oral communication, in happy spontaneous speech we can verify that the group withelectronic larynx presents a greater number of movements at the level of the eyes and mouth, as compared with the other types of oral communication; these results can be supported with the importance of the non-verbal resources in communication for the complementation of what is said orally by these individuals ¹⁹ . This complementation of non-verbal resources, this type of oral communication, may also justify the fact that these individuals have a higher number of units of action, at the level of the eyes, associated with happiness, i.e. once the aim of happy spontaneous speechwas to direct thespeech and the behavior of the face for the expression of happiness, these individuals needed to use more resources and, in turn, to activate more movements that transmit this emotion. However, the fact that these individuals present less movements associated withhappiness, at the level of the mouth, can be justified by the fact that the their voice is partially modeled and amplified by an electronic device ¹⁹ and, therefore, it leads to that the individual does not make full usage of all their articulators to produce voice and full decrease the activity of this segment. On the other hand, we can verify that individuals who use tracheoesophageal voice communication presented, at the level of the mouth, long movements and activated more codes associated with the expression of happiness in this segment, being this fact supported by previous studies, where it was found that these individuals have achieved a functional speech and of good quality, which leads to a less difficulty in express themselves ¹⁹ .

Concerning sad spontaneous speech, we can verify a great variability of results in the various types of oral communication whereby at the level of the eyes, it was found that the group with tracheoesophageal voice presented more and longer movements being these findings substantiated by other studies ¹⁹ . The good functionality of the speech acquired by these individuals is also related to the fact that we have observed in them more codes associated with the expression of sadness, in this spontaneous speech, because this facility in expressing themselves leads to that the same individuals, after the access to the emotions of the spontaneous speech, would be able to lengthen the feeling, by the analyzed segments ²² . At the level of the mouth we can verify that the group with whispered speech presents the largest and longer number of movements, as these results may be related to the fact the whispered speech is only modulated by articulators, through the exploitation of the circulating air within the resonance spaces ¹⁹ and they require a more and a larger temporal support of the movements of this segment to communicate. We could also verify that this group was more expressive through the mouth in order to express happiness, in this spontaneous speech, which can be related to the production of voice bythe absence of sound vibration,and, inherently, leads to the fact that these individuals present more strategies ofcoping, which increases optimism ²⁷ , positively expressed by mouth. 

Facial expression in laryngectomized with low and high intelligibility

Analyzing the behavior of the face in the groups with low and high intelligibility (Table 6), we canverify that the same did not present statistically significant differences in facial expression in both evocations (p≥0.05), except in the average duration of each unit of the eyes in the sad spontaneous speech (p-value= 0.01), and the group with low intelligibility presented significantly more durable movements. This fact can be justified by the role the eyes play, which is to intensify the main role the mouth plays in the transmission of emotions of happiness and sad ²⁵ , which leads to the fact that, in this spontaneous speech, the activated units have had a great variability in terms of temporal support, while they helpedother segments. Nevertheless, this finding may also highlight the role the eyes plays in the transmission of sadness, which fact is confirmed in another study ¹⁴ , and that in the absence of oral competence, by low intelligibility, it becomes more active in assisting the desired transmission of emotional message.

Table 6
- Comparisonof facial expression in thegroupsoflowandhighintelligibilityduringhappyandsadspontaneousspeech

	Happiness			Sad
	Lowintellig.	Highintellig.	p-value	Lowintellig.	Highintellig.	p-value
	MeanRank	MeanRank		MeanRank	MeanRank
Number of eye movements	4,92	5,17	0,893	4,92	6,38	0,43
Total duration of eye movements (sec)	4,33	6,33	0,291	5,08	6,13	0,59
Average duration of each unit eyes (sec)	6,17	2,67	0,071	7,33	2,75	0,01
Number of movements of the mouth	7,56	6,10	0,504	5,38	8,75	0,12
Total duration of mouth movements (sec)	6,44	7,90	0,508	6,19	7,13	0,67
Average duration of each unit mouth (sec)	7,13	6,80	0,884	5,38	8,75	0,12
Number of movements of the eyebrows	6,94	5,63	0,548	7,94	5,50	0,25
Total duration of eyebrows movements (sec)	5,88	7,75	0,394	7,81	5,70	0,34
Average duration of each unit eyebrows (sec)	6,13	7,25	0,610	7,88	5,60	0,30

Statisticaltest: Mann-Whitney (p≤0,05)

Concerning the observed units of action that refers to the expression of happiness, the group with low intelligibility presented more complex movements, greater in number at the level of the mouth movements and longer, either for the eyes, or for the mouth, which can be related with the time of operation, since this group presents a greater average time of surgery ,andthen refuting the idea that, when he/she passes the minimum time after the operation (six months), the individual thus increases the ability to accept himself/herselfinthe society, increasing their social relations, transmission of emotions and improve communicative act ²³ . To these facts one can also associate the increase of strategies of coping, as they show a greater adaptability to situations they are exposed daily ²⁷ ; the increase of these strategies also justifies the presence of codes associated with the expression of happiness in spontaneous speech sad. On the other hand, we could verify that the group with high intelligibility presented more movements associated with the expression of sadness, and these results can be corroborated with the ideathat improving the functionality of speech the individual is more capable tocommunicate and make a full usage of all the communication skills (verbal and non-verbal) daily ² . In This way, in addition to the rehabilitation of oral communication, it is important to promote the facial expressiveness in postoperative period to make the process of adaptation to the new condition easier ⁵ .

On the other hand, the same groups did not present statistically significant differences in emotional demonstration of happiness and sadness (p≥0.05). These findings meet the results in depressive population ¹¹ or even, in population with vocal pathology ¹⁴ , and in the area of schizophrenia ²⁸ , once in these studies there were differences in emotional demonstration in one or in both emotions. Taking into account that there are no differences in emotional demonstration, data do not corroborate the study, where the author ¹³ mentions that the expression of emotions can be altered in populations which exhibit unique changes of the physical structures, once the ability to express emotional states can be interrupted or even lost because of andeficiency underlying a physiological or neurologicalsubsystem. The absence of changes at this level also leadsto the small impact of communication in the transmission of emotions and to the lasting quality of life. These abilities are also likely to play an important role in our daily social interactionsand to promotesuccess in events such as job interviews, personal relationships and social position ²⁹ .

FINAL COMMENTS

We founded that laryngectomized, regardless their communicative way, are less facially expressive than normospeakers, which translates a non-usage of their non-verbal communicative potential as an aid to oral communication which could lead to a compromise in the transmission of emotions and inherently of family life, social and affective. 

One can understand that, in spite of all the consequent changes, the surgery and the possibility to continue, or even,to establish new plans, predominated over the perceptions of any losses and obstacles with which the individuals were confronted with. It would be expected that laryngectomized, by presenting a change in oral communication, optimized the emotional transmission by the face, once this transmission is bimodal, and such a fact was not verified; however, these results should not be generalized, once the sample is small and should not be ruled out the idea that this area of intervention should be assessed and studied, by encouraging these individuals to make usage of all non-verbalpotentials, promoting a more effectively communicative act in interpersonal relations.

Since this is an exploratory study and also by the lack of studies conducted with the same methodology, it has become difficult to carry out the discussion of the results, and this is one of the limitations of this study. Another limitation was the time of completion of the study and the time required for the analysis of facial expression that it doesn’t made possible to obtain a sample with a significant number, for further generalizations. The fact that we need to record the individuals and they don’t feel comfortable with a video camera and, consequently, the fact that there isn’t a therapeutic relationship with the researcher, leads to the condition that they are not totally in a natural environment, which can then compromise the faithful transmission of emotions during the application of the methodology of the study. Despite these limitations, the problematic questions initially placed were answered and the goals achieved.

We can, therefore, walk in the direction of improving the communicative resources, in order to seek a more natural speech and decrease the strange feelingby the listener. In this perspective, the expressiveness should be worked to benefit the communication and future studies should be conducted in this area, in order to help in the rehabilitation of these patients and to better understand the impact that verbal communication disrupted by laryngectomy has in non-verbal communication, presenting more effective therapeutic approaches in rehabilitation of the communication of these individuals. It is then proposed that future studies should be performed in women, because there is a certain prevalence of this pathology in these individuals and we do not know if there are differences among them. Finally, to avoid some methodological errors, it is proposed that in future studies, the recording of the tasks is not held in the same days and that the tasks of spontaneous speech are not requested by the researcher of the study, in order to avoid the surprise factor in individuals, not compromising the self-perception of the emotions and obtaining a greater time of speaking to further analysis of the emotional facial expression. 

CONCLUSION

According to emotional quantification, laryngectomized patients with electronic larynx were those who felt happiness and sadness with greater intensity, and those who used esophageal voice showed higher intensity of sadness. Either normospeakers or laryngectomized perceived the positive emotion in a more intense way than the negative. 

On the other hand, according to the emotional qualification, the majority of individualsidentified the positive emotion as happiness, but, in negative context, the identification of the surprise and aversion were more frequent.

Communication through electronic larynx and tracheoesophageal prosthesis showed good intelligibility, but esophageal voice and whispered speechshowed worse intelligibility. Tracheoesophageal voice showed a significantly higher intelligibility.

In automatic language, it was found that in neutral context normospeakers are less facially expressive compared to laryngectomized, but in emotional  context happiness and sad, we could found that normospeakers presented more and lasting movements,and also more units of action associated with the expression of happiness.

During happy and  sad spontaneous speech, there were differences in descriptive facial expression in both groups, however no statistically significant differences were found between them.

In bothemotional contexts it was found that normospeakers are more facially expressive.

There were no statistically significant differences in facial expression in both evocations between the groups of low and high intelligibility, exceptingthe average duration of each unit of eyes in sad spontaneous speech; the same groups did not show statistically significant differences in emotional demonstration of happiness and sadness.

ACKNOWLEDGEMENTS

To Prof. Dr. Michael Kipp by permitting the access to Anvil program; to Hospital Garcia de Orta by participating and allowingthe collection of sample; to all our colleagues for their help, encouragement and availability provided and, finally, to all the participants who accepted enthusiastically participate in this study, in particular the laryngectomized patients for their encouragement and life lessons given to us.

REFERÊNCIAS

• ¹
  Carvalho D. Cancro e actividade profissional: contributo para o estudo nos cancros de cabeça e pescoço epulmão, brônquios e traqueia na região centro, de 2002 a 2006. [Dissertação] Coimbra (Portugal): Faculdade de Medicina da Universidade de Coimbra; 2009.
• ²
  Jungerman I. Qualidade da voz e da comunicação em pacientestratados do câncer de laringe e hipofaringe:análise de diferentes ouvintes e sua relação coma qualidade de vida[Dissertação]. São Paulo (SP):Fundação António Prudente; 2009.
• ³
  Steyn H. The laryngectomy patient’s need for support groups in a hospital setting: a social work perspective [Dissertation]. Stellenbosch (País):University of Stellenbosch; 2009.
• ⁴
  Farrand P, Duncan F. Generic health-related quality of life amongst patients employing different voice restoration methods following total laryngectomy. Psychol, Health Med. 2007;12(3):255-65.
• ⁵
  Noonan B, Hegarty J. The Impact of Total Laryngectomy: The Patient’s Perspective. Oncol Nursing Forum. 2010;37(3):293-301.
• ⁶
  Bliss-Moreau E, Owren MJ, Barrett F. I like the sound of your voice: affective learning about vocal signals. J Exp Soc Psychol. 2010;46(3):557-63.
• ⁷
  Biele C, Grabowska A. Sex differences in perception of emotion intensity in dynamic and static facial expressions. Exp Brain Research. 2006:171(1):1-6.
• ⁸
  Graham SM. Facial Expression of Emotion.EncycSoc Psychol. Sage: publications; 2007.
• ⁹
  Ekman P, Keltner D. Facial Expressions of Emotion. In: Lewis M,Haviland-Jones J.Handbook of emotion. 2ª ed. New York: Guilford Publications, Inc; 2000. p. 236-49.
• ¹⁰
  Surguladze SA, Senior C, Young AW, Brébion G, Travis MJ, Phillips ML. Recognition accuracy and response bias to happy and sad facial expressions in patients with major depression. AmericanPsychol Assoc. 2004;18(2):212-8.
• ¹¹
  Kan Y, Mimura M, Kamijiima K, Kawamura M. Recognition of emotion from moving facial and prosodic stimuli in depressed patients. J NeurolNeurosurgPsychi.2004;75:1667-71.
• ¹²
  Ebner NC, Riediger M, Lindenberger U. Faces – a database of facial expressions in young, middle-aged, and older women and men: development and validation. J Behav Research Methods. 2010;42(1):351-62.
• ¹³
  Izdebski K. Emotions in the Human Voice: Clinical Evidence. 1ª ed. San Diego: Plural Publishing; 2008.
• ¹⁴
  Correia PG. Sob o signo das emoções: expressões faciais e prosódia em indivíduos com perturbação vocal [Dissertação].Lisboa (Portugla): Instituto de Ciências da Saúde da Universidade Católica Portuguesa e Escola Superior de Saúde de Alcoitão; 2007.
• ¹⁵
  Gonçalves B, Fagulha T. The Portuguese version of the center for epidemiologic studies depression scale (CES-D). Europ J Psychol Assessment. 2004;20(4):339-48.
• ¹⁶
  Kipp M.Gesture generation by imitation – from human behavior to computer character animation. Boca Raton, Florida: Dissertation.com; 2004.
• ¹⁷
  Ekman P, Friesen W. Measuring facial movement. Environ Psychol Nonverbal Behav. 1976;1(1):56-75.
• ¹⁸
  Sonnemans J, Frijda NH. The structure of subjective emotional intensity.Cognition and emotion.1994;8(4):329-50.
• ¹⁹
  McAuliffe MJ, Ward EC, Bassett L, Perkin K. Functional speech outcomes after laryngectomy and pharyngolaryngectomy. Arch Otolaryngol Head Neck Surg. 2000; 126(6):705-9.
• ²⁰
  Pietruch RW, Grzanka AD. Vowel recognition of patients after total laryngectomy using mel frequency cepstral coefficients and mouth contour. J Automatic Control. 2010;20(1):33-8.
• ²¹
  Legal E. Expressões faciais de emoções: desenvolvimento das percepções de assimetrias [Dissertação]. São Paulo (SP): Instituto de Psicologia da Universidade de São Paulo; 1996.
• ²²
  Schnall S, Laird JD. Keep smiling: enduring effects of facial expressions and postures on emotional experience and memory. Cognition and Emotion.2003;17(5):787-97.
• ²³
  Hannickel S, Zago MM, Barbeira CS, Sawada NO. O comportamento dos laringectomizados frente à imagem corporal. Rev Bras Canc. 2002;48(3):333-9.
• ²⁴
  Paes C, Toledo PN, Silva, HJ. Fonoaudiologia e estética facial: estudo de casos. Rev CEFAC. 2007;9(2):213-20.
• ²⁵
  Busso C, Narayanan SS. Interplay between linguistic and affective goals in facial expression during emotional utterances. Proceedings of 7thInternational Seminar on Speech Production; 2006; Ubatuba, Brasil. Los Angeles (CA): Inproceedings; 2006. p.549-56 .
• ²⁶
  Kohler CG, Turner T, Stolar NM, Bilker WB, Brensinger CM, Guer RE, et al. Differences in facial expressions of four universal emotions. Psychi Research. 2004;128(3):235-44.
• ²⁷
  Tamayo MR, Tróccoli BT. Exaustão emocional: relações com a percepção de suporte organizacional e com as estratégias de coping no trabalho. Estudos Psicol.2002;7(1):37-46.
• ²⁸
  Silver H, Shlomo N, Turner T, Gur RC. Perception of happy and sad facial expressions in chronic schizophrenia: evidence for two evaluative systems. Schizop Research.2002;55(1-2):171-7.
• ²⁹
  Lõhmus M, Sundström L, Björklund M. Dress for success: human facial expressions are importante signals of emotions. AnnZoolFennici. 2009:46:75-80.

Publication Dates

• Publication in this collection
  Mar 2014

History

• Received
  19 Sept 2012
• Accepted
  10 May 2013