perception of the emotional content of speech by canadian and mexican children, adolescents, and...

7/28/2019 PERCEPTION OF THE EMOTIONAL CONTENT OF SPEECH BY CANADIAN AND MEXICAN CHILDREN, ADOLESCENTS, AND ADULTS

http://slidepdf.com/reader/full/perception-of-the-emotional-content-of-speech-by-canadian-and-mexican-children 1/15

international Journalof Psychology 16 (1981) 119-132

North-Holland Publishing Company

119

PERCEPTION OF THE EMOTIONAL CONTENT OF SPEECH BY

CANADIAN AND MEXICAN CHILDREN, ADOLESCENTS,

AND ADULTS *

Ken W. McCLUSKEYLord Selkirk School Division, Canada

Daniel C. ALBASUniversity of Manitoba, G m d a

Revised version received September 1980

The study at hand was undertaken to assess and compare the respective abilities of Canadian

and Mexican subjects to identify emotion in speech. Canadian and Mexican female teachers,

speaking whatever words they wished in their ow n languages,attempted to simulate four em*

tional states ( i e . , happiness, sadness, love, and anger). After thc initial recording, these samples

were passed through an electronic filter which removed the semantic content w u e leaving

intact the tonal qualities of the speech. The fdtered Vocal expressions were played to Canadian

and Mexican subjects 5 , 9 , 13, 17,25 ,45 , and 65 y w s of age. From 5 through 2 5 , there was a

progressive increase with age in ability to identify the emotion expressed. Ability did level off,

however, and eventually even decreased Ouch that the 65 year-~lds erformed less accurately

than the other adult groups). Overall, the Mexican subjects were significantly more sensitive to

emotion in speech than their Canadian counterparts. As well, both Canadian and Mexican sub-

jects judged speech samples from Mexican speakers more accurately than those from Canadian

speakers.

Any spoken communication between two or more people involves bothverbal and nonverbal elements. Indeed, much of the research in the areaof communication is based upon the notion that speech is composed oftwo distinct channels: the verbal and the vocal. Supposedly, the verbal

channel is composed of sound patterns resulting in syllables, words,phrases, sentences, and other linguistic forms conveying the semantic

meaning of speech. If the words spoken in a particular message are



K . W . McCluskey, D.C. Albas /Emotional w n t e n t of peech 121

was a progressive increase with age in ability to identify correctly the

emotional content of the voice samples.The next study with children along these lines was performed almost

four decades later by Dimitrovsky ( 1 964), who attempted to isolate the

vocal channel of speech by holding the verbal channel constant. In her

speech samples, speakers were asked to recite the same standard para-

graph while simulating the emotions of happiness, sadness, love, andanger. When these samples were played to children aged five throughtwelve, it was likewise found that the ability to judge vocal expression

of emotion was positively related to age. Fenster ( 1 967), using a similar

standard content (non-filtered) method, and McCluskey (1974), em-ploying both standard content and filtered speech samples, also found

that the ability to identify emotion in vocal expressions improves with

age. Most recently, Rosenthal and his co-workers have also noticed a

pronounced age effect with their PONS test, such that subjects exhib-

ited increased sensitivity to both filtered and randomly spliced speech

with age (Rosenthal e t al. 1979).To date, most developmental studies have dealt with American or

Canadian subjects. It would seem useful, then, to extend the scope ofinvestigation and approach the development of emotional sensitivityfrom a cross-cultural perspective. In the research at hand, Canada andMexico were chosen as the comparison cultures, largely because theyare so radically different from one another. Despite the fact that theyare both technically North Americans, pronounced differences in “emo-

tional style” seem to exist between Canadians and Mexicans. From ananecdotal point of view, Canadians often describe Mexicans as being“Latin blooded” and “highly emotional”. Conversely, Canadians andother Anglo races are typically stereotyped by Mexicans as being cold,

aloof, and “mechanial” (cf: Lewis 196 1 ) .

At an empirical level, there is evidence to indicate that emotional

expression does in fact vary from culture to culture. Specifically, with

respect to nonverbal communicative patterns, clear cultural differences

exist in terms of size of personal space zones, posture and angle of ori-

entation in interpersonal interaction, volume and tone of voice, and

amount of gesturing, eye-contact, and bodily-contact (cf: Albas e t al.

1976;Birdwhistell 1963;Hall 1959; McCluskey and Albas 1978; Mehra-



122 K .W.Mccluskey, D.C. lbas /Emotional content of speech

Mexican children, adolescents, and adults had smaller personal space

zones (and engaged in more bodily-contact) than either black or whiteAmericans. The nonverbal styles of expressing emotion were different

between the Mexican and American subjects, and differences found in

childhood tended to persist into adult life. Further, in a preliminarystudy examining sensitivity to tone of voice, Mexican children aged six

through 11 were found to identify the emotional content of filteredspeech more accurately than Canadian children of corresponding age

(McCluskey et al. 1975).For the purposes of the present study, the issue is whether or not the

subtle nonverbal differences between the Canadian (Anglo) and Mexi-can (Latin) cultures extend to even the tonal level. Although there is

some empirical support for the intuitive notion that communication of

affect differs in the two cultures (cf:Argyle 1972; Mehrabian 1972),the question as to whether there are cross-cultural differences in sensi-tivity to the emotional content of speech has not been adequately ex-plored. Indeed, there seems to be only one study addressing itself tothis issue (McCluskey e t al. 1975). The major amof the present

research, then, was to compare the ability of Canadian and Mexicansubjects of various ages to judge the affective content of filtered vocalexpressions recorded by Canadian (English-speaking) and Mexican(Spanish-speaking) communicators.

Virtually all research involving recitation of non-meaningful material

and standard content has shown that children are able to judge emotionin speech more accurately as they grow older (Dimitrovsky 1964;Fenster 1967; Gates 1927). Also, with filtered speech where there were

no words or verbal content for the children to “hook onto”, the samegeneral trend has been reported (McCluskey 1974). In light of suchresearch, it was hypothesized that:

(1) there would be a progressive increase with age in children’s abilityto identify the emotional content of filtered speech in both theCanadian and Mexican cultures.

As well, since no complete developmental comparison between Cana-

dians and Mexicans has been undertaken, it was decided to presentspeech samples to Canadian and Mexican subjects of various ages (ie.,



K.W. McCluskey,D.C. Ibas Emotional mntent of speech 123

learned early in th e chi ldhood years tend t o persis t into ad ul tho od, it

was felt that any cross-cultural differences in sensitivity to tone might

well perseverate through the entire life cycle. It is known that certain

differences exist in th e comm unication of e mo tion in the Anglo and

Latin cultures (Argyle 19 72 ; Baxter 19 70 ; Hall 1 9 5 9; McCluskey et al.

19 75 ; McCluskey et al. 19 73 ; Mehrabian 1972). Given what research

there is in the area, and assuming that the widespread popular belief

th at Mexicans are extremely em otio na l may perhaps have som e basis in

fact, th e most logical and parsimonious ex pe ctati on was tha t:

(2 ) Mexican subjects a t all age levels would ide ntify th e em otio nal con-ten t of th e f i l tered speech mo re accurately than Canadians.

Although no t mu ch work has been do ne in th e area, there are indica-tions t ha t cross-cultural differences exist in t h e transmission as well as

in the reception of affective information (McCluskey e t al. 1975) .

Although i t might be premature to s ta te i t as a formal hypothesis , it

was expected th at Mexican speakers would be able t o comm unicate th e

em otio n th ey intended more effectively tha n Canadians.

The experimmt

M e t h o d

S u b j e c t s

Sixty boys from the Lord Selkirk School Division no. 11,Selkirk, Manitoba, Can-

ada, and 60 from schools in Colonia del Valle, Mexico City, Mexico, served as Ss.

More specifically, 15 boys from each country at each of the ages 5 , 9, 13, and 17took part. N o S at any age level was within two months of a birthday at the time

the study was run.

As well, 15 male teachers at each of the mean age levels 25, 45, and 65 were

selected from bot h Manitoba and Mexico City schools. Due to the difficulty in find-

ing enough individuals at each age level, Ss ranged (plus or minus two years) around

the mean ages listed above. That is, the ages actually ranged from 23-27,43-47,

and 63-67. All Ss in the 65 year-old group had retired from active teaching (but

were contacted through their former schools). No S whose I.Q. was known to be

above or below the normal range (on th e Peabody Picture Vocabulary Test, Quick

Test, or any group intelligence test) was included in the sample. As well, the hearing

of every S participating in the study was tested within normal limits on a Bell Tone



12 4 K. .Mccluskey, D.C. Albas 1Emotional content of peech

tro l fo r th is variable. All Ss in Selkirk w ere chosen from “m iddle class” schools, and

each student’s cumulative file was checked with regard to father’s occupation and

family econo mic background. T he same general procedure was followed in Mexico,where all b u t the lowest class schools are private institutions . Since all Mexican stu-

dents in the study attended private schools, their families most probably had rela-

tively good incomes. And since all adult Ss in both cultures were (o r had been)

teachers, they too would likely fall roughly in the “m iddle” socioe conom ic class

bracket.

App arat us and measuring instrument

Th e original speech samples were recorded on a Sony Stere o Cassette Deck, modelTC-199SD. Th ese samples were the n rendered unintelligibIe thro ugh t h e use of an

elec tron ic filter, which passed frequencies from 100-4 50 cycles per secon d witha 60-decibel per octave attenuation at th e uppe r level. A second Sony Stereo Cas-sette Deck was used during the filtering proced ure. After filtering, the fin al samples

were presented to Ss via ex ternal speakers (S ony Spea ker Sy stem Model S S - 2 2 5 0 ) .

In several studie s of emotional communication, adult Ss are asked t o choo se th e

one particular word from a checklist of adjectives that they feel best describes the

emotional content of each vocal expression (Davitz and Davitz 1959; Soskin and

Kauffman 1961). However, when dealing with Ss of various ages, it is inappropria te

to use an exp erim enta l task involving either reading or w riting, since age differences

in these abili t ies may confound results . Consequ ently, in something of an attem pt

to control for this problem, all Ss were asked to make their identification of each

speech sample by pointing to one of fo ur stick figure drawings that represented thefo ur em otio nal states under consideration (Le. . happiness, sadness, love, and anger).

Each picture was drawn on a separate card, and the cards were presented in random

order to each S. The stick figure drawings used here were identical to those origi-

nated by Dimitrovsky (1964) .

Speech samples

Thre e Canadian and three Mexican elementary school teachers, all female and all

selected b y the ir principal fo r their “sensitivity”, were asked t o express vocally t heemotions of happiness, sadness, love, and anger. Each of these female teache rs was

between 25 and 35 years of age. Since the three teachers in each cu lture expressed

each of the four emotional meanings, there was a total of 24 speech samplesrecorded. The order in which the four emotions were acted was randomized for

each speaker. However, i t was stipulated th at n o m ore th an tw o speakers could be

asked t o comm unicate the same emotion f irst . In addi t ion , the order in which the

emotions were expressed had to be different for al l teachers.

T o avoid artif iciali ty arising from recitation of meaningless or neutral material,

each Canadian teacher was allowed t o say any tw o sentences (in English) th at came

to mind in a n a t t e mp t to simulate each of th e four emotions. Similarly, th e Mexi-can teachers attempted to express the four emotional states in Spanish. Each



K.W.McCluskey. D.C. Albas /Emotional content of peech 12 5

samples. Each vocal expression was considered to communicate a particular emo-

tional state simply because the speaker intended to express the emotion in ques-

tion. Also, prior to filtering, the speech samples were played to and rated by tenmale teachers (Canadian samples were rated by Canadian teachers and Mexican sam-

ples by Mexican teachers). Before any sample was included for use in th e study, at

least eight of the ten judges had t o agree that the speaker was indeed communi-

cating the emotion she intended to communicate.

After the 24 samples were recorded, they were arranged on tape in random order

(with the restrictions that no emotion could appear more than twice in a row, no

more than two speech samples from any one speaker could appear in succession,

and no more than two teachers from the same culture could show up in succession).

Following this randomization procedure, the semantic content of the speech was

removed by means of the electronic filtering device.

Procedure

The procedure paralleled Dimitrovsky’s (1964) methodology as closely as possible.

All Canadian S s in every age group were seen individually by the same experimenter

(who of course gave the instructions in English). Similarly, all Mexican Ss were seen

by a single Mexican experimenter (giving exactly the same instructions translated

in to Spanish). The experimental situation was as similar as possible in both coun-

tries.

S s in each age group were trained to associate each of the four pictorial represen-

tations with the appropriate corresponding emotional state. In other words, theywere required to point to the happy woman, the sad woman, the angry woman, and

the loving woman. This procedure was repeated until a criterion of two correct

identifications of all four drawings was reached N o S had any difficulty with this

task.

Canadian Ss were told that they were going to hear the voices of ladies speaking

from “far away”, and that they would not be able to make out the words clearly.

They were then given a slightly modified version of Dimitrovsky’s (1964) instruc-

tions, which read:

You are going to hear different ladies say th ings in different ways. Some will sound happy,some sad, some angry, and some loving. Listen to each one and point to the lady it sounds ike.

Does it sound like the happy lady, the sad lady, the angry lady, or the loving lady?

All Mexican S s were given exactly the same instructions in Spanish.

The filtered recordings were than played to Ss, who made their 24 choices.

After hearing each speech sample, each S had a lOsec interval to point to the stick

figure drawing that he thought best represented that particular vocal expression.

Each response made by each S to each stimulus was recorded. In the rare instances

when an S took longer than 10 sec to respond, the tape was stopped until he made

his judgment. No speech sample was played more than once.

Resul t s



126 K. .Mccluskey, D.C. Albas 1Emot ional content of speech

th e necessary err or values, an analysis o f variance containing seven levels of age, tw o

levels of cult ure of listene r, and tw o levels of culture of speaker was performed on

these d ata. Since directional hy potheses w ere indicated, tests were done by [Ipriori(planned) comparisons in order to maximize power. In essence, the comparisons

revealed that all main effects were significant. When all Ss in both cultures were

considered, tests of trend yielded significant linear ( F =61.90, df= 1,196, p <0.01) a n d q u a d r a t i c ( F = 8 1 . 4 7 , df= 1,196, p < O . O I ) age trends. As well, overallplanned comparisons revealed significant effects for culture of listener ( F =25.35,df= 1,196, p <0.01) and culture of speaker ( F =163.71, df= 1,196, p <0.01).

The mean scores for all age levels in both cultures are presented graphically infig. 1 .

Fig. 1 , then shows the responses of Canadian and Mexican Ss of v arious ages to

th e 12 vocal expressions simulated by C anadian speakers and to th e 12 samplesenacted by Mexican speakers. The significant age effect is accounted for by the

gradual and progressive improvement in ability as Ss get older. From 5 through 25

years of age, there was a n increase in accuracy from every younger to every olderage level. At the 25 year-old level, however, there was a levelling off, and for the

oldest age groups (i.e.. the 45 and 6 5 year-olds), there was even a decrease in accu-

racy as compared to the performance of the 2 5 year-old group.

t

11

/--

a 1w

2 1

Speaker Listener

Mexican

Canadian Mexican HCanadian D-----cI

Mexican

Canadian



K . W. McCluskey, D.C. Ibas /Emotional mntent o f speech 12 7

In terms of age effects, tests for trend involved pooling of error terms (cf Kirk

1969). These tests for trend showed that Ss in general became more accurate with

age in all four conditions shown in fig. 1. That is, there were significant linear

trends for Canadian Ss listening to Canadian speakers ( F=30.50, df =1,388, p <0.01), for Canadian Ss listening to Mexican speakers ( F =29.55, d f = 1,388, p <0.01), for Mexican Ss listening to Canadian speakers ( F = 24.81, df= 1,388, p <O.Ol), and for Mexican Ss listening to Mexican speakers ( F = 26.13, df= 1,388,

However, as fig. 1 indicates, scores do tend to level off with increasing age, peak-

ing at 25 years and falling off somewhat thereafter. This decrease in sensitivity after

25 years of age appeared to be meaningful, as tests of trend also revealed significant

quadratic trends for Canadians listening to Canadian samples ( F =31.05, df=

1,388, p <0.01), Canadians listening to Mexican samples ( F = 45.30, df= 1,388,

p <0.01), Mexicans listening to Canadian samples ( F = 33.54, df= 1,388, p <0.01), and Mexicans listening to Mexican speech (F=36.88, d f = 1,388, p <0.01).

As fig. 1 shows, the significant listener effect is accounted for by the obvious

superiority of Mexican Ss in ability t o identify the emotional content of bot h Cana-

dian and Mexican speech samples. Indeed, at every age level examined in the study,

the Mexican Ss were more accurate in judging the emotional content of the voice

samples than were the Canadians. Planned comparisons revealed that this cultural

difference was significant at the 5 year-old ( F =4.67, df= 1,196, p <0.05) and

9year-old (F=8 .29,d f=l ,1 96,p< O.Ol) level s , but not at the 13 (F =3 .6 9, df =1,196, p>0.05), 17 (F =3.69, df= 1,196, p > 0 . 0 5 ) , 25 ( F =2.43,'df = 1,196,

p >0.05), 45 ( F =2.07, df= 1,196, p >O.OS), and 65 year-olds ( F =2.07, df =

1,196, p >0.05) levels.

Fig. 1 also depicts the significant culture of speaker effect. Both Canadian and

Mexican Ss were clearly more sensitive to the emotional content of the Mexican

speech samples than to the Canadian speech. Planned comparisons showed that

both Canadian Ss ( F = 80.30, df= 1,196, p <0.01) and Mexican Ss ( F =83.43,

df= 1,196, p<O.Ol ) were more sensitive to the Mexican speech than to the

Canadian speech.

In summary, he major findings of t he present study were: first, discriminationability was a progressively increasing function of age from childhood through t o the

25 year-old level. At this point, things seemed t o level off, until eventually the 65

year-old group performed significantly more poorly than the other adult groups.

Second, Mexican Ss were able to identify the emotional content of speech more

accurately than the Canadians. Third, Ss at all age levels in both cultures identified

the emotional content of Mexican vocal expressions more accurately than Canadian

speech.

p <0.01).

Discussion



128 K .W . Mccluskey, D.C. Albas 1Emotional content of peech

(Dimitrovsky 1964; Fenster 1967; Gates 1927; McCluskey 1974;

McCluskey e t al. 1975). Interestingly, the age trend was the same in

both cultures; from 5 through to 25 years of age, Canadian and Mexi-

can subjects showed a progressive increase in ability to identify the

emotional content of filtered speech. It was also interesting to note that

subjects tended to plateau and level off a t the 17 through 45 year-old

age groups. This finding is consistent with Rosenthal e t aZ.’s (1979)

work with the PONS test, which also showed that nonverbal decoding

ability levelled off between 20 and 30 years of age. Even though there

were some controls in terms of hearing and I.Q., it still appears that

humans function less efficiently as they reach “old age”. Certainly, the

data here indicate, at least with respect to tone of voice, that 6 5 year-

olds perform noticeably less accurateIy than the other adult groups.

Since hypothesis (2 ) was also upheld ( i e . , Mexican subjects of all

ages judged the speech samples significantly more accurately than their

Canadian counterparts), the results provide empirical evidence for the

intuitive notion that Mexicans are generally more sensitive to the emo-

tional content of the communication than are Canadians. Since this dif-

ference does in fact hold true at every single age level examined, itwould seem that cultural differences in the vocal reception of emotion

are established early in childhood.and perserverate through to old age.

As Baxter (1970) contends, cultural responses learned in childhood

appear to have a lasting effect.

In addition, the data show that Mexican speakers communicated

emotion more effectively than the Canadians (ie.. they were more fre-

quently able to communicate what they intended). Although there

were only three speakers from each country, this finding lends supportto the idea that differences exist in the transmission (encoding) as well

as in the reception (decoding) of affective information. Like Argyle

( 1 9 7 3 , who noted that Japanese are more sensitive to facial expres-

sions from cultures other than their own, the results here show that

Canadians are more accurate at judging speech from Mexicans than sam-

ples from other Canadians. Canadian speakers may simply be less

expressive than Mexicans, or there may be a different code operating

(ie., different “style” of expressing emotion). In any case, it does

indeed seem to matter greatly “who communicates to whom” (cf Mc-



K .W.McCluskey. D.C. Albas f Emotional content of speech 129

individuals from cultures that were “dissimilar” to the U.S. scoredpoorly on the decoding test (Rosenthal et al. 1979). That is, the more

cultural dissimilarity between the encoder and decoder, the less accu-rate the communication. This finding is not entirely consistent with thepresent results, which showed that Mexicans were more sensitive totone than Canadians regardless of the nationality of the encoder. This

could be due to the lack of comparability in Rosenthal’s Mexican andCanadian samples (there did not appear to be careful matching on ageand other variables), or to a difference between auditoryanly decoding(used in the present study) and decoding involving both auditory and

visual cues. Still, while the results of this study differ somewhat fromRosenthal’s, they are nonetheless in accord with Argyle’s (1975) inter-pretation that people can be more sensitive to nonverbal communica-tion from “outsiders” than to intra-cultural messages.

In an attempt to keep the design manageable and t o highlight what-ever differences might exist (cf.McCluskey and Albas 1978), the studyat hand focused on a single communication pattern: female speaker tomale receiver. While this approach did in fact facilitate these ends, it

comes at the cost of limiting generalizability, in that we cannot simplyassume that the same results would have been obtained ushg a differ-ent communication pattern. Another note of caution - pointing outthat the performance of subjects from one culture should not be viewedas being “better” or “worse” than that of subjects from the other -

might also be in order here. As Pike (1966) argues, to understand a par-ticular system or culture, it is necessary to look at it from within (ie.,

take an “emic” perspective) rather than from without (an external,“etic” view). In many ways, “better” and “worse” are etic and ethno-centric terms - “different” is perhaps more appropriate.

The findings here suggest that cultural differences in emotional per-ception are established ear1y.h life. Indeed, it may well be that Mexi-cans are more sensitive to speech because of learning experiences duringsocialization in childhood; the Mexican life style may simply be more“emotional” than that in Canadian society. Should this indeed be thecase, the implications for international understanding (especially withrespect to travelling, political negotiations, and the like) are self-evi-

dent. If, as this research suggests, people from different cultures inter-pret emotion in speech differently, effective cross-cultural communica-



130 K . W. Mccluskey, D.C. Albas 1Emotional content of speech

Obviously, it can be difficult to communicate between cultures. For

one thing, tonal aspects of speech (such as pitch, strength, length, and

pause) differ from language to language, and may create problems incross-cultural communication (Key 1975). With whites and Native

Indians, at least, it appears that individuals from one culture are more

adept at judging tonal affect from members of their own group (Albas

e t al. 1976). Possibly, through socialization and historical twists of fate,the Mexican culture, language, or whatever has become more “emo-tionalyy han ours. On the other hand, it may even be conceivable thatspeakers of English are more likely to express affect in the verbal than

in the vocal channel. In any case, for representatives of the two culturesto communicate effectively, it would be tremendously helpful for each

to “take the place of the other” and try to become more aware of how

the other reacts nonverbally and emotionally.It may pay researchers to try to get more actively and pragmatically

involved in improving intercultural communication through nonverbaltraining; for it has been shown that with practice, subjects can becomemore sensitive to nonverbal cues (Rosenthal e t al. 1979). Already, non-

verbal signals have been used effectively to help Spanish-speakingpeople learn to speak English, and vice versa. Triandis (1975), going yet

a step further in this regard, speaks of “cultural assimilators” - raining

devices designed to help individuals become more familiar with and

more sensitive to the attitudes, values, and communicative styles of

people from other cultures. Work along these lines - imed at improv-

ing inter-cultural communication - as a great deal of promise in termsof helping people from different cultures relate and communicate with

each other in more realistic and meaningful ways.

References

Albas, D.C., K.W. McCluskey and C.A. Albas, 19 76 . Perception of the emotional content of

speech: a comparison of two Canadian groups. Journal of CrossCultural Psychology 7,

481 -490.

Argyle, M., 197 2. The psychology of interpersonal behaviour. Baltimore: Penguin.

Argyle, M., 197 5. Bodily communication. London: Methuen.

Barcus, F.E., 19 59 . Comm unications content: analysis of the research, 1900-19 58. Ph.D.

dissertation, University of Illinois.



K . W. Mccluskey, D.C.Albas 1Emotional content of speech 131

Carney, T.F., 1972. Content analysis: a technique for systematic inference from communi-

Davitz, J.R. (ed.), 1964. The communication of emotional meaning. New York: McGraw-Hill.Davitz, J.R. and L.J. Davitz, 1959. The communication of feelings by content-free speech.

Journal of Communication 9.6-13.Dimitrovsky, L., 1964. The ability to identify the emotional meaning of vocal expressions at

successive age levels’. In: J.R. Davitz (ed.),The communication of emotional meaning. NewYork: McGraw-Hill.

Fenster, C.A., 1967. Vocal communication of emotional meaning among adults and children.Dissertation Abstracts In tematiod 28,1964-1965.

Gates, G.S., 1927. The role of the auditory element m the interpretation of emotions. psy-chological Bulletin 24, 175.

Godlstein, S.and L. Sies, 1974. The communication contract. Springfield, IL: Thomas.

Hall, E.T., 1959. The silent language. Garden City, NY: Doubleday.Key, N.R., 1975. Paralanguage and kinesics (nonverbal communication). Metuchen, NJ: The

Scarecrow Press.

Kirk, R.E., 1969. Experimental design: procedures for the behavioral sciences. Belmont, CA:Brooks/Cole.

Knight, G.P. and S. Kagan, 1977. Development of prosocial and competitive behaviors in

AngleAmerican and Mexican-American children. Child Development 48,1385-1394.Kramer, E., 1963. Judgement of personal characteristics and emotions from nonverbal proper-

ties of speech. Psychological Bulletin 60,408-420.Lewis, O., 1961. The children of Sanchez: autobiography of a Mexican family. New Yo&:

Random House.

McCluskey, K.W., 1974. Age differences in the perception of the emotional content of non-

fdtered and mtered vocal expressions. Master’s thesis, University of Manitoba.McCluskey, K.W. and D.C. Albas, 1978. Differential sensitivity to contradictory communica-

tion as a function of age and culture. Journal of CrossCultural Psychology 9,167-178.McCluskey, K.W., R.R. Niemi and D.C. Albas, 1978. Vocal communication of emotional mean-

ing among normal and disturbed children. Journal of Special Education 12,443-449.McCluskey, K.W., R.R. Niemi and C. Ferrer, 1973. Diferendas culturales en la percepcion de las

emociones. Informacion Nacional9,30-31. ,

McCluskey, K.W., D.C. mas,R.R. Niemu, C. Cuevas and C. Ferrer, 1975. Cross-cultural differ-ences in the perception of the emotional content of speech: a study of the development of

sensitivity m,Canadian and Mexican children. Developmental Psychology 11 ,551-555.Mehrabian, A., 1972. Nonverbal communication. Chicago: Aldine-Atherton.Pike, K.L., 1966. Etic and emic standpoints for the description of behavior’. In:A.G. Smith

(ed.), ommunication and culture:. readings in the codes of human interaction. New York:

Holt, Rinehart and Winston.

Rosenthal, R., J.A. Ha,M.R. DiMatteo, P.L. Rogers and D. Archer, 1979. Sensitivity to non-verbal communication: the PONS test. Baltimore: Johns Hopkins University Press.

Soskin, W.F. and P.E. Kauffman, 1961. Judgment of emotion in word-free voice samples. Jour-nal of Communication 11,73-80.

Starkweather, J.A., 1956. Content-free speech as a source of information about the speaker.Journal of Abnormal and SocialPsychology 52,394-402.

Triandis, H.C., 1975. ‘Culture training, cognitive complexity and interpersonal attitudes’. In:R.W. Brislin, S. Bochner and W J . Lonna (eds.), Cross-cultural perspectives on learning.

cations. Winnipeg: University of Manitoba Press.



132 K . W . McCluskey, D.C. Albas 1Em otional content of speech

Le but de la prksente recherche est d’dvaluer et de comparer les capacitds respectives chez des

sujets canadiens et mexicains h identifier les dm otion s vihic ulies par le parole. Des enseignantes

canadiennes et mexicaines, utilisant des mots qu’elles choisissaient dans leur propre langue, onttent6 de simuler quatre 6tats hmotifs, a savoir la joie, la tristesse, l’amour et la coltke. Apris

I’enregistrement initial, ces 6chantillons furent ffltrds dlectroniquement d i n d’en &miner le

contenu s imant ique , tout en laissant intactes les tonalitds d u discours. Ces expressions vocales

fd trie s fure nt ensuite dco uties par des sujets canadiens et mexicains igds de 5 . 9 , 1 3 , 1 7 , 2 5 , 4 5

e t 65 ans. Chez les groupes Bgds de 5 a 25 ans, a capacitd d’identifier I’imotion exprimie aug-

mente progressivement avec l’ige. Par co ntr e, cett e capacitd atteignit un plafond et dventuelle-

ment mEme diminua, & tel que les sujets Igds de 65 ans exdc utirent cet te t lch e avec moins de

pr6cision q ue les autres groupes adultes. Dans l’ensemble les sujets mexicains se sont riv61ds plus

sensibles i 1’6motion vdh iculk par la parole que les sujets canadiens, et ce d e facon significative.

De plus, tous les sujets, canadiens et mexicains, on t jug6 les ichantillons des “orateurs” mexi-

cains avec plus de precision que ceux des “orateurs” canadiens,



Copyright of International Journal of Psychology is the property of Routledge and its content may not be copied

or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.

However, users may print, download, or email articles for individual use.

perception of the emotional content of speech by canadian and mexican children, adolescents, and...

Documents