Narrative Analysis is a systematic procedure for the analysis of recording the naturally occurring talk produced in everyday human interaction. "Narrative" is sequence and consequence: events are selected, organised, connected, and evaluated as meaningful for a particular audience. Storytellers interpret the world and experience in it; they sometimes create moral tales - how the world should be. Narratives represent storied ways of knowing and communicating (Hinchman & Hinchman, 1997). Westby (1989) stated that narrative skills forms the bridge between oral language and literacy by providing examples of the extended, decontextualized, cohesive discourse units that a child will encounter in written texts. In the field of Speech Language Pathology, narratives have served distinctive evaluative functions. A narrative is some kind of retelling, often in words, of something that happened (e.g. a story). Gloria (2009) studied the narrative skills in the age group of 5-6 year old typically developing children and concluded that children had more phonological errors and they used reduction and unrelated repair strategy. Studies show that the normal development of phonology and morpho-syntax continues by the age of 6 years.
Aging is a natural phenomenon and is inevitable. As an individual starts aging, there is a gradual wear and tear of one's body systems. Language and communication is no exception to this. The linguistic and communication patterns of the old are quite different from that of the younger groups. Geriatric individuals need to put in a lot of effort to get their message across to others. As we age, our ability to communicate effectively will get influence by some normal changes in speech, language, swallowing, and hearing. (Busacco, 1999). Also there is a great deal of variation in the language skills of older adults. The differences are most likely due to the individual's life history, language competence, communication environment, and cognitive abilities. Some age-related changes in language are due to normal cognitive declines that occur as one grows older, including a decrease in attention, decline in memory, and slower speed of processing information. A large percentage of older adults at one time or another will experience problems with word retrieval. They may have difficulty remembering names and may have problems retrieving well-known words. Often, this can be a frustrating communication experience, particularly when an older adult is having increased difficulty finding words or is having problems selecting the correct word for a familiar object (Busacco, 1999). In general, older adults tend to use simpler sentence structure than younger adults. They tend to use more fragmented sentences in conversation, especially as age increases. Aging seems to have little or no effect on the manner in which older adults engage in conversation. For the most part, older adults tend to be adept at conversational turn taking, maintaining the topic, and modifying the content of their message depending on the listener's needs. The majority of older adults maintain basic conversational skills well into the 8th or 9th decade of life unless a pathological condition is present. (Busacco, 1999). So it is important to know how narrative skills will differ with different age groups and different clinical population.
Studies focusing on comparison of narrative skills between traumatic brain injured patients and control groups have revealed that, there was a significant difference between participants with and without TBI for all measures in the monologic narrative. In the jointly-produced narrative, there was no significant difference in performance and participation between individuals with TBI and control participants. Participants with TBI demonstrated a significant improvement between the monologic and the jointly-produced task in story grammar and informational content. Wetherell, Botting & Ramsden (2007) did a study on narratives in adolescent subjects with specific language impairment (SLI), they compared the 99 typically developing adolescents and 19 peers with specific language impairment. The subjects were given two different types of narrative task: a storyâ€telling condition and a conversational condition. Based on the assessment of the four areas of narrative (productivity, syntactic complexity, syntactic errors and performance) they concluded that the group with specific language impairment was poorer on most aspects of narrative skills. A number of measures also showed interactions between group and genre, with storyâ€telling proving to be a disproportionately more difficult task for the specific language impairment group. Error analysis also suggested that the specific language impairment group was making qualitatively different errors to the typically developing group, even within a genre. Labov (1972) described narration as a form of discourse that allows the speaker to give interpretative meaning to a sequence of events. Studies have shown that the internal narrative skills will be deteriorating as age increases, but there will be no reduction of the interpretative capacity of one's life story. Also one important factor he observed that older adults used their experiences to appreciate events and describe emotional states than younger adults. The reason may be due to the fact that older adults had much more experience than younger ones. In a study of linguistic features and discourse organisation in 52 functionally illiterate French men and women, Eme, Lacroix & Almeciia (2010) reported that the subjects had great difficulty handling morphosyntactic rules, referential cohesions and the narrative schema. The authors concluded that individuals who have not succeeded in learning to read also have impaired oral language abilities.
Mathias (2008) used conversational analysis to study language characteristics between normal geriatrics and middle aged adults. Srinivasacharya (2008) used verbal narratives to describe literacy use of modern Telugu novels and the creative use to which the disorders are put in the imaginative field of fiction. Heble & Karanth (1985) used conversational analysis to examine the pragmatic strategies used by aphasics. John, Veena, George & Rajashekhar (2008) did a study on comparison of narrative and procedural discourse in normal young adults and elderly subjects. They also studied the influence of age, gender and education in the Malayalam speakers. The results indicated that the discourse skills alter with the advancing age in all the parameters taken. There was an increase in the number of words, mean length of utterance (MLU), speaking rate and reduction in the number of the sentences produced in elderly as compared to young adults.
NEED FOR THE STUDY
Since the average life span of human being increase, it is important to know the narrative skills of older individual and it is also taken into consideration when we assess the older individual. There are not many studies reported in Malayalam, so the present study investigates the narrative skills of older individuals speaking Malayalam.
AIM
To describe the narrative skills in geriatrics.
To describe the narrative skills in middle aged adults.
To compare these subjects with normal middle aged adults.
REVIEW OF LITERATURE
It is well known that communication is a part of our everyday life. Effective communication influenced not only by aging but also due to changes in speech. One of the most common problems in old age is tooth loss and compromised dentition. Aging also causes tissue, glandular, and muscular changes in the jaw, tongue, salivary glands, and throat. There is a decrease in the number of salivary glands and reduced taste sensation (Sonies, 1987). Although both genders age, the change in vocal pitch differ with gender while the vocal pitch of males increases, that of females decreases. There is also an increase in jittering of the voice with age because of a decrease in neuromuscular control of the muscles that support the larynx (Carus & Mueller, 1997). There is a great deal of variation in the language skills of older adults. The differences are most likely due to the individual's life history, language competence, communication environment, and cognitive abilities. Some age-related changes in language are due to normal cognitive declines that occur as one grows older, including a decrease in attention, decline in memory, and slower speed of processing information. A large percentage of older adults at one time or the other will experience problems with word retrieval. They may have difficulty remembering names and may have problems retrieving well-known words. Often, this can be a frustrating communication experience. If an older adult is having increased difficulty finding words or is having problems selecting the correct word for a familiar object (ASHA, 1999). In general, older adults tend to use simpler sentence structure than younger adults. They tend to use more fragmented sentences in conversation, especially as age increases. Aging seems to have little or no effect on the manner in which older adults engage in conversation. For the most part, older adults tend to be adept at conversational turn taking, maintaining the topic, and modifying the content of their message depending on the listener's needs. The majority of older adults maintain basic conversational skills well into the 8th or 9th decade of life unless a pathological condition is present. (Busacco, 1999).
Narrative Analysis
A survey of literature on scientific studies of narrative analysis showed that there are not many published work. Narrative Analysis is a systematic procedure for the analysis of recording the naturally occurring talk produced in everyday human interaction. "Narrative" is sequence and consequence: events are selected, organised, connected, and evaluated as meaningful for a particular audience. Storytellers interpret the world and experience in it; they sometimes create moral tales - how the world should be. Narratives represent storied ways of knowing and communicating (Hinchman &Hinchman, 1997). Jorgenson &Togher (2009) compared the narrative skills between traumatic brain injured patients and control groups and they found out that, there was a significant difference between participants with and without TBI for all measures in the monologic narrative. In the jointly-produced narrative, there was no significant difference in performance and participation between individuals with TBI and control participants. Participants with TBI demonstrated a significant improvement between the monologic and the jointly-produced task in story grammar and informational content. In a study of narrative skills in William syndrome and its neuropsychological correlates, done by Marini, Martellj, Gagliardj, Fabbro & Borgatti (2009) and they compared the narrative skills of WS patient with typically developing children's and found out WS participants showed visual-spatial deficits but scored within the normal range, according to their mental age, in the linguistic assessment. For the narrative task, they showed good phonological, lexical and syntactic skills, but their story descriptions were less effective than those produced by the TD group on measures assessing global coherence and lexical informativeness, showing dissociation between macro and micro linguistic abilities.
Randall (1999) suggested that the "internal narrative" seems to continue expanding in older age, and that there was no reduction of the interpretative capacity of one's life story. In fact, older adults appeared to be interested in making sense of life experiences, expressing a subjective appreciation of events and describing emotional states, more than younger adults seem to do. This may be due to the greater amount of personal experience and to the way that older adults might represent themselves as communicative actors in context. In a study of linguistic features and discourse organisation in 52 functionally illiterate French men and women, Eme, Lacroix & Almeciia (2010) reported that the subjects had great difficulty handling morphosyntactic rules, referential cohesions and the narrative schema. The authors concluded that individuals who have not succeeded in learning to read also have impaired oral language abilities.
In another study by Heilmann, Miller, Nockerts & Dunaway (2010) on properties of the narrative scoring scheme using narrative retells in young school-age children, the authors concluded that the NSS was significantly correlated with age and each of the microstructural measures. Murray (2010) tried narrative analysis to distinguish clinical depression from early Alzheimer's disease in elderly people, and they found out significant group differences on the informativeness discourse measures. AD participants producing less-informative samples than DEP and control participants. DEP and control groups did not significantly differ on any discourse variable.
Conversational Analysis
Conversational Analysis (CA) is a systematic procedure for the analysis of recorded, naturally occurring talk produced in everyday human interaction. The principal aim is to discover how participants understood and responded to one another in their turns at talk and how such turns are organized into sequences of interaction. CA approach makes use of recording the naturally occurring conversations, which would have taken place even if they had not been recorded. CA occurs in real contexts and is a bottom up, data-driven approach. It aims to describe and explain how the participants displayed their interpretations of each other's talk. Yim & Bs 2006 did a study on conversation analysis of communication between patients with dementia and their professional nurses and found that among a total of 532 episodes, 440(82.7%) were identified as nurse-involved episodes. In addition, 66 of the 440 episodes were selected based on the significance of the conversation. The communicative problems between nurses and patients in terms of expressions were identified as "directive and authoritative expressions", "emotional and competitive expressions", "evasive and on-looking expressions", and "excessive use of title only", such as calling them granny or grandpa without proper names. In terms of content and relationships, "lack of themes in psychosocial areas" and "nurse-led relations" was identified respectively as communicative problems. CA can be used as an assessment tool in people with aphasia. Wilkinson (2006) applied conversation analysis to aphasic talk and found out the CA has been made by researchers drawing on conversation analytic findings into the structure of aspects of ordinary, non-aphasic, talk such as repair organization and turn organization, and indicates some of the ways in which this approach to aphasia has been used within intervention studies and everyday professional practice. Beeke, Maxim & Wilkinson (2007) in a study using Conversation Analysis to Assess and treat people with Aphasia brought about the implications of using CA as a tool for assessment and treatment in aphasia. Studies by Rousseaux, Verigneaux & Kozlowiski (2010) on the analysis of communication in conversation after severe traumatic brain injury and revealed the patients with TBI were impaired in their participation to communication, especially in greeting behaviour. Verbal communication was mostly affected by difficulties in producing fluent and intelligible language and using pragmatics (responding to open questions, presenting new information and introducing new themes, organizing discourse and adapting to interlocutor knowledge). Non-verbal communication was impaired by difficulties in using pragmatics (mostly adapted prosody). Participation and verbal communication correlated with the executive functions, language and behavioural assessment.
Mathias (2008) used conversational analysis to study language characteristics between normal geriatrics and middle aged adults, and she stated that the geriatrics had increased trouble sources and decreased resolution strategies. Srinivasacharya, (2008) used verbal narratives to describe literacy use of modern Telugu novels and the creative use to which the disorders are put in the imaginative but in the imaginary field of fiction.
James, Burke, Austin & Hulme (1998) compared the performance of older and younger adults in the production of personal narrative and picture description. Their results demonstrated age related increase in "off-topic verbosity" (OTV) levels in personal narrative, which is a discourse gender that emphasizes the exposition of life events. In turn, picture descriptions, which provided more objective communication situations, presented lower OTV scores. In addition, older adults' narratives were rated by younger and older adults as more interesting, informative, and simply as better stories. Although older adults produced more OTV, their narratives appeared to have greater communicative value than that of younger adults, suggesting that there was not a decrement in the quality of older adults' discourse. Korolija, (2000) found that conversations amongst older adults presented different coherence-inducing strategies.
A comparative study done by John, Veena, George & Rajashekhar (2008) of narrative and procedural discourse in normal young adults and elderly subjects, together with of the influence of age, gender and education in the Malayalam speakers provided results which that the discourse skills alter with the advancing age in all the parameters taken. There was an increase in the number of words, mean length of utterance (MLU). Speaking rate and reduction in the number of the sentences produced in elderly as compared to young adults.
Wetherell, Botting & Ramsden (2007) in a study on narratives in adolescent subjects with specific language impairment (SLI), compared the 99 typically developing adolescents and 19 peers with specific language impairment. The subjects were given two different types of narrative task: one, a storyâ€telling condition and the other a conversational condition. Four areas of narrative (productivity, syntactic complexity, syntactic errors and performance) were assessed. They concluded that the group with specific language impairment was poorer on most aspects of narrative skills. A number of measures also showed interactions between group and genre, with storyâ€telling proving to be a disproportionately more difficult task for the specific language impairment group. Error analysis also suggested that the specific language impairment group was making qualitatively different errors to the typically developing group, even within a genre. In a study on pragmatic skills in the early stages of Alzheimer's disease, Feyereisen, Berrewaerts & Hupet (2007) concluded that persons with dementia of Alzheimer's type (DAT) produced a larger number of words than control participants and they benefited from the task repetition. However, they were less able to take into account previously shared information, used no definite referential expressions and were more idiosyncratic in their descriptions of the referent. This decline of communicative effectiveness was found not to relate closely to executive deficits.
METHODOLOGY
Participants
A group of 8 healthy geriatric persons aged 60-70 years consisting of 4 males and 4 females and a group of 8 healthy middle aged adults (40-55 years) matched for age, gender, socio-economic status, education and linguistic background participated in the study. The participants had 15-18 years of education and used Malayalam as their language.
Inclusion criteria
Subjects did not have any neurological impairment.
PROCEDURE
Data collection
Narrative sample of 5-10 minutes was recorded for each subject regarding their past events of both familiar & unfamiliar contexts. Familiar contexts involved story narration & past experiences of life (eg: About marriage, any traditional festivals etc.) Non-familiar contexts involved asking subjects to imagine the specified situation and narrate with respect to different characters/personality (eg: Clinician would ask 'what developments will you do if you become a Chief Minister?). A quiet sound treated room was selected for recording purpose. The subjects were seated comfortably on the chair at a distance of 1 feet from the laptop placed on the table. Each client's speech was recorded individually using a standard laptop computer with inbuilt microphone with the help of the PRAAT voice recording and analysis software 5.1 Version. (Boersma & Weenick, 2009). Sampling rate was 44100 Hz and quantization level set at 16 bits.
Data coding and analysis
Obtained data of 5 minutes (hundred utterances were selected) were transcribed and analyzed to study types of trouble sources such as phonological, morphological-syntactic, semantic, discourse and also repair strategies such as repetition, unrelated, elaboration, reduction and substitution along with the complexity and success of resolution like most successful, successful and unsuccessful and type token ratio using the systematic procedure [Orange, Lubinski & Higginbotham (1996)].
Analysis of trouble sources, repairs and resolutions
According to the guidelines by Orange, Lubinski & Higginbotham (1996), the data was analysed to study the trouble sources, repairs, along with complexity and success of resolutions. Trouble sources (TS) were divided into phonological, morphological-syntactic, semantic, discourse and other trouble sources. The categories for describing repair types were repetition, elaboration, reduction, substitution and unrelated. Repair resolutions were analysed as most successful, successful and unsuccessful. Repair complexity was coded as simple or complex and the repair sequences were analysed based on self initiated self repairs (SI-SR) and SI-incomplete utterances (SI-INCO).
Analysis of type token ratio
The data was analysed in terms of Type Token Ratio (TTR). Based on the classification given by Yule (2002), open class words (content words) like nouns, verbs and adjectives and closed class words (functional words) like conjunctions, prepositions, articles and pronouns were used.
According to Wren, Martin & Rao (2008), once the closed and open class words were identified the total number of words; total number of different words and type token ratio of each category were calculated using the ratio:
TTR= Total number of different words
Total number of words.
The data was then treated with statistical analysis using Man Whitney U test.
RESULTS
The obtained audio recordings are transcribed into native Malayalam and hundred successive utterances were selected for analysis. The analysis was carried out using the guidelines provided by Orange, Lubinski & Higginbotham (1996). The major issues considered in analysing severe measurement of types of trouble sources, repairs, resolution and their complexity for both familiar and unfamiliar tasks. As per the guidelines of Wren, Martin & Rao (2008), only the words were taken for analysis and they were divided into seven categories nouns, verbs, adjectives, conjunctions, prepositions, articles and pronouns. Other measure considered was TTR which was solved for each category.
Familiar and Unfamiliar tasks
Comparison of trouble sources
Familiar tasks
Geriatrics had higher (48) trouble sources than adults (40), geriatric population showed higher morpho-syntactic task (32) than adults (25).
Unfamiliar tasks
Geriatrics had higher (47) trouble sources than adults (33), again geriatric population showed higher morpho-syntactic task (31) than adults (21).
Type of repair strategy
Familiar tasks
Total number of repair strategies was equal in both adults and geriatrics (24), in elaboration part geriatrics showed higher number (20) than adults (15).
Unfamiliar tasks
Geriatrics showed higher repair strategies (17) than adults (15), geriatrics showed more elaboration and repetitions (9) than adults (5).
Type of trouble source repair sequences
Familiar tasks
Geriatrics had higher (16) repair sequences than adults (15).
Unfamiliar tasks
The total number in repair sequences was more in geriatrics (19) than adults (11)
Types of resolutions
Familiar tasks
Geriatrics showed higher resolutions (17) than adults (16).
Unfamiliar tasks
Geriatrics showed higher resolution (19) than adults (11).
The overall results showed that geriatrics populations have more trouble sources, repair sequences and resolutions except in repair strategies. The familiar tasks in repair strategies showed that both adults and geriatrics have the same score.
Type token ratio
The comparison of adult and geriatrics using type token ratio revealed that there has significant difference in the unfamiliar task of both open (p=.008 sig) and close (p=.008 sig) words in adults.
GRP
AGE
N
Mean
Std. Deviation
T
Familiar
OPEN
Geriatric
8
.7825
.06182
.56700
Adults
8
.7637
.07029
p=.58 ns
CLOSE
Geriatric
8
.2175
.06182
.56700
Adults
8
.2363
.07029
p=.58 ns
Unfamiliar
OPEN
Geriatric
8
.7425
.03454
3.10800
Adults
8
.6863
.03777
p=.008 sig
CLOSE
Geriatric
8
.2575
.03454
3.10800
Adults
8
.3137
.03777
p=.008 sig
Statistical analysis
Mann - Whitney U test was done to obtain significance values across all the parameters. The result showed that in trouble sources, repair strategy, repair sequences and type of resolutions in familiar topics has no significant difference between adults and geriatrics, while the trouble sources in unfamiliar task has significant difference in morpho-syntactic tasks (p=0.046 ). In adults while comparing familiar and unfamiliar tasks there is no significant difference in the above mentioned errors with respect to narrative skills. But in geriatric populations there is some significant difference in unfamiliar topics with respect to repair strategies (elaboration, p=.026) and repair sequences (SI-INCO, p=.246).
DISCUSSION
The result of the present study indicates that the performance of geriatrics and middle aged adult population can be distinguished from one another using narrative skill analysis. The performance of geriatrics population on the trouble sources and repair strategies is similar to the results reported by Mathias (2008). In this study, observations in younger subjects show that they have few errors. Another study by Hegde, Shruthy & Subba Rao (2010) shows that the number of trouble sources and the repair strategies used were higher in young adults than the middle aged adults and geriatrics which was contradictory to this present study. This may be because of the developmental changes in the usage of language.
Since there are only few studies on narrative skill analysis it is difficult to validate the result in more stringent fashion. Overall, geriatrics has increased trouble sources and resolution strategies. It is important to notice that such differences are seen in unfamiliar topic narration than familiar topic narration. This probably indicates a decline in their cognitive abilities and memory. Overall the technique of narrative analysis provides detailed description which can be clinically used. The values reported in this study are out of hundred consecutive utterances or sentences and hence further studies can be carried out in different language and cultural groups to understand the phenomenon of narration of language.
CONCLUSION
In conclusion, narrative analysis is a good tool which can be used for both assessment and treatment of language disordered population, irrespective of age groups. In a diverse culture ranging in differences in languages, education levels, place, geographic location, socio economic status etc, there is a need to understand the narrative skills of the normal elderly in the Indian settings and thus form a normative database for the language disordered population. It gives us an idea of individual's communication skills in a natural environment. It also gives information about the parameters of language. Narrative analysis can also be used among bilingual communities, as it helps to know the deficits across languages. Future research is required to study the narrative aspects across various groups of adults and geriatrics and can be compared with the different Indian language disordered group.
