Paola Casadio
IRCCS S Lucia Rome
Via Isonzo 38, 00198 ROME (Italy)
Tel : 0039/03470812310
Fax : 0039/065032097

The observation of communication disorders in patients with traumatic head injury as a result of focal or widespread brain damage, has highlighted the presence of a series of phenomena which are difficult to classify within a unitary definition: they involve language in its many manifestations and at different levels: formation of thought, conceptual and linguistic organisation of discourse, appropriate interpretation and production as a function of the linguistic context. Since the aetiology of traumatic brain injury (TBI) is frequently a result of motorcycle accidents, this population is characterised by a preponderance of frontal and temporal lobe pathology. The impact of frontal lobe deficit on language skills must therefore be considered.

Pragmatic disturbances therefore refer to communicative behaviours that involve cognitive and social processes such as attention, planning, self monitoring, abstraction and the ability to regulate one’s behaviour. Communication deficits in patients with traumatic brain injury only infrequently involve language at the level of processors, as described in different types of aphasia.

In fact, unlike what is observed in aphasic syndromes, pragmatic disturbances of phonological, lexical and syntactic language processors involve a reduction in flexibility, efficacy and clarity. The linguistic deficit affects the quality of procedural and narrative discourse which as a result is fragmented, tangential, irrelevant, incoherent when not invented. These patients have difficulty introducing, selecting, maintaining or changing a topic; furthermore, during conversation, they have a hard time taking turns, providing the appropriate amount of information or using it. So, because of a combination of relatively spared structural verbal skills and decreased cognitive abilities, individuals who have sustained head injury often appear to have better communication skills than they functionally have in real-life situations.

Pragmatic communication lies outside the focus of conventional language tests, which typically assess structural aspects of language performance; therefore, there is a need for more wide-ranging assessment tools which take into account the context in which language occurs. An exploration of communication problems after closed head injury also requires an assessment of neuropsychological functioning since it is inextricably linked to more general social and cognitive impairment.

Many efforts have recently been applied to the development of alternative measures of communication behaviours, by taking into account social, neuropsychological and pragmatic principles. Researchers and clinicians have developed a wide range of tasks and methods of analysis but, to date, results are essentially descriptive and qualitative and few norms are available. This observation may be related to a lack of agreement as to the nature of this communicative disorder, given the complexity of several interrelated clinical symptoms, the variability in individual styles related to the patient’s age, his/her level of education and occupational status, the necessity to take into account the subject’s psychological condition and motivation. Snow et al (1995) state that, in the absence of norms, "..speech pathologists need to make quite arbitrary distinctions between "normal" and "abnormal", but because of their own subculture characteristics, may well be erring on the side of false positives (or negatives) more than is currently realised" (p. 368). On this subject, Body & Perkins (1998), using two sets of raters, one clinically trained in language pathology and the other drawn from peers of the TBI population, highlighted that clinicians are nevertheless sensitive to cognitive and linguistic skills across wider sociological groups, especially where poor performance is concerned.

Those tools used to assess pragmatic communication are generally grounded in a social interaction model of communication which involves the evaluation of contextual aspects in the form of natural conversation and role-plays. Check lists are used to record the presence/absence of specific pragmatic behaviours and rating scales also take into account the efficacy and degree of precision of messages from a qualitative point of view. We will only mention the most well-known tools: e.g. the Profile of Communicative Appropriateness (PCA) (Penn, 1983), the Pragmatic Protocol (Prutting & Kirchner, 1983), and the Rating Scale for Communication Deficits (Erlich & Barry, 1989).

These protocols analyse some communication parameters such as:

• nonverbal aspects (prosody, mime and gestures),
• topic selection and maintenance,
• taking turns in terms of resumption/revision,
• pause time,
• feedback to speaker and quantity.

More recently, McDonald & Pearce (1995) proposed the "dice game". In this test, the patient is required to explain how to play a game, formulating a procedural semi-structured discourse whose content is analysed, in order to evaluate whether deficits in planning, monitoring and impulse control are disrupting the ability to communicate effectively.

In clinical practice, a diagnosis of pragmatic deficit is frequently made without distinguishing between disturbances in "comprehension" and " production " skills. Furthermore it is unclear whether a comprehension deficit, deriving from a linguistic deficit and/or a cognitive deficit which is difficult to ascertain, could have an impact on production capabilities.

When we talk about "pragmatic comprehension deficit" in TBI, we refer to the difficulty which patients demonstrate in analysing the interaction of extralinguistic, paralinguistic and linguistic contexts in order to interpret the real meaning of the speaker’s utterances.

The extralinguistic context includes external elements such as: different settings which influence both conversational style and lexical choice; other participants who contribute their knowledge and their beliefs regarding a topic and who influence the situation through their respective roles (e.g. mother or son, doctor or patient); movements that include postural stances and facial gesturing. The internal context involves the knowledge of the world and emotional state of the individual speaker or listener.

The second category, the paralinguistic context, involves the prosodic variations that convey semantic and syntactic information, as well as the emotional state.

The interaction between extralinguistic and linguistic processes conveys the speaker’s meaning. It has been differentiated from utterance meaning as a way of characterising the common difference between what a speaker means, with reference to the non-literal meaning, and what a speaker says, that is the literal sense.

In clinical practice, within individual sessions, we observed that our patients were unable to process those aspects of language which were more abstract and symbolic and their meaning within specific contexts: subjects were trapped into the literal sense of discourse, the concrete interpretation of situations, and they seemed unable to grasp an alternative meaning even when it was suggested by the therapist. We therefore decided to conduct a series of group therapy sessions to observe and train the comprehension abilities of our patients by presenting them with several film sequences.

Participants were four consecutive chronic survivors of severe closed head injury (M/F: 3/1), average age 27 years (range 22-33), average educational level 13.5 years (range13-15), attending the Physical and Rehabilitation Program at the Santa Lucia Hospital in Rome. These TBI individuals had suffered a severe head trauma at least 6 months earlier (the Glasgow Coma Scale at Hospital admission was < 8). All patients were out of the post-traumatic amnesia (PTA) period. We excluded those patients with a coma duration of less than 3 days and those who were unable to follow the procedure due to very severe attentional or memory deficits, or aphasic problems. Personal and clinical data as well as TBI patients’ performance scores on some neuropsychological tests are summarised in Table 1.

Since it was our first attempt to observe our patients’ comprehension abilities, the group therapy protocol was developed according to clinical parameters rather than research criteria, although we tried to respect the most basic research principles.

The order of presentation of the 16 film sequences, which followed a graduation in difficulty of interpretation, was established after we showed them to ten control subjects selected from our speech pathology students and patients’ relatives or friends, so that we could match TBI patients in terms of age and educational status. We explained to control subjects what the communicative pragmatic features were and recommended that they focus their attention while viewing the films exclusively on these features. The rates were given a simple scale ranging from 1 to 5, in which each description was matched with a score:

1= very easy to understand

2= easy to understand

3= some parts are easy and others are difficult

4= difficult to understand

5= very difficult to understand

A score of 1 was given, for example, to a sequence with blatant and redundant behaviour. In contrast, a maximum score of 5 was given to a sequence which presented scenes where metalinguistic translation skills were necessary.

The films mainly belonged to the Italian culture, though we did include a few French, Spanish and American films, as we are accustomed to seeing films from these countries. Some of the sequences selected broke pragmatic rules at different linguistic, paralinguistic and extralinguistic levels as defined previously, or they did not respect Grice’s model of conversational practice ( 1978) in which speakers cooperate and expect cooperation when conversing with each other (Fig. 1) or, finally, they did not conform to social behaviours such as good manners and politeness.

Film sequences were used to facilitate patients’ understanding: first by showing them situations where verbalisation was emphasised by the professional use of mime, gestures and prosody; second by using the videotape recorder, which gave patients the opportunity to review film segments that were unclear to them.

At the beginning of each session, a response form was given to the participants with questions to guide their observation of the film segment. That was:

Express your opinion regarding:

what the actors say;

the gestures they use;

the way they behave.

Each question was thoroughly explained, so the patients knew exactly what to focus on, then the sequence was introduced by a brief explanation of the situational context. The short film was shown twice to facilitate patients’ attention and recall and help them answer the questions in writing. At the end of this first sequence, the participants read their first answer which was in turn discussed by the group on the basis of each member’s opinion. The same process was followed for the other questions. The role of the therapist conducting the group was to synthesize the group’s comments whether divergent or convergent opinions were expressed, writing a summary on a blackboard. At the end of the discussion, the therapist showed the film sequence again, making her own comments on crucial passages and comparing them with the opinions written by the patients. Before finishing the session, there was a final phase of comparison with what the therapist expressed. Each session lasted approximately an hour and a half. The meetings were held twice a week for two months.

We chose a group format because clinical experience in the field of neuropsychological rehabilitation of brain injured patients has demonstrated the validity of this group exercise method which induces a more naturalistic style of communication and is also more similar to everyday life or social experiences. Structural group treatment contains certain features which make it possible for patients to have a more realistic view of their deficits that are often minimised or misattributed to others’ behaviours. The objectives of therapy are to encourage patients:

- to check and to correct their own behaviour in a social situation, accepting positive and negative feed-back from the group, and overcoming personal resistances which can occur in the one-to-one patient-therapist approach;

- to follow the examples of others who demonstrate how to perform certain specific therapeutic tasks;

- to act in ways explicitly prescribed by the therapist so as to modify unacceptable or maladaptive behaviours;

- to train the ability to judge other persons’ verbal and non-verbal communication;

- to assimilate and personalize the prescribed behaviours;

- to evaluate one’s own progress as well as the remaining unresolved problems.

In this way, role diversification is promoted, individual resources can be integrated, relationships are established, the common focus being the acquisition of more independence and efficacy by becoming aware of one’s ability to adjust, thus enhancing acceptance of the consequences of the traumatic event, and achieving realistic ways of improving functional competencies.

Results

We analysed the different answers given by the patients in terms of their ability to identify errors in actors’ discourse, gestures and behaviours and to make explicit the pragmatic rules that had not been followed in the situations presented. We scored separately the answers regarding "identification" and "judgement" according to the following scale:

0 = absent or wrong

0.5 = fixation on details

1 = correct.

At the beginning of treatment, group members only described the content of sequences without realising that something was wrong. Even the therapist’s critical comments were not really understood: group members accepted the therapist’s opinion without any criticism and without providing any judgement, as was the case during their previous performance. In the third week, subjects started to show individual differences in their performances. Thus, at the end of treatment, we could identify some processes and draw individualised profiles (Table 2):

• patient n° 4 continued to describe situations without grasping inadequacies in the different communicative levels.
• patient n° 1 initially succeeded in perceiving some details but would lose sight of the global situation and could not propose adequate behaviours; eventually, he was able to identify broken rules but could not provide an explanation for his opinion.
• patient n° 3 could not capture the meaning of wrong social behaviours for which he persisted in giving concrete interpretations, although he could detect verbal and gestural anomalies
• only patient n° 2 completely modified his performance: he was able to identify errors and to explain those behaviours which went against the different pragmatic rules, finally proposing a correct model.

The major source of disturbance observed in our patients seemed to lie in their inability to integrate information which derives from global communicative behaviours shown in the film sequences and then to produce inferential links in order to formulate a correct judgement on these sequences. This difficulty seems to originate in their inability to inhibit concrete interpretation and therefore to analyse the same information from another point of view; the problem is also related to their inability to take into account the different contents at a higher conceptual level, through abstract reasoning and the formulation of a judgement. Since these functions are typically attributed to the frontal lobes, it is likely that the particular types of errors which characterised the performance of our patients were linked to those frontal lobe deficits exhibited on neuropsychological assessment. McDonald and Van Sommers (1993), observing TBI patients, showed the significant influence of frontal lobe damage on the management of pragmatic tasks. The authors stressed the importance of problem solving abilities and of the integrity of executive functions for pragmatic skills to function adequately. Other studies, still focusing on brain-damaged subjects, indicated that a correct interpretation of the non-literal sense and of indirect requests may involve the right hemisphere to some extent. This fact could explain the performance of patient n° 4, with a double left-right lesion, who was not able to modify her literal interpretations and continued failing in her judgement of the adequacy of behaviours.

In general, even though a model of pragmatic competency is not yet available, these disturbances observed in TBI patients seem to be influenced by several factors which are not exclusively linguistic, particularly with regard to the patients’ reduced ability to monitor and integrate those processes which are needed to link language and communicational context. Further effort is needed to define a pragmatic framework within which each component and its specific role can be taken into account. It will then be possible to develop an appropriate rehabilitation protocol which will improve pragmatic communication and support the social integration of TBI patients.

Avanzi, S., Mazzucchi, A. Disturbi post-traumatici del linguaggio. In A. Mazzucchi (ed.), La Riabilitazione Neuropsicologica dei Traumatizzati Cranici. Milano: Masson, 1997.

Body, R. and Perkins, M.R. Ecological validity in assessment of discourse in traumatic brain injury: ratings by clinicians and non-clinicians. Brain Injury, vol.12,11, 963-976, 1998.

Grice, H.P. Further notes on logic and conversation. In P. Cole (ed.), Syntax and semantics: Pragmatics. New York: Academic Press, vol.9, 1978.

McDonald, S. Communication disorders following closed head injury: new approaches to assessment and rehabilitation. Brain Injury, 6, 283-292, 1992a.

McDonald, S. Differential pragmatic language loss following severe closed head injury: inability to comprehend conversational implicature. Applied Psycholinguistics, 13,295-312, 1992b.

McDonald, S. Pragmatic language skills after closed head injury: Ability to meet the informational needs of the listener. Brain and Language, 44, 28-46, 1993.

McDonald, S. & Pearce, S. The "dice" game: a new test of pragmatic language skills after closed head injury. Brain Injury,3,255-271, 1995.

Penn, C. & Cleary, J. Compensatory strategies in the language of closed head injured patients. Brain Injury, 2, 3-18, 1988.

Prigatano, G. Neuropsychological rehabilitation after brain injury. Baltimore: Johns Hopkins University Press, 1986.

Prutting, C.A. & Kirchner, D.M. Applied pragmatics. In Gallagher T. & Prutting C. (eds.), Pragmatic assessment and intervention issues in language. San Diego, College Hill Press, pp. 26-68, 1983.

Prutting, C.A. & Kirchner, D.M. A clinical appraisal of the pragmatic aspects of language. Journal of Speech and Hearing Disorders, 52, 105-119, 1987.

Sarno, M.T. Head injury: language and speech defects. Scandinavian Journal of Rehabilitation. (Med. Supply.), 17, 55-64, 1988.

Sgaramella, T.M., Zettin, M. & Bisiacchi, P. Disturbi del linguaggio nei traumi cranici: comprensione e uso degli aspetti pragmatici. In Zettin, M.& Rago, R. (eds.), Trauma Cranico. Conseguenze neuropsicologiche e comportamentali. Torino, Bollati Boringhieri, pp. 108-119, 1995.

Snow, P., Douglas, J. & Ponsford, J. Discourse assessment following traumatic brain injury: a pilot study examining some demographic and methodological issues. Aphasiology,9, 365-380,1995.

Sohlberg, M.M. & Mateer, C.A. The remediation of language impairments associated with head-injury. In Sohlberg, M.M. & Mateer, C.A. (eds.), Introduction to Cognitive Rehabilitation:Theory and Practice. New York, The Guilford Press, pp. 212-231, 1989/b.

Sohlberg, M.M. & Mateer, C.A. Evaluation and Treatment of Communicative Skills. In Kreutzer J.S. & Wehman, P. (eds.), Community-integration following Traumatic Brain Injury. Paul H. Brooke Publishing Co., pp. 67-83, 1990.

Sohlberg, M.M., Perlewitz P.G., Johansen A., Schultz J., Johnson L. & Hartry A. Improving pragmatic skills in persons with head injury. Tucson, Communication Skill Builders Inc, 1992/c.