Activity-based communication analysis – focusing on context in communication partner training

ABSTRACT

Background: There is an increased awareness of contextual influence on communication involving persons with aphasia (PWA) and there is a need for frameworks addressing how to analyse contextual factors and address them in, for example, communication partner training (CPT). Activity-based communication analysis (ACA) is a theoretically based method providing a vocabulary and a structure for analysing the influence of context in interaction and how it relates to communication.

Aims: The purpose of this paper is to present the framework of ACA as a means of stimulating further enrichment of CPT practices.

Methods and Procedures: The paper provides a brief background to the ACA approach, including examples, as well as a checklist and a model for analysis and assessment of intervention addressing communication involving PWA.

Outcomes and Results: This approach enables an explicit account of different contextual factors and how they relate to communication and may, thus, guide the design of CPT for particular participants and motivate behavioural change.

Conclusions: ACA is a framework that addresses the influence of context on communicative interaction and provides a way of including contextual factors in a systematic way when analysing communication involving PWA. In this way, it can contribute to a richer analysis and intervention and is clearly applicable for CPT.

KEYWORDS:

• Aphasia
• communication
• context
• social activity
• communication partner training

In this paper, we discuss how context, physical, social, and psychological can be systematically considered in communication partner training (CPT) to enhance awareness of what influences communication and how it can be supported in various social activities.

Taking more of “the whole picture” into consideration in studying and supporting communication involving persons with aphasia (PWA) can be addressed in different ways and by different frameworks, which can all be used for enriching CPT. One way is to use a social model approach, in the sense where “social” involves a birds-eye view of the life situation of the PWA and his/her communication partners (cf. Simmons-Mackie, 2000) and aim to make this situation optimal for communication, given the circumstances. Another way is to use a detailed sequential analytic perspective, analysing specific interactions in detail in order to identify communication patterns that can be improved for the participants and assist in making changes, using Conversation Analysis (CA) (e.g. Wilkinson, 2014).

Activity-based communication analysis (ACA) is an approach that focuses on the context of communication in different social activities (Allwood, 2013). It can, by doing this, provide a link between a social model approach to the life situation and a qualitative (and quantitative) analysis of communicative behaviour in contexts and social activities. This approach is presented below.

We will start with an example illustrating communication in a social activity, where ACA could be used in CPT. Then follows a presentation of the model of analysis of ACA, including influencing background factors as well as influenced behaviour in a specific social activity. ACA is briefly compared to some other, related ways of analysing communication in context, which may be used in CPT. Further examples are given and elaborated in relation to CPT. We argue that ACA provides a useful framework for identifying barriers and facilitators for communication, related to particular social activities and to identify activities to target in CPT.

Let us first consider a specific communication context in Example 1 (see also Figure 1).

Figure 1. Nurse asking husband what food his wife likes.

Example 1

The husband of a woman with aphasia comes to visit his wife at the hospital. Why does a nurse ask the husband what his wife likes to eat, although his wife is present?

In this situation, the PWA is sitting in her bed and the husband is standing on one side of the bed while the nurse is standing on the right side of the bed, facing the husband and having eye contact with him. The nurse starts to ask the husband what his wife likes to eat. The wife listens but says nothing. If we analyse this activity, we can identify some critical background factors and see how they can affect communication. The physical background factors of where the participants stand, sit, lie or move around in the room play an important role, since the nurse and the husband are facing each other, while the wife is sitting on the bed (at a lower level), The psychological individual factors of the nurse finding it easier to ask the husband rather than the wife with aphasia, perhaps being unsure about his wife’s ability to understand and answer the question, also plays a role. The individual (and activity-based) goal of getting information about the PWA is in focus. In short, the physical placement and eye contact between the nurse and the husband, the potentially less problematic approach of asking the husband, together with the physical placement of the PWA sitting on the bed, without face-to-face contact with the others and the individual background factor of her aphasia lead to an interaction pattern involving only the nurse and the husband. A more structured ACA analysis of background factors and communicative behaviour and how it can inform CPT in order to improve a communication situation like the one in the example will be further elaborated below.

Activity-based communication analysis in relation to other frameworks and its role in CPT

A main goal of ACA in relation to CPT is to provide a vocabulary and a structure for analysis of contextual influence on communicative behaviour, which can be used for enhancing the understanding of barriers and facilitators and for designing and training successful communication support. The framework addresses communication in different social activities and is related to other frameworks dealing with communication in everyday interactions. A social activity can be any situation where people are communicating, like in example 1, “conversation involving a PWA, her husband and a nurse in a hospital ward”. This activity can, in turn, contain smaller sub-activities, like “finding out food preferences of a PWA” in this conversation.

Taking the goals, roles, physical, and psychological conditions into consideration in analysing and supporting communication in different social activities is the main focus of applying ACA to CPT. There are the goals, roles, physical, and psychological factors of a social activity in itself, but also the goals, roles, physical, and psychological factors linked to each of the participants. These factors influence the patterns of interaction and by analysing them the conditions for successful communication may be identified and supportive changes can be made.

Figure 2, below, illustrates the background factors of a social activity and of a PWA and a communication partner and how they interact with interactive communication patterns. CPT may address changes of background factors and how they can affect communication patterns, as well as how changes in communication patterns can, in turn, affect background factors, as well as individual production and comprehension.

Figure 2. An overview of the ACA model: The arrows indicate the interaction between: (1) the background factors of the social activity and of the participating individuals’ background factors; (2) the background factors and the behaviour; and (3) participants’ means of communication and the interaction patterns in the activity.

Returning to closely related approaches, the awareness of the importance of analysing communication in naturalistic interactions has considerably increased and is by now firmly established in aphasiology. Some of the clinical motivation for this comes from the International Classification of Functioning, Disability, and Health (ICF; World Health Organization, 2001) inspired focus on “activity” and “participation” aspects of communication. In the schematic presentation of the ICF, contextual factors are included as two separate boxes representing “environmental factors” and “personal factors”, placed below the three entities “functions and structure”, “activity”, and “participation”. The ICF scheme has been criticised for many reasons, one of them being the placement of the contextual factors in the base as detached entities separated from, for example, “activity” and “participation” (Heerkens et al., 2017; Maribo et al., 2016). The ICF scheme thus reinforces the impression that each of these entities could be regarded as discrete units, which may be approached individually, instead of visualising the context impact on functioning.

Qualitative sequential analysis of patterns of communication in video-recorded interaction has added important information on phenomena, such as turn management, the use of gestures, and repair sequences in interactions involving PWA (see, for example, Goodwin, 2003; Wilkinson, 2015). This tradition, influenced by sociology, anthropology, and also by ethnomethodology (Garfinkel, 1967) is applied to the analysis of interaction involving PWA and also to CPT (cf. Beeke et al., 2015, 2014; Lock et al., 2001; Saldert, Johansson, & Wilkinson, 2015; Wilkinson, 2014). This type of approach, CA (Sacks, Schegloff, & Jefferson, 1974), often, like ACA, uses analysis of video-recorded interaction as a basis for suggesting changes in communication, mainly focusing on interaction patterns which are analysed in detail. It provides a good insight in interaction patterns at the local level of interaction, without focusing on external context and background factors. It is, thus, very strong on interaction patterns on a detailed, sequential level, analysis, providing information about the local sequential context. There has been reluctance within CA to draw on contextual factors beyond what is visible on a sequential level in the analysis of conversational interaction (Hutchby & Wooffitt, 2008, p. 27 ff.). The use of pre-set ideas of the influence from contextual factor such as, for example, the speaker’s cultural background or power relations, as a base for the understanding of social actions has been questioned within CA. Although applied CA, for example, the study of interaction in relation to aphasia, may include external context in the analysis (Wilkinson, 2015), the detailed analysis focusing on the context on a sequential level may still be considered as the focus and main ground-breaking contribution of the CA approach.

A social model approach to communication, taking the life situation and social network into consideration, is also advocated by many, in order to map out how the life of PWA and their communication partners, especially family and health professionals is affected (Simmons-Mackie, 2000). This approach has its main strength in outlining the social life of the PWA and thereby linking communication to their life situation and social network as a prerequisite for CPT and other types of intervention. It is, thus very strong on the overall context of the everyday life situation.

These two lines of inquiry and main concerns (the focus on detailed sequential analysis of conversations of CA and the focus on analysis of everyday life situations of the social model) are both closely related to the design of optimally relevant CPT (Kagan, 1998; Lock et al., 2001) and have been combined in a number of studies (cf. Simmons-Mackie, Raymer, & Cherney, 2016).

The direct influence of environmental factors on communication is much less studied. Studies by Howe, Worrall, and Hickson (2004, 2008a, 2008b), O’Halloran, Worrall, & Hickson (2011), and O’Halloran, Grohn, & Worrall (2012) have pointed to the need to identify environmental barriers and facilitators. This would facilitate the design of aphasia friendly environments. These studies identified requirements for such an environment in hospitals and more in general, by interviewing PWA and by observing communicative interaction between PWA and hospital staff. In categorising barriers and facilitators for communication in the hospital, two main themes were identified: (i) health care providers (i.e. their knowledge, skills, attitude, experience, etc.), (ii) structure and processes of the stroke unit (e.g. possibilities to communicate, access to the family, use of communication aids, physical environment). The authors conclude that CPT for hospital staff as well as improvements in the government of hospital processes are very important, both of them affecting the context of communication for PWA (O’Halloran, Grohn, & Worrall, 2012).

ACA relates to all of these approaches, taking the social activity context, with its background factors, as point of departure for the analysis of communication patterns and for identifying barriers and facilitators. Several studies using ACA in the direct analysis of video-recorded interactions involving PWA with different communication partners have been published (see Ahlsén, 2007, for a summary of some of them). These studies, have, for example, focused on multimodal communication by the same PWA in different social activities involving different communication partners and, thus, giving the PWA very different activity roles and thereby very different possibilities to communicate (Ahlsén, 2002). Comparative studies, using ACA, have also been made of communication involving other non-speaking participants with and without the use of AAC in different everyday social activities (e.g. Ferm, 2006; Rydeman, 2010; Thunberg, Ahlsén, & Dahlgren Sandberg, 2007, 2011) and of different methods for analysis and assessment of communicative interaction (Saldert et al., 2012). ACA is presented more in detail below.

Description of the activity-based communication analysis framework

ACA (Allwood, 1976, 1995, 2000, 2001, 2013) carries influences from a number of theories and methods, which have in different ways put language as action in context in focus, for example, Wittgenstein’s (1953) ideas of language as action and language games, Russian activity analysis (Vygotsky, 1978), anthropology (e.g. Malinowski, 1922), speech act theory (Austin, 1962; Searle, 1969), the conversational principle and maxims (Grice, 1975), and CA (e.g. Sacks et al., 1974), stressing patterns of interaction and sequential analysis. The ACA approach is also well in harmony with the SPEAKING (setting, participants, ends, act sequences, key, instruments, norms, genres) grid suggested by Hymes in his description of the ethnograpy of communication (Hymes, 1972, 1974) with the purpose of structuring field work, but also later applied in discourse analysis of different areas. The cultural aspect is also present in ACA, which is, for example, well suited for the analysis of intercultural communication (see, for example, Allwood, Berbyuk Lindström, & Lu, 2011).

The main assumption of ACA is that social activities come with a number of determining or influencing factors that can be identified and provide different conditions for communicative interaction. There are, thus, a number of collective influencing background factors that are associated with the type of activity and affect all participants, as well as a number of individual influencing background factors, that come with the individual participants. These factors provide the conditions and affordances that go into the interaction (see Figure 2). In studying the actual interaction, the communicative behaviour is also continuously providing a local “internal” context for the following interaction, thus, the preceding interaction changes the context for the succeeding parts of the interaction and becomes part of the dynamic context of the social activity.

Starting from the context, of the social activity, ACA stresses the fact that the conditions influencing communication can vary considerably between different social activities in daily life, where people interact. This can be specified as a number of influencing background factors in the particular context. An important task is, thus, to keep track of and analyse what these factors are and how they interact in each situation. This provides a scaffolding of the communicative behaviour in a social activity, which helps in understanding and explaining what will happen, as well as why it happens and this also enables predictions for different social activities. By understanding this, it becomes possible to tailor CPT to meet the specific communicative demands of different social activities.

There are four main types of influencing background factors linked to the social activity as such:

First, we have the goals of the social activity itself, that is, mainly the conventional goal, like selling–buying in a shop, teaching–learning in a classroom, dinner conversation at home, etc. The participants have to communicate towards these goals and it is important to know what the requirements are to do so and how to make it possible to participate in reaching this goal.

Second, we have the roles that are given by the social activity, for example, assistant–customer in a shop, teacher–pupils in a classroom or husband–wife having dinner at home. The participants have to be able to fulfil these roles.

The third type of background factors are the physical circumstances. Where is the interaction taking place, what are the conditions for seeing and hearing each other, reading and writing messages, using objects or communication aids, etc.? These factors are sometimes given by the activity, but can often be changed in order to make communication easier.

The fourth type of background factors are the psychological conditions linked to the social activity (e.g. how the activity and the activity roles to be filled affect emotions and attitudes of the participants). For example, a police interrogation usually involves more tension and worry and less enjoyment than a play interaction between a parent and a child.

Identifying actual and potential social activities where a person with aphasia and a communication partner will interact, how the background factors may influence their communication and how background factors can be changed is an important part of ACA.

Turning to the participants in the social activity, four individual types of influencing background factors, “parallel” to those of the social activity as such, can also be identified:

Goals of the participating individuals (individual goals in general carried into the interaction by the participants). In the analysis, the matching of these goals and the goals of the social activity and of the other participant(s) need to be considered. It is not always the case that the goals of all the participants are the same as the goal of the activity as such and different individual goals have to be identified. For example, the goals of a seller and a buyer can be different in terms of the desired price for what is sold/bought and the individual goals of teachers and pupils are not always the same.

Roles of the participating individuals (e.g. their identity based on age, education, income, family role, profession, health condition, etc.) How these individual roles match the roles given by the social activity type and the roles of the other participants should be analysed. For example, a person who used to be a managing director might have difficulties assuming the role of a patient in a hospital.

Physical characteristics of the individual participants (e.g. size, looks, mobility, etc.). The analysis should find out how they match the physical conditions of the activity and the physical characteristics of the other participants, in order to assess their possible influence. In this category, some of the characteristics having to do with aphasia and other physical limitations after a stroke come in as factors that may be more or less compatible with the physical conditions of a particular social activity.

Psychological characteristics of the individual participants (e.g. mood, personality, any psychological problems, etc.) ACA also analyses how the individual psychological characteristics match the psychological conditions of the activity and the psychological characteristics of the other participant. Also here, factors having to do with the aphasia of the PWA and the characteristics of a CP can be identified and their effect on the interaction can be analysed.

As can be seen above, the analysis of background factors involves (i) identifying the factors for targeted social activities, and (ii) analysing how these factors may interact with each other. For example, (i) could involve activities identified in a diary- type of analysis of daily life situations in a social model approach, for the individuals in focus, while (ii) could involve how collective social activity factors and individual factors may interact and how the individual factors of the participants may interact, that is, a “matching exercise”. This involves obtaining some control of how the social activities in daily life can be handled in the best possible way, given activity factors and individual factors.

Having analysed the influencing background factors of the collective activity and individual participants and how they can interact, we turn to their influence on the communicative behaviour of an interaction in a specific social activity.

The actual behaviour can be observed, for example, in a video recorded interaction. It can also, to some extent, be predicted from the constellation of background factors. What is, then, the communicative behaviour that can be studied?

The communicative interaction patterns are a very important part of the communication. What is analysed in ACA are the same types of patterns that are analysed in CA, the main difference being the focus on linking the patterns to influencing background factors in ACA. One part is, thus, turn management, which studies how contributions are distributed between the participants, that is, who has “the right to speak/contribute” or “the floor” when in the interaction and why, something that needs to be attended to in CPT. A second characteristic of interaction patterns is how feedback is elicited, given and interpreted. Feedback consists mainly of, often small and unobtrusive, words or sounds (e.g. yes, no, mm), head movements (e.g. nods and shakes), shoulder shrugs and facial expressions, showing reactions to contributions from other participants. There are also many typical sequences of contributions that occur in different social activities. Some of these are referred to as “adjacency pairs” in CA, for example greeting–greeting, question–answer, but there are also longer sequences typical, for example, for buying a ticket, phoning about a hotel room, etc.

Finally, the production and comprehension of communicative contributions by each of the participants, using speech, as well as other means of communication (e.g. gesture, drawing, writing, and other AAC) is analysed and, in turn, related to interaction patterns and influencing background factors.

With the help of an ACA analysis of communication in different social activities, problems, and possibilities can be discussed among PWA, communication partners, and clinicians, in order to choose what to focus on in intervention and training. Certain social activities can be focused because they offer favourable conditions (or facilitators) for communication, others can be chosen as challenges (or barriers) where strategies and support have to be provided. The overview in Table 1 can be used as a checklist for this type of analysis.

Table 1. Types of factors in ACA – overview.

	Collective – of the social activity	Individual – of each of the participants
Influencing background- factors	- Goals of the activity - Roles given by the activity - Physical environment of the activity - Social-psychological factors of the activity	- Goals of each participant - Roles of each participant - Physical/biological features of each participant - Social-psychological features of each participant
Behaviour (interactive and individual)	Interaction patterns in the activity: - Turn management - Feedback giving and elicitation - Sequences of contributions	Production of contributions: (pragmatics, semantics-lexicon, grammar and phonology-phonetics, gesture, Alternative and Augmentative Communication (AAC), etc.) Perception/Comprehension of contributions: (pragmatics, semantics-lexicon, grammar and phonology-phonetics, gesture, AAC, etc.)

The interaction between the influencing background factors and the communicative behaviour observed in the interaction is illustrated by the arrows in Figure 2.

Intervention can, thus, be directed towards the interaction between different components of the ACA model: (1) the interaction between background factors of the social activity and those of the individual participant, and, in particular, (2) the interaction between background factors and behaviour (especially interaction patterns), and (3) the interaction between interaction patterns and individual production and comprehension.

An activity-based communication analysis of Example 1 and suggestions for communication partner training

Based on Example 1 above we now look more closely at a systematic analysis of background factors and behaviour according to ACA.

Table 2 illustrates a possible overview ACA analysis of influencing background factors of the activity and the participants and of how the interaction patterns may be influenced by these factors.

Table 2. Overview of results from an ACA analysis of background factors and communicative behaviour in Example 1.

Background factors
Collective background factorsof the social activity	Individual background factors of each of the participants
- Goals of the activity For the PWA and her husband – to see each other (Activity: Visit) For the nurse – to find out facts about the PWA (=subactivity)		- Goals of each participant Nurse: to find out facts about the PWA PWA: to be able to communicate herself, to see husband Husband: to be polite and helpful in contact with the nurse, to see wife, to give and find out facts
- Roles given by the activity Nurse, patient, visiting spouse		- Roles of each participant PWA: patient, given passive role Nurse: nurse, information seeking Husband: spouse, information giving
- Physical environment of the activity PWA sitting on bed, visitor and nurse standing face-to-face “talking over her head”		-Physical/biological features of each participants PWA: Aphasia (communication difficulties) Nurse: healthy adult Husband: healthy adult
-Social-psychological factors of the activity The nurse has the initiative. The healthy participants – fully able to communicate – find it natural to do this. The PWA in the role of a patient with limited communication is not expected to be able to supply information.		-Social-psychological features of each participant Nurse: may be anxious about creating an awkward situation if she and the PWA cannot communicate successfully Husband: wants to respond politely to the nurse PWA: not able to claim the turn, left out of conversation
Behaviour
Collective communicative behaviour of the social activity: Patterns of interaction		Individual communicative behaviour: Production and Perception
- Turn management Turns exchanged between the nurse and the husband, verbally and nonverbally, since they have eyecontact, facing each other. The PWA is excluded from the turn management, since the others are not facing her or addressing her verbally. Typically, an adult person in the role of patient, would be addressed directly, but the fact that the PWA has aphasia has changed this pattern		- Production of contributions Nurse: typical in dyad with husband, no production directed at the PWA Husband: typical in dyad with nurse, no production directed at the PWA PWA: not able to take turn, can possibly give some verbal feedback but is not encouraged to do so, has no communication support
- Feedback giving and elicitation - Feedback is elicited and given by the nurse and the husband to each other’s contributions. Feedback from the PWA is not considered		- Perception/understanding of contributions Nurse: typical dyad with husband, no perception of contributions from PWA Husband: typical dyad with nurse, no perception of contributions from PWA PWA: Cannot be observed here, but, known to be fairly good
- Sequences of contributions Sequences contain only contributions from nurse and husband

Going through the ACA analysis table is a structured way of identifying barriers and facilitators to identify what can be changed in the context to improve communication. In the example provided above, the ACA would have identified that the nurse’s lack of knowledge about how to support communication could cause anxiety, which might make her stay with the “safe” option (to address the husband only); and this condition could cause social exclusion for the wife, who would be likely to feel disempowered. CPT would, in this case, have focused on giving priority to directly addressing the woman with aphasia from a face-to-face position (a change in physical conditions), making this a goal of the activity (a change of activity goal), and, if needed, providing communication aids, such as pen and paper and pictures, thus further improving the physical conditions for communication. This would also have improved the psychological conditions, especially for the PWA, who would have felt more included, respected, autonomous and able to fulfil her goals (cf. Pound & Jensen on “humanisation”, in this issue). It would have created increased awareness and a change of goals in the nurse and the husband and made the situation less awkward and more inclusive, probably also releasing potential tension or embarrassment in them, since the nurse, as well as the husband might well be (more or less consciously) feeling guilty about excluding the PWA. These changes in background factors would, in turn, have changed the interaction patterns radically, by including the PWA in the turn management, feedback and sequencing of communication.

Returning to the ACA model presented in Figure 2, CPT can, thus, enrich the analysis of the interaction mainly in three places

1. in the analysis of context, that is, background factors for the social activity and how background factors of the social activity relate to those of the individual participants

2. in the relation between background factor and behaviour, in order to make changes for facilitating communication (interaction patterns and communicative production and comprehension)

3. in the analysis of communication patterns and how they are related to production and comprehension, in order to make changes improving this relation

A case illustration of activity-based communication analysis in communication partner training

To illustrate the relation between the different features of the activity and the individual background factors and the effect of these factors on the interaction patterns, we here present a case featuring everyday interaction in a dyad including an elderly woman with aphasia, resident in a nursing home, and her key nursing assistant (NA), who was involved in an eight sessions long CPT programme. The case also illustrates how changes in the interaction patterns allow the participants to alter their roles and goals in the activity. Note that the effects of this CPT programme have been explored elsewhere (see Eriksson, Forsgren, Hartelius & Saldert, 2016, for details regarding the design and results of the intervention study). The case is presented here with the purpose of displaying how ACA may be used to provide important information as the basis for the planning of CPT and to describe the effects from CPT on the participants’ interaction and roles in an activity.

The PWA in the dyad presents with a severe aphasia. Her mother tongue was Swedish. Besides this, the PWA brings several other relevant influencing individual background factors into the communicative encounter (see overview in Table 3). The PWA had been working as a shop assistant and had always taken good care of her looks and it meant a lot to her to look her best. Another individual background factor which she brought into the interaction was that her comprehension was severely impaired. She only scored 36 of 231 points on Token test (De Renzi & Vignolo, 1962). These background factors affected her communication and her participation in the conversational interaction. Her speech output was sparse and due to phonological errors, it was difficult to understand for her communication partners. She did not manage to produce any words in tests of phonological and semantic word fluency. In conversational interaction, she would respond with minimal responses, such as yes/no or produce subtle nods or head shakes. Sometimes she made attempts to produce verbal utterances but those attempts were quickly abandoned. She occasionally expressed herself by deictic pointing gestures; she used iconic gestures, but only infrequently.

Table 3. Overview of results from an ACA analysis of background factors of the social activity and the participants before participating in communication partner training (Nursing assistant – NA; the person with aphasia – PWA).

Background factors
Collective background factorsof the social activity	Individual background factors of each of the participants
- Goals of the activity Primarily to get the PWA groomed and dressed, but also an opportunity for doing sub activities like physical training and social interaction		- Goals of each participant NA: to get PWA ready as quick as possible, to do a good job and take advantage of the opportunity to facilitate training and social interaction PWA: to get the help she needs with her grooming; to be as independent as possible and to communicate
- Roles given by the activity NA is a care provider and a facilitator of training and communication, PWA is a receiver of help, both are potential communication partners		- Roles of each participant NA: Facilitator of care, training activity and communication PWA: Can be more or less a passive or active receiver of care and as participant in communication
- Physical environment of the activity PWA sitting in a wheel chair in front of the mirror in a quite small bathroom. Several objects are involved in the different sub activities		-Physical/biological features of each participants NA: A healthy adult PWA: has communication difficulties and hemiparesis due to stroke
-Social-psychological factors of the activity The NA takes the initiative, both in terms of the care activities and social interaction. NA often experience time constraints due to shortage of staff. The PWA in the role of being a resident in need of care is expected to accept the provided help and any initiatives to social interaction		-Social-psychological features of each participant NA: professionally experienced but may want to avoid creating an awkward situation if she and PWA cannot communicate successfully. PWA: dependent on NA in the activity, both in the care activities and in communication
Behaviour
Collective communicative behaviour of the social activity: Patterns of interaction		Individual communicative behaviour: Production and Perception
- Turn management Turns are exchanged between the NA and the PWA. The NA is the one who initiates different sub activities including communication. PWA takes no initiatives.		- Production of contributions NA: unimpaired verbal and non-verbal means of communication; rarely offers PWA communicative support PWA: Mainly produces yes and no responses verbally and non-verbally; sometimes uses deictic (e.g. pointing); impaired production of phonology-phonetics and word retrieval
- Feedback giving and elicitation Feedback is elicited, but NA is not always awaiting a response from PWA.		- Perception and Comprehension of contributions NA: reduced attention for contributions from PWA PWA: responds to requests from NA in activity, but formal tests expose impaired comprehension on word and sentence levels
- Sequences of contributions Sequences involve NA contributions monitoring the care activity and requesting and encouraging physical training activity on behalf of the PWA

The NA was a woman in her forties with Swedish as her mother tongue. She had been working with the PWA since her arrival at the nursing home nine months earlier. The NA had 12 years of education, including the 18 months long NA basic care training programme. Just as the resident she had a personal interest in the world of different beauty products and from her appearance in the video recordings she seems to put some efforts into her own looks. Her language background and experience of working with people with communication disorders as well as her own personal interests are parts of her individual background factors which she brings into the communicative encounter and which may affect her communicative interaction with the PWA.

The social activity context for the interaction described below is a morning routine practice, which is similar every day and takes place in a bathroom in the PWA’s apartment. One of the main goals of this activity is to get the residents ready for the day. The main activity consists of several sub activities where the NA helps this particular resident to, for example, have a wash, brush her teeth and get dressed. The NA brushes the resident’s hair; the routine also includes putting on face cream, body lotion and deodorant. A main goal in the activity is related to the fact that the activity provides a natural opportunity for physical training for the residents, and they are supposed to be encouraged to perform some of the tasks themselves, even if it takes more time. Another main goal of this activity is to make the most of the opportunity for staff to interact socially with the residents. The morning routine is usually the only occasion where the nursing staff has a little more time alone with the residents and this is an opportunity for establishing interpersonal relations (Forsgren, Hartelius, Skott & Saldert, 2016). Besides the inherent goals, the activity is affected by factors such as time constraints as other residents are waiting for help, and the physical environment, that is, the use of objects related to the care routine in a quite small room with a mirror, and these factors may interact with the individual background factors of the PWA and the NA.

The NA’s role in the activity includes helping the PWA to accomplish the morning routine, to encourage and coach her physical training and to be a communication partner in social interaction. The PWA has the role of being the one who gets help with issues, which may be quite intimate, and which adults normally are expected to be able to perform themselves. Thus, there may be a certain amount of awkwardness attached to the role of being the recipient of care. Still, typically, being a resident in a nursing home has been described as being obliged to accept care when it is offered, and in the form, it is provided (see, for example, Lagacé, Tanguay, Lavallée, Laplante, & Robichaud, 2012; Timonen & O’Dwyer, 2009) . Further, the PWA is also obliged to be a conversation partner when the staff takes the initiative to interact.

Below we explore the effects from the background factors on interaction patterns in:

1. one extract of the video-recorded interaction between the resident and the NA from before the NA participated in CPT,

2. one extract from after the training.

A video camera was placed in the bathroom by a researcher, who made sure that both the NA and the PWA were visible and that the camera was running before leaving the dyad alone during the recording. The participants were instructed to interact and perform as they usually did during this activity. The transcriptions of the extracts included non-vocal features, such as gestures and other body movements, as well as talk, using standard CA transcription conventions. A key to the transcription symbols used is provided in Appendix 1. The transcriptions were translated into English by the authors.

Extract 1 (see Table 4) illustrates how the NA before the CPT is focusing on the role of being the one who initiates the sub-activities and decides in which order carried out. In line 1 the NA moves on from one sub-activity (washing and drying) to the next. In this case, she initiates work directed at the goal of the PWA’s physical training and she presents her decision as a question in line 1. Although the NA is wording her contribution as a suggestion, with a rising intonation, she has no eye contact with the resident while she presents the question. The PWA’s communication disorder impedes her ability to produce a verbal response. She is raising her head and opens her mouth but no sound is heard (line 2) and her initiative is not explicitly acknowledged by the NA. However, after this, the NA glances quickly at the PWA and in the absence of an explicit objection she gives the deodorant to the PWA in line 3. The lack of eye contact, along with the lack of explicit negotiation or verification of the PWA’s view regarding the chosen sub-activity, puts the resident in the role of being a passive object of care, see Pound & Jensen in this issue. Although the NA is reporting her intentions (e.g. line 5), she again starts handling the PWA’s body without providing any opportunity for the PWA to acknowledge the decision. The NA does not make eye contact with the PWA until after two seconds (in line 8) when she is asking the PWA if it is ok that she is applying lotion while the PWA is putting on roll-on, and now she also receives a verbal response from the resident.

Table 4. Extracts of conversational interaction between the nursing assistant (NA) and the person with aphasia (PWA) before and after communication partner training.

Extract 1: Before communication partner training The PWA is sitting in front of the mirror in the bathroom during the morning routine task. The NA is standing beside her and has just finished drying the PWA’s back with a towel. The PWA has her head tilted forward.			Extract 2: After communication partner training The PWA is sitting in front of the mirror in the bathroom during the morning routine. The NA is standing beside her and has just finished drying the resident’s back with a towel. She takes a bottle of lotion and holds it in front of the resident and asks if she wants some of it.
1	NA	shall you try to put on eh deodorant shall you? (gaze first directed at PWA’s back, then down when putting towel in PWA’s lap, then turning to basin searching for the deodorant)	1	NA	shall I apply some lotion? (leans towards and looking at PWA)
2	PWA	(lifts her head and opens mouth but no sound is heard)	2	PWA	(nods)
3	NA	(a quick look at PWA then takes of the lid of deodorant and gives it to PWA)	3	NA	(opens bottle and starts putting some lotion in her own hand then turns to PWA with bottle)
4	PWA	(takes the bottle in left hand)	4	NA	yes you do you want to apply some to your hands and arms? (moves bottle between PWA’s hands and looks at her)
5	NA	I thought I’ll put lotion on your shoulder here at the same time (takes lotion, still without eye contact with PWA, and starts applying on shoulder)	5	PWA	°hh yes° (nods, takes bottle in left hand, looks down)
6	PWA	(starts putting on deodorant under the right arm)	6	NA	sometimes you enjoy that (moves behind PWA, starts applying lotion on PWA’s back)
7		(2.0) (PWA putting on deodorant, NA applying lotion without eye contact)	7	NA	would you like to take some of this (.) on your arms and hands (moves besides PWA, points at bottle, then touches PWA’s left hand and arm then right hand, then looking at resident)
8	NA	does it work for you if I do this at the same time (looks at PWA)	8		(12.0) (PWA starts putting lotion on her right hand; NA applying lotion on PWA’s back with intermittent quick glances at PWA’s activity)
9	PWA	°.hh yes° (applying deodorant under the right arm)	9	NA	it smells good this one (quick glance at PWA, still applies lotion on PWA’s shoulder)
10	NA	Yes (looking at PWA’s activity through mirror applying lotion)	10	PWA	°.hh yes° (puts lotion into her hand then on right arm)
11		(2.5) (NA looking at PWA putting on deodorant while rubbing shoulder)	11	NA	you do get dry skin don’t you
12	NA	this is going just fine (looking at PWA’s activity through mirror)	12	PWA	(looks up at NA, rubs lotion on hands)
			13	NA	when you sit outside (looking at PWA)
			14		a lot in the sun
			15	PWA	°.hh yes° (applies lotion on hands)
			16		(2.0, both applies lotion)

The NA fulfils her role as a coach in the PWA’s physical training by watching her trying to put on the roll on in lines 10–12, and she is also encouraging the PWA in a positive manner in line 12 (this is going just fine). However, by choosing this role she also focuses the conversational interaction on an evaluation of the PWA’s physical ability, or lack of ability. In this encounter, the PWA’s role is reduced to being a passive object in need of care and support, and this prevents a more personal and equal social interaction. The interactive pattern is characterised by initiative on behalf of the NA, but she rarely waits for, or considers, any feedback from the PWA.

During the intervention (see Eriksson et al., 2016, for a description of the intervention) it became clear that the NA was not aware of how impaired the PWA’s comprehension was. Her interaction style was not adapted to the PWA’s needs in terms of time and support for her communication and the NA often missed the PWA’s subtle contributions. As in CPT in general, the main objective of the training was to facilitate the PWA’s ability to express herself and participate in the interaction in a more equal manner by means of better communicative support from the NA (see Saldert, Jensen, Blom Johansson, & Simmons-Mackie, in this issue). The ACA here reveals how roles may be altered and how other factors in the activity itself and features of the individual background factors may be utilised to promote a more active participation on the behalf of the PWA. The training targeted the NA’s skill in supporting comprehension with multimodal cues; the NA was encouraged to try to be more attentive and to allow more time for the PWA to respond to the NA’s techniques for facilitating communication.

Extract 2, see Table 4, features an example of an interaction video-recorded after the NA had participated in CPT. In this extract, the NA is utilising the techniques she learned in the training. She has slowed down her speech, provides support for the PWA’s comprehension, and her contributions in this video are more often followed by a pause before her subsequent contribution. She is now more careful to preserve the PWA’s sense of agency when she tries to get her consent to suggested actions (line 1). This reduces the number of communicative exchanges where the PWA may be perceived as a passive object of care. The NA makes sure to establish eye contact, which enables her to notice any feelings displayed by the PWA and to monitor her comprehension. Although the NA is still taking on her role as coach in PWA’s physical training in extract 2, she presents the task as a shared activity which they will perform together, when she asks consent from the PWA to apply lotion herself, and also awaits a response and an initiative from PWA (line 5). Furthermore, she supports the PWA’s comprehension, when she shows the bottle to her (line 4) and repeats the question and touches her hands and arm with the bottle in line 7. In line 6 the NA displays her view of the PWA as a unique person, one of which she has personal knowledge about, by saying that “sometimes you enjoy that”. She is monitoring the PWA’s achievements but in a discrete manner while they are working together on the task. In line 9, the NA is focusing on their shared experience when she refers to the aroma of the lotion. She then links the activity to the world outside the context of providing or receiving care in a nursing home when she brings in the topic of their shared experience of getting dry skin from sitting outside in the sun a lot (lines 11–12).

To sum up: Although the PWA is still confined to responding by providing minimal vocal and non vocal acknowledging responses, the change in the NA’s behavioural patterns enables her to participate more actively in the interaction. The focus is redirected from the PWA’s bodily impairment to the shared experience of the activity. The PWA’s uniqueness as a person is affirmed by the NA who references her knowledge about the resident’s preferences and feelings and the focus is thus shifted from her inability to social interaction on personal matters.

The change in their interactional pattern is both dependent on and a prerequisite for the NA to focus more on the sub goal of facilitating social interaction than on the physical training. Furthermore, this means that the NA elaborates her role as facilitator of communication besides providing care in the activity.

Discussion and conclusion

As the examples above illustrate, the analysis of background factors can explain observed interaction behaviour in different social activity contexts and thus answer questions about what happens in communication. The examples of interaction presented here also illustrate how ACA may guide and motivate change in a communication partner’s behavioural patterns and enable a PWA to communicate and participate more in the interaction. In the case illustration, the change in the NA’s understanding of how the PWA’s individual background factors affect her communicative behaviour, and her learning of supportive communication allows her to adapt her own communication to the needs of the PWA. She also utilises their shared individual background factors, that is, their shared personal interest in beauty products and physical appearance as well as the available objects in the morning routine, as resources in the interaction.

The ACA perspective involves relating factors at different aspects of analysis (background factors and behavioural factors, individual, collective, and interactional). An increased focus, especially on linking influencing background factors to what happens in interaction is a main characterising feature of the framework.

For CPT, this means that the focus in analysing interaction and initiating changes in behaviour considers interaction patterns in relation to influencing background factors. It preferably starts by the analysis of the background factors of the social activities that are addressed and the individuals that are involved in them. Following this, the interaction patterns and production/comprehension of communication are analysed with respect to how they may be influenced by the background factors and how they can be changed in relation to these factors. The analysis of interaction is, in this way, always placed in the context of the social activity. The focus is, thus, not only on how to change interaction patterns. It is also on how to change interaction patterns in relation to background factors and also how to change the conditions of the social activities, that is, background factors, if possible.

This emphasis on the effects of the interaction between different contextual factors and behaviour in the ACA may in many ways supplement the social model approach and the ICF perspective in rehabilitation practises in aphasia. Further, the conscious and explicit structured analysis of the context of everyday life may add on to the detailed sequential analysis of conversations in applied CA.

The advantage of using ACA is that it facilitates a more thorough understanding of how different factors interact in determining what happens in interaction and how these factors can be affected by changes in contextual factors, thereby also facilitating follow-up and evaluation of CPT, as well as prediction of CPT effects in relation to different social activities. In the planning of CPT, the results from ACA may enable the SLP and the participants to discuss the conditions for the conversational interaction in different activities and the effect of this on, for example, the roles in the interaction. Such insights may motivate the communication partners to change their interaction patterns by use of the tools provided in the CPT. In the examples provided here we have illustrated encounters in hospital and residential care contexts. However, insights about one’s own and communication partners’ roles and interaction patterns in communicative activities may also be important to achieve, for example, a reduction in occurrences of too many didactic sequences in spouses’ everyday conversations (Bauer & Kulke, 2004). That is, understanding the effect on communication of different roles, goals and behaviour in specific activities may prevent the non-aphasic spouse from taking on the role of speech trainer in social activities where this is not optimal, such as in everyday conversations, and allow a more equal and balanced social interaction between spouses.

Some of the main areas of application of ACA are

• to search for targets in social activities in relation to therapy, considering different constellations of background factors and their influence

• to predict communication in different types of contexts, based on the relevant background factors, and prepare support accordingly

• to identify why tools for communication, for example communication aids, are used or are not used in a certain social activity context

• to identify physical as well as psychological barriers and facilitators for communication and use ACA as a basis for understanding, discussing and planning support and therapy.

The structured analysis of communication in context provided by the use of ACA, thus enables the pinpointing of possible effects from CPT beyond goals related to getting a message across in a communicative activity.

The high degree of ecological validity also has to be evaluated in relation to the possible variation in everyday contexts. This means that while the analysis will, in general, make possible comparisons of communication in a particular social activity before and after intervention and also of more specific features of the interaction, it may be difficult to obtain high reliability numbers or perfect agreement between analysts, since the reliability of a method is dependent of the possibility of keeping several factors constant. However, since the purpose of ACA is to deal with everyday communication contexts, and since variability is a condition of such contexts, the validity of the method is still considered a strength.

Regarding clinical feasibility, the ACA analysis can be made on different levels of ambition and degrees of detail, depending on the purpose of the analysis. Since it contains many parts, it can be quite time consuming. The cost of time spent on the analysis, which is preferably based on observation of video-recordings in combination with interviews about communication in everyday social activities, however, has to be considered in relation to the benefits of fairly comprehensive coverage of relevant factors. In increasingly multicultural societies, a focus on the influence of background factors can also assist in designing CPT.

With respect to time for analysis weighted against depth and breadth of coverage, as well as with respect to ecological validity versus strict reliability, a cost-benefit analysis has to be made in relation to the purpose at hand and what outcome information is most useful. Central questions to consider are what is to be evaluated and why and for whom the outcome evaluation is made.

Acknowledgements

With thanks for use of the images to the Norfolk Conversation Partners, the Supported Communication Research Team, *cast and MAKE at the University of East Anglia and NIHR RfPB (PB-PG-06-09-17264) for funding the DVD from which these images were extracted.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Danish Council for Independent Research in Humanities under grant number DFF-4180-00046.

Appendix 1

Key to symbols used in transcriptions:

(0.7) Numbers in parentheses indicate silence in tenth of second

(.) Silence shorter than 0.5 s

.hh Audible in-breath

? Rising intonation

words Underline indicates emphasised word or syllable

°no° Degree signs indicate a passage of talk which is quieter than surrounding talk