Applying the intervention model for fostering affective involvement with persons who are congenitally deafblind: an effect study.
In a previous paper we introduced the Intervention Model for Affective Involvement (IMAI) to improve affective involvement during interaction and communication between persons who are congenitally deafblind and their communication partners (Martens, Janssen, Ruijssenaars, Huisman, & Riksen-Walraven, 2014a). An intervention based on this model was evaluated in a single-case study (Martens et al., 2014b). After onset of the intervention, affective involvement between the participant and his communication partners increased, as did the participant's positive emotions, while his negative emotions decreased.
The aim of the present study was to apply the IMAI to four participants who are congenitally deafblind (henceforth referred to as "clients") and their 16 communication partners in different settings and interactional situations, and to examine the effectiveness of the intervention. We examined: (1) whether the intervention increased affective involvement and positive emotions and reduced negative emotions across clients, and (2) whether the intervention was effective across communication partners (teachers, teacher assistants, caregivers, caregiver assistants), interactional situations, and settings (schools, group homes, daytime activities centers).
PARTICIPANTS AND SETTINGS
The study followed the tenets of the World Medical Association Declaration of Helsinki on Ethical Principles for Medical Research Involving Human Subjects and was approved by the Institutional Review Board of Royal Dutch Kentalis. Informed consent was obtained from the clients' parents, legal representatives, and communication partners.
Four clients (Clarice, Bryan, Dewy, and Nick, see Table 1; all names are pseudonyms) and 16 female communication partners (see Table 2) participated. The clients were selected for participation based on three criteria: (a) a dual sensory impairment from birth on, (b) showing difficult emotional behaviors, and (c) a request for coaching by the communication partners. The communication partners were selected based on the following criteria: (a) working frequently with the client, and (b) having difficulties in sharing emotions with the client.
The communication partners received coaching from three coaches (one for Clarice and Nick, one for Bryan, and one for Dewy). The coaches were familiar with the aim, principles, and protocol of the IMAI, and they were skilled in conducting video analysis and coaching with video feedback. They had extensive work experience and were specialized in interaction and communication with persons who are congenitally deafblind.
The study was conducted in structured settings involving a fixed-day program (school: Bryan, and Dewy; daytime activities center: Clarice) and in unstructured settings involving a more flexible day program (group home: Clarice, Dewy, and Nick). All these settings are part of Royal Dutch Kentalis, an organization specializing in auditory and visual disabilities.
The IMAI-based intervention aimed to foster affective involvement between the clients and their communication partners during interaction (for phase I of the intervention, see steps 1 through 4 in the Intervention protocol section) and communication (for phase II of the intervention, see steps 5 through 8 in the Intervention protocol section). The intervention principles were to improve the communication partners' competence in: (a) recognizing affective behaviors, (b) attuning to interactive behaviors, (c) sharing meaning for better understanding, (d) sharing emotions and evaluating the adequacy of their own affective behavior during interaction and communication, and (e) adapting the context to promote affective involvement.
The coaches followed the seven-step protocol to foster affective involvement and trained the communication partners using team and individual coaching in steps 4 and 6. Coaching included video analysis, information transfer, and role-playing (Martens et al., 2014a).
The first phase of the intervention protocol, "Phase I: Intervention focusing on Interaction," that was followed by the coaches involved the following steps: Step 1. Determining the question. The communication partners of the four clients requested coaching.
Step 2: Clarifying the question. Information on the clients' characteristics was gathered and relevant interactional situations were chosen in consultation with the communication partners. Table 3 lists the target questions for the clients.
Step 3: Analyzing interactions. Recordings of interaction situations with each client were analyzed to formulate intervention aims based on the four core categories, defined by Martens et al. (2014a) as: (a) attention: focusing on the interaction partner, the content of the interaction, and the persons or objects within the interaction context; (b) initiatives: starting an interaction or raising a new idea or issue as part of a reaction; (c) regulating intensity: waiting while the client is adapting the intensity or pace of the interaction or is processing new information; and (d) affective involvement: recognizing positive and negative emotions and sharing these emotions in a positive way that is perceivable for the client. Table 3 shows the intervention aims and interactional situations for all clients.
Step 4: Implementing intervention focusing on interaction. During two 120minute team coaching sessions and three 60-minute individual sessions over 10 weeks, the communication partners were trained to change their interactive behaviors. The client's communication modalities were taken into account (see the communicative functioning column in Table 1) and the communication partners were coached to foster affective involvement while simultaneously focusing on the four core categories. The following are examples of activities for each category: (1) attention: communication partners were taught to tap on an object together with the client or to co-actively roll a ball on the floor; (2) initiatives: communication partners were taught to reduce or repeat initiatives; (3) regulating intensity: communication partners were taught to wait for the client's initiative or to exaggerate their movements; (4) affective involvement: communication partners were taught to share emotions tactilely by rubbing the client's arm or imitating muscle tension. The steps for "Phase II: Intervention focusing on communication" of the intervention protocol were as follows:
Step 5: Analyzing communication. For the communication phase, new videos were analyzed to formulate intervention aims according to the three core categories of behavior, defined by Martens et al. (2014a) as: (a) sharing experiences: elaborating on events and introducing new events so that the client becomes motivated, feels secure, and knows what is going to happen; (b) sharing meaning: interpreting and affirming the client's expressions of communication and taking turns to negotiate about the correct meaning of an expression; and (c) affective involvement: recognizing positive and negative emotions and sharing these emotions in a positive way that is perceivable to the client. See Table 3 for the intervention aims of the communication phase.
Step 6: Implementing interventions focusing on communication. In a procedure similar to Step 4, the communication partners were trained to change their communicative behaviors. The following are examples of target behaviors: (1) sharing experiences: increasing coactive acting or checking whether information was perceived; (2) sharing meaning: using more turns to negotiate about a certain topic; (3) affective involvement: using friendly facial expressions or using rhythmical vocalizations and higher pitch.
Step 7: Evaluating. The coach evaluated the intervention with the communication partners in a separate team session. Video fragments of the first recordings at baseline and the last recordings of phases I and II were used to support the evaluation.
Intervention fidelity was secured through close collaboration between the coaches and researcher during the implementation of the intervention, and by monitoring (video recording all coaching sessions) and supervising the coaches. Moreover, the coaches followed the seven-step IMAI intervention protocol. In consultation with the communication partners, they formulated aims and target behaviors to ensure the engagement of communication partners. The coaches monitored changes in the behavior of communication partners by reviewing the most recent videos recorded during the intervention.
A multiple-baseline design across subjects was used to examine the functional relationship between intervention conditions (baseline, intervention phase I, intervention phase II, and follow-up) and possible changes in affective involvement between the clients and their communication partners as well as the clients' expressions of negative and positive emotions (described under Observational categories) (Barlow, Nock, & Hersen, 2009). The baseline started in the same week for all clients and was continued while the intervention was randomly introduced for Clarice (week 8), Bryan (week 9), Dewy (week 10), and Nick (week 12). The intervention was followed by follow-up measures after 2, 4, and 6 months.
Observation and measurements
Video recordings of interaction situations of at least 20 minutes were used to observe the effects of the intervention. The situations differed per client (see Table 3). The activities of each communication partner were videotaped weekly during weeks 28-32 and during follow-up. From these videos, 182 recordings in different settings and with different clients were selected for analysis. The observers were not informed about the observation phase, definitions were read before each recording session, and interobserver reliability was checked continuously.
Time sampling was used for observation: for the first 10 minutes of each recording, the occurrence of the observational categories was noted on a coding form in 30-second intervals (see also Martens et al., 2014b). Three trained observers (the first author, a psychologist, and a master's student in educational psychology) rated the recordings.
Five behavioral categories were observed. The first category, affective involvement, is a client-communication partner category: sharing negative and positive emotions in a positive way so that they are perceivable to the client (for instance, rubbing the client's arm when he or she is showing discomfort). The other four categories pertain to emotions displayed by the client: very negative emotions (aggressive and self-abusive behaviors); negative emotions (negatively tense, bad-tempered, compulsive, or noncooperative behaviors); positive emotions (exploring and cooperative behaviors); and very positive emotions (laughing and behaviors expressing excitement). Table 4 lists examples of each client's behavior.
Training and reliability
Before data collection, the observers were trained to reach 80% interobserver agreement. Reliability was calculated for 25% of the formal observations (Barlow et al., 2009). Retraining was provided when the percentage of agreement was below 80%. Interobserver agreement ranged from 75% to 100% (SD = 7.98) across clients, with a mean agreement of 95% across observation categories. The mean agreement for affective involvement was 98% (range 80%-100%), for very negative emotions 97% (range 75%-100%), for negative emotions 94% (range 82%-100%), for positive emotions 92% (range 77%-100%), and for very positive emotions 91% (range 77%-100%).
The communication partners' acceptance of and commitment to the program were secured by consulting the partners repeatedly before, during, and after the intervention. The evolving process was evaluated during the coaching sessions, and the intervention outcomes were evaluated afterwards. The communication partners' perceptions of satisfaction and the usefulness and effectiveness of the intervention were measured with an adapted version of the Social Validity Scale (Martens & Janssen, 2011, according to Seys, 1987).
EFFECTS ON BEHAVIOR
Figures 1 and 2 show the observed occurrence of dyadic affective involvement (client--communication partner) and Figures 3 and 4 show the clients' very positive and (very) negative emotions. Each target behavior is depicted in two figures because of the different nature of settings (see Participants and Settings section): structured (school and daytime activities center) and unstructured (group home).
The left panel of each figure presents the mean occurrence of the target behaviors in the four intervention conditions: baseline, phase I, phase II, and follow-up. The right panel shows the actual data and trend line within the different conditions. Trend lines are not included if less than three observation sessions were videotaped during follow-up.
Figure 1 shows that affective involvement for Clarice appears to increase during phase I, showing an uptrend line. During phase II, the mean occurrence increases, although a downtrend line is displayed in the right panel. During follow-up, affective involvement decreases with a steeper downtrend line than in phase II.
For Bryan, affective involvement appears during phase I and increases during phase II with upward trends during both phases. During follow-up, affective involvement slightly increases but shows a downward trend line.
For Dewy, affective involvement increases considerably during phase I (showing a downward trend) and slightly decreases during phase II (including an upward trend). During follow-up, affective involvement further decreases but remains above baseline.
Figure 2 shows that affective involvement varies for Clarice during phases I and II and follow-up, remaining above baseline conditions. The trendlines during these conditions are all upward.
For Dewy, affective involvement increases during phases I and II, with uptrend lines, while it decreases but remains above baseline conditions during follow-up.
For Nick, affective involvement is observed during one session only in phase I (three times in session 13); it is absent during phase II but appears twice during one of the follow-up sessions.
Very positive and (very) negative emotions
Figures 3 and 4 present the occurrence of (very) negative emotions (negative and very negative emotions summed) and very positive emotions. In the right panel, the occurrence of (very) negative emotions is subtracted from the occurrence of very positive emotions. Hence, positive values indicate that very positive emotions predominate while negative values indicate that negative emotions predominate.
Figure 3 shows that Clarice's (very) negative emotions decrease in phase I, increase in phase II (remaining below baseline), and disappear during follow-up. Very positive emotions increase during phase I, decrease slightly during phase II, and increase considerably during follow-up. The right panel shows that the decrease of very positive emotions in phase II is due to a high peak of (very) negative emotions in session 19. Unlike the downward trend during follow-up, upward trends are shown during phases I and II.
For Bryan, (very) negative emotions steadily decrease during phases I, II, and follow-up. Very positive emotions increase during the three intervention conditions. The right panel shows an upward trend during baseline. Phase I shows a slight downward trend, and phase II a slight upward trend, followed by a strong upward trend during follow-up.
For Dewy, (very) negative emotions decrease in phase I, increase slightly in phase II (remaining below baseline), and disappear during follow-up. Very positive emotions increase during phase I, decrease slightly during phase II, and decrease during follow-up. The right panel shows a downward trend during baseline, suggesting that Dewy's (very) negative emotions increase during baseline. Upward trend lines are shown during phases I and II. A downward trend reappears during follow-up.
Figure 4 shows that for Clarice (very) negative emotions decrease during phases I and II and increase again during follow-up (remaining below baseline). Very positive emotions increase during phase I, decrease during phase II, and increase again during follow-up. The right panel shows an upward trend line during phase I, a stable trend line during phase II, and another upward trend during follow-up.
For Dewy, (very) negative emotions decrease during phase I, disappear during phase II, and reappear during follow-up (remaining below baseline). Very positive emotions increase during phases I and II, and decrease during follow-up. The right panel shows upward trends during phases I and II.
For Nick, (very) negative emotions disappear during phases I and II and reappear during follow-up but remain below baseline. Very positive emotions appear during phase I and increase during phase II and follow-up. The right panel shows stable trend lines for the conditions during baseline and phases I and II.
The ratings on the five-point Social Validity Scale (Martens & Janssen, 2011; 1 = low; 5 = high) were averaged across communication partners. The coaching was judged as "effective" to "highly effective" with preferences for individual coaching (M = 4.5) over team coaching (4.2). The communication partners judged their attitude and communicative skills (4.2) and the client's behavior (3.9) as "positively changed" to "very positively changed," with the best results being for Bryan (4.5 and 4.4). Affective involvement (3.9) was judged to be "somewhat easy to implement" to "easy to implement," with the highest scores being for Clarice (4.2) and the lowest scores for Nick (3.3). In the open comment box in the survey, the communication partners indicated that they could more easily share positive emotions than negative emotions. Sharing attention (3.8), recognizing and interpreting initiatives (3.8), and regulating intensity (3.7) were also judged to be "somewhat easy to implement" to "easy to implement." Sharing experiences (2.9) and sharing meaning (2.7) were judged to be "rather difficult" to implement, with the lowest scores being for Nick (both 2.3).
The present study shows that affective involvement and very positive emotions can be improved and (very) negative emotions can be decreased in persons who are congenitally deafblind by coaching their communication partners according to the IMAI. The communication partners learned to attune to their clients' interactive behaviors and emotions and to share emotions and meanings. Three clients showed an increase of affective involvement, and all clients showed an increase of very positive emotions and a decrease of (very) negative emotions. The intervention also proved to be effective for different communication partners and in different settings and interactional situations. These promising results agree with those of a previous study by Martens et al. (2014b).
For Dewy, affective involvement improved at school immediately after the onset of the intervention (see Figure 3), with high peaks of affective involvement when her communication partner could focus her on one topic
at a time. This shows how important it is for communication partners to be able to share a focus on a third element outside the dyadic space with the person with congenital deafblindness. This communication pattern is the most difficult in establishing social interaction with people with congenital deafblindness (Janssen & Rodbroe, 2007).
It is remarkable that affective involvement only increased during follow-up in Clarice's case (group home; Figure 2). Her communication partners continued to apply the IMAI skills learned during coaching (for instance, maintaining the flow of interaction by lowering tempo and repeating initiatives; making use of opportunities for communication by mutually choosing between activities; sharing emotions by persevering in sharing very positive emotions visually and tactilely).
For Bryan, despite the absence of affective involvement during baseline, the upward trend (see Figure 3, right panel) suggests that his emotional behavior had already improved during this condition. Spontaneous improvement of behavior before the intervention begins is a quite common phenomenon in single-case research (Barlow et al., 2009). One explanation for this sudden improvement may be that the communication partners had already- unintentionally-adapted their behavior to Bryan's behavior during baseline.
For Nick, no clear effects for affective involvement were found because the frequency of affective involvement was too low to draw conclusions. But his emotional behavior improved: (very) negative emotions disappeared completely and very positive emotions appeared during phase I and further increased during phase II (see Figure 4). Nick's communication partners indicated that during intervention they could more easily attune their behavior to Nick's behavior, but sharing experiences and sharing meaning (phase II of the intervention) remained difficult due to the few and brief moments of mutual contact. A possible explanation for this is that he has more difficulty making contact because he had few opportunities to develop social communicative relationships during his 40-year stay in group homes run by two different organizations specialized in intellectual disabilities. In those settings, there was a lack of deafblind-specific education involving communicational development within social engagement. A second possible explanation could be that the interaction situations were not the best for eliciting affective involvement with Nick. During evaluation, communication partners reported that affective involvement occurred in other interaction situations such as drinking a beer in a pub. Such situations could be chosen for future interventions with Nick.
As noted above, we could have made a more optimal selection of interaction situations to improve affective involvement.
Second, the small number of clients involved restricts generalization to other persons who are congenitally deafblind. Replication of the intervention is therefore recommended (Barlow et al., 2009).
Third, the present study only allows conclusions regarding the effectiveness of the intervention program as a whole on the affective involvement and emotions of the clients. We cannot draw any conclusions about which categories of communication partner behaviors or which specific communication partner behaviors that were assessed during coaching specifically brought about the observed changes in the affective involvement and emotions of the clients. This topic could be explored for future studies directed at increasing the effectiveness of such interventions.
IMPLICATIONS FOR PRACTICE
The IMAI is based on the assumption that focusing on affective involvement during interaction should precede focusing on affective involvement during communication. The communication partners in this study confirmed that this approach was effective, which supports the practicality of the two-phase IMAI-approach.
The social validity outcomes provide useful information on how to maximize the impact of the intervention on practice (Barlow et al., 2009). For example, the communication partners experienced more difficulties in phase II of the intervention: they found sharing experiences and meaning rather difficult to implement. This finding suggests that the intervention could possibly be improved by reducing phase I and extending phase II.
Consistent with previous studies (Janssen et al., 2003; Martens et al., 2014b), our findings indicate that affective involvement can be improved during intervention, but that it is difficult to maintain affective involvement over time. This is not surprising, because it is a constant challenge for the hearing and seeing communication partners to imagine the experience of the person who is congenitally deafblind (Miles, 1999). Affective involvement, though, serves as the most informative and helpful route to understanding another person's emotions. Therefore, we recommend that communication partners of people who are deafblind be constantly coached in fostering affective involvement or at least be coached at regular intervals.
This study was funded by Royal Dutch Kentalis. We would like to acknowledge the contributions of Marijse Pol and Maartje Hofman at the University of Groningen; and the staff members of Kentalis Rafael (school), the Department of the Daytime Activities Center, and the Orthopedagogics Center at Royal Dutch Kentalis, Netherlands.
Barlow, D. H., Nock, M. K., & Hersen, M. (2009). Single case experimental designs: Strategies for studying behavior change (3rd ed.). Boston: Allyn & Bacon.
Chen, D., Klein, D., & Haney, M. (2007). Promoting interactions with infants who have complex multiple disabilities: Development and field-testing of the PLAI curriculum. Infants & Young Children: An Interdisciplinary Journal of Special Care Practices, 20(2), 149-162.
Diamond, L. M., & Aspinwall, L. G. (2003). Emotion regulation across the life span: An integrative perspective emphasizing self-regulation, positive affect, and dyadic processes. Motivation and Emotion, 27(2), 125-157.
Janssen, M. J., Riksen-Walraven, J. M., & van Dijk, J. P. M. (2003). Contact: Effects of an intervention program to foster harmonious interaction between deaf-blind children and their educators. Journal of Visual Impairment & Blindness, 97(4), 215-229.
Janssen, M. J., & Rndbroe, I. (2007). Communication and congenital deafblindness. Contact and social interaction. Sint-Michielsgestel, the Netherlands: Danish Resource Center on Congenital Deafblindness and Viataal.
Martens, M. A. W., & Janssen, M. J. (2011). Social Validity Scale. Unpublished manuscript.
Martens, M. A. W., Janssen, M. J., Ruijssenaars, A. J. J. M., Huisman, M., & RiksenWalraven, J. M. (2014a). Introducing an intervention model for fostering affective involvement with persons who are congenitally deafblind. Journal of Visual Impairment & Blindness, 108(1), 29-41.
Martens, M. A. W., Janssen, M. J., Ruijssenaars, A. J. J. M., Huisman, M., & Riksen-Walraven, J. M. (2014b). Intervening on affective involvement and expression of emotions in an adult with congenital deafblindness. Communication Disorders Quarterly, 35(3), 1-10.
Miles, B. (1999). What is communication? In B. Miles & M. Riggio (Eds.), Remarkable conversations. A guide to developing meaningful communication with children and young adults who are deafblind (pp. 10-19). Watertown, MA: Perkins School for the Blind.
Rodbroe, I., & Souriau, J. (1999). Communication. In J. M. McInnes (Ed.), A guide to planning and support for individuals who are deafblind (pp. 119-149). To ronto, London: University of Toronto Press.
Seys, D. M. (1987). Kwaliteit van zorg: zorg voor kwaliteit [Quality of care: Care for quality]. Unpublished doctoral thesis. Nijmegen, the Netherlands: University of Nijmegen.
Marga A. W. Martens, Ph.D., researcher, Kentalis Deafblindness Center of Excellence, Royal Dutch Kentalis, Petrus Dondersplein 1, 5271 AA Sint-Michielsgestel, the Netherlands, and Department of Special Needs Education and Youth Care, University of Groningen, Grote Rozenstraat 38, 9712 TJ Groningen, the Netherlands; e-mail: <firstname.lastname@example.org>. Marleen J. Janssen, Ph.D., professor, Department of Special Needs Education and Youth Care, University of Groningen, and Kentalis Deafblindness Center of Excellence, Royal Dutch Kentalis, Sint-Michielsgestel, the Netherlands; e-mail: <email@example.com>. Wied A. J. J. M. Ruijssenaars, Ph.D., professor, Department of Special Needs Education and Youth Care, University of Groningen, Groningen, the Netherlands; e-mail: <firstname.lastname@example.org>. Mark Huisman, Ph.D., assistant professor, Department of Sociology, University of Groningen, Groningen, the Netherlands; e-mail: <email@example.com>. J. Marianne Riksen-Walraven, Ph.D., professor, Department of Developmental Psychology, Radboud University, Montessorilaan 3, 6525 HR Nijmegen, the Netherlands; e-mail: <firstname.lastname@example.org>. Please address all correspondence to Dr. Martens.
Table 1 Characteristics of the clients. Client, gender, Sensory age Diagnosis impairment Clarice, female, Congenital Visually 26 rubella impaired and deaf Bryan, male, 15 Marden Walker Visually and syndrome hearing impaired Dewy, female, 15 Severe Visually and shortage of hearing oxygen at impaired birth Nick, male, 54 Congenital Blind and rubella deaf Client, gender, Communicative Behaviors difficult age functioning to interpret and attune to Clarice, female, Pushing, pulling, Agitation, inactive, 26 touching, gesturing, self-stimulation understanding some conventional signs, augmentative communication (full-color pictogram) Bryan, male, 15 Pushing, pulling, Protesting, non touching, gesturing, -cooperative, vocalizing comfort self-stimulation and discomfort, understanding some concrete objects of reference, reacting to name calling Dewy, female, 15 Speech, gesturing, some Aggressive, conventional signs, reactive, augmentative compulsive, communication easily (drawing, black-and distracted -white pictogram) Nick, male, 54 Pushing, pulling, Aggressive, touching, gesturing, agitation, self some objects of -injurious and reference, dominant understanding two behaviors conventional signs (pushing and (drinking, finished) pulling) Client, gender, Physical age functioning Clarice, female, Spastic 26 paralysis of the legs Bryan, male, 15 Motor and balance problems Dewy, female, 15 Epilepsy Nick, male, 54 Diabetes at older age; high blood pressure Table 2 Characteristics of the clients' communication partners. Age of Years of work communication experience Number of partners with clients communication Client partners Range M (SD) Range M (SD) Clarice 4 21-41 1.5-10 30.25 (8.46) 5 (4.1) Bryan 3 46-54 9-25 50.67 (4.17) 14.67 (7.3) Dewy 5 35-55 4-25 44.6 (8.27) 16.4 (8.88) Nick 4 22-49 1-25 35 (12.94) 9.25 (9.81) Total 16 21-55 1-25 39.75 (11.93) 11.87 (9.32) Client Function Clarice Caregiver (3) Caregiver assistant (1) Bryan Teacher (2) Teacher assistant (1) Dewy Teacher (2) Caregiver (1) Caregiver assistant (2) Nick Caregiver assistant (4) Total Teacher (4) Teacher assistant (1) Caregiver (4) Caregiver assistant (7) Table 3 Target questions for coaching, intervention aims, and interactional situations in the different settings per client. Intervention aims Target questions Client for coaching Phase I: Interaction Clarice (a) When does she * Recognizing and express negative interpreting emotions? positive and (b) How can we negative emotions regulate agitation * Regulating negative and self-stimulation? emotions (c) How can we * Sharing positive establish enjoyment emotions and mutual * Increasing mutuality interactions with her? and enjoyment Bryan (a) How can we be * Interpreting and explicit without attuning to different being harsh? states of emotions (b) How can we * Evoking positive enhance his positive emotions emotions? * Sharing negative (c) How can we share emotions different states of emotions? Dewy (a) How can we * Providing improve predictable independent acting interactions without losing * Reducing negative control of the tension situation? * Increasing positive (b) How can we emotions establish mutual * Sharing positive trust and have a emotions joyful contact? Nick (a) How can we prevent * Reducing negative him from having emotions+ negative emotions? * Regulating negative (b) How can we emotions regulate his negative * Increasing mutuality emotions? and fostering (c) How can we positive emotions promote mutuality and foster positive emotions? Intervention aims Phase II: Settings and Client Communication interactional situations Clarice * Sharing Group home: leisure positive and household and activities; daytime negative activities center: emotions unstructured and structured * Sharing activities positive interactions more often Bryan * Expressing School: eating intentions yoghurt in the clearly and afternoon interpreting his emotional behaviors * Sharing joyful experiences Dewy * Sharing School: making positive music, arts and emotions crafts more often * Reducing Group home: leisure negative and household experiences activities Nick * Being an Group home: having interesting afternoon coffee, and having dinner enjoyable partner * Sharing meaningful experiences * Sharing positive emotions Table 4 Examples of emotional behaviors per client. (Very) negative Client emotions Negative emotions Clarice Screaming, pinching, Ignoring or pushing and hitting communication communication partners away, partners vocalizations of discomfort Bryan Persistently biting Walking away, bib, falling onto the pulling on floor, stiffening communication body partner, vocalizations of discomfort Dewy Spitting, yelling or Clinging to screaming, communication throwing objects partner (clothing, hair), restless moving, repeating questions Nick Stamping feet, hitting Pulling and head, vocalizations pushing, of discomfort increasing tempo, increasing muscle tension Very positive Client Positive emotions emotions Clarice Looking at and touching Smiling, vocalizing communication comfort partner, affirming initiatives Bryan Looking at Laughing, flapping communication arms or jiggling partner, participating legs in acting, exploring objects Dewy Friendly smiling, making Laughing, joking eye contact, affirmatively responding Nick Repeatedly and softly Brief moments of touching smiling, asking communication for head partner, vocalizations caresses of comfort