The Break (and Summary) in Solution-Focused Brief Therapy: Its Importance and Client Experiences *

Journal of Solution-Focused Brief Therapy — 1

The end-of-session break was nominated by Steve de Shazer as an essential component of Solution-Focused Brief Therapy; yet it is an aspect often ignored or eschewed by many Solution-Focused therapists. After reviewing the history and development of the break and the end-of-session message, this paper argues that the recency effect in cognitive psychology highlights the importance of how therapists conclude their sessions and that, if the way the session ends is important, perhaps it warrants some time to consider and plan. An qualitative study suggests that the break is not only useful to therapists but that clients report that the therapist taking a break and then providing a summary message enhances the benefit of the session for them. Limitations of the study are discussed and it is suggested that the findings contribute to discussion and ongoing evolution of the Solution-Focused approach.
Steve de Shazer, one of the founders of Solution-Focused Brief Therapy, and someone who claimed that he used the Miracle Question "almost always in the first therapy session" (de Shazer, 1999, p.1), is reported to have said, "If I was forced to make the choice … I would give up the Miracle Question before I would give up taking the break!" That is quite a strong, and perhaps provocative, statement, particularly given that informal, anecdotal enquiry suggests in a systematic way (de Shazer et al., 1986), the authors describe the typical functioning of the Milwaukee team. "After 30 to 40 minutes the therapist excuses himself to consult with the team … After an intermission of 10 minutes or less, the therapist returns and gives the formal intervention." (p. 216). The intervention (the "message from the team") was seen as the primary agent of change.
Lipchik and her colleagues comment that, later, as Solution-Focused Brief Therapy developed as a definable approach, "the interview rather than the intervention became the primary agent of change" (Lipchik et al., 2012, p.9). As will be shown below, the message given by the therapist after the break changed in nature but was still viewed as important. Thus, the break began to be seen primarily as a chance to think carefully and prepare the end-of-session message or summary. Consistent with this, it began to make sense to take a break to prepare the summary message even if the therapist did not have a team to consult. Cade (2001, p. 203) observed, "Solution-focused therapists typically take a break before ending each session, whether or not there is a team behind a one-way mirror with whom to consult." Turnell and Hopwood (1994) suggest that the time before the break is where the therapist asks questions and listens, but the client talks. After the break, they suggest, the therapist talks and the client listens. They describe the typical therapist explanation as: I like to take a break since you've said a lot that is very important and before I give you my [the team's] thoughts/some feedback, I want to spend a few minutes considering everything you have told me. (Turnell & Hopwood, 1994, p. 48) In 1997, de Shazer and Berg, proposing a "research definition" of Solution-Focused Brief Therapy, posit four "characteristic features" of the approach. They nominate the Miracle Question and Scaling Questions as the first and second feature. The other two "characteristic features" are: (3) At some point during the interview, the therapist will take a break.
(4) After this intermission, the therapist will give the client some compliments which will sometimes (frequently) be followed by a suggestion or homework task (frequently called an 'experiment'). (de Shazer & Berg, 1997, p. 123).
That is, they saw the break and the subsequent feedback to the client as essential defining characteristics of Solution-Focused Brief Therapy. In their review of outcome research on Solution-Focused Brief Therapy, Gingerich that many therapists who describe themselves as Solution-Focused do not routinely take a break towards the end of their therapy sessions.
Thus, it might be useful to reconsider the importance of "the break" (and the subsequent end-of-session summary to the client) in Solution-Focused Brief Therapy and to ask what the experience of our clients is about the break.

History and Development of The Break
One of the developments that characterised the early Family Therapy (and Brief Therapy) movement was the use of one-way mirrors and observing teams. Weakland, Fisch, Watzlawick, and Bodin (1974) write that the use of a one-way viewing screen was part of their practice from 1967, initially using a therapist and one observer and later preferring a team of observers behind the mirror. Minuchin (quoted by Lappin, 1988) explains that, as part of developing his family therapy approach, "We broke through a wall in our treatment room and put in a one-way mirror and began to observe one another …" (p. 225). Taking a break began simply as part of doing therapy with a team and a one-way mirror. Pragmatically, the therapist needed to take a break in order to consult with the team behind the mirror so as to benefit from their observations.
As more strategic brief and family therapy approaches developed, the break and the team became part of the STRATEGY of therapy. In Strategic Family Therapy (Nichols & Schwartz, 2001;Papp, 1980) and MRI Brief Therapy (Weakland et al., 1974), the team, the break and "the intervention" were used to attempt to influence the client and/or interrupt a systemic pattern. Cade (1980) reported using "contrived team conflict" during the break as a way to break therapeutic impasses. Selvini Pallazoli and her colleagues saw the purpose of the "intersession break" was for the team to agree upon a comprehensive systemic hypothesis about the development and function of the symptom, leading to "the intervention" which usually offered the family this hypothesis as a systemic explanation of their predicament (Tomm, 1984).
Observation of therapy by a team and consultation with that team were essential parts of the early days of what was to become Solution-Focused Brief Therapy. Four of the original Milwaukee team remember that, initially, "The interviews were conducted in Steve and Insoo's living room by one person while a team observed." (Lipchik, Derks, Lacourt, & Nunnally, 2012, p.5). Lipchik and her colleagues recall that, "… after the mirror was installed …", consultations between the interviewer and the members of the team became commonplace.
In the seminal paper that first presented Solution-Focused Brief Therapy and  referring to a "thinking pause". They comment that many practitioners will take a break and that some will leave the room while others may pause but remain in the room. Thus, they seem to be retreating from de Shazer and Berg's stance that taking the break is a distinguishing characteristic of the approach. Nonetheless, they still suggest that taking a break is pragmatically useful for the therapist purely as providing space to think.

The End-of-session summary
In the early days of family therapy, particularly strategic family therapy, the interview was largely seen as a process of gathering information which would be synthesized by the team during the break and the therapist would then return to the room to "deliver the intervention" (Nichols & Schwartz, 2001;Papp, 1980). Haley (1993) described "strategic therapy … as a name for the types of therapy where the therapist takes responsibility for directly influencing people" (p. 17) and said that the therapist has to design interventions to achieve those goals. Clearly, "the intervention" was viewed as the primary tool for achieving change. Weakland and Fisch (2009) describe a process in which the therapist uses the end-of-session message to deliberately reframe behaviour, to instruct clients to do certain things and paradoxically to advise clients against making changes. In describing the operation of the MRI Brief Therapy Center, Weakland et al. (1974) discuss various aspects of the message (frequently "the message from the team") that are specifically designed to influence behaviour. Watzlawick (2009) describes the MRI team's interventions as "injunctions" or "prescriptions" and nominates direct behaviour prescriptions, paradoxical interventions (also called symptom prescription) and positive connotations as the three categories of intervention.
Given that Solution-Focused Brief Therapy was a direct descendant of the MRI Brief Therapy approach (Cade, 2001;de Shazer et al., 1986), it is not surprising that the prevailing view of the message delivered by the therapist following the break as being "the intervention" was carried forward. As noted earlier, de Shazer et al. (1986) describe that, following the break, "the therapist returns and gives the formal intervention." (p. 216). The intervention (the "message from the team") was still seen as the primary agent of change. Discussing the functioning of the team behind the one-way mirror, de Shazer and Molnar (1984) emphasise the planned, directive and strategic nature of the intervention, "the intervention delivered after the consulting break are usually phrased in terms of 'we …' rather than 'I'." (p. 297).
de Shazer and Berg (1997) write that "the task" will often be framed as "an and Eisengart (2000) similarly nominate seven components necessary for classification as Solution-Focused Brief Therapy, including "(6) A consulting break, and (7) a message including compliments and task" (p. 479). Ten years after de Shazer and Berg saw it as characteristic, the break is still seen as a normal and helpful part of the Solution-Focused therapy process. de Shazer and Dolan (2007) discuss taking a break as if it is still an expected part of the approach and again as an opportunity to think about what the client has said and frame the summary message. They assert that, even if there is not a team, "the therapist will still take a break to collect his or her thoughts, and then come up with compliments and ideas for possible experiments" (p. 11). De Jong and Berg (2008) similarly comment that, "When interviewing clients in a solution-focused manner, practitioners generally take a break of 5 to 10 minutes before giving clients feedback. This will have definite benefits for you and your clients" (p.115).
Eve Lipchik stressed the importance of taking a break in order to think carefully about what the therapist plans to say to the clients.
Those of us who are accustomed to taking breaks to formulate a closing message … usually have stories to tell about the occasions we decided to forgo the break to save time. (Lipchik, 2002, p.100) I would urge those therapists who feel uncomfortable about shortening their sessions to reconsider. The benefits clients get from a carefully designed summation message may well outweigh the extra 10 minutes of conversation. (Lipchik, 2002 p. 103) Lipchik's clear support for taking a break was based on her experience that a well-thought-out summary message is of benefit to clients and that a wellthought-out summary message requires some space to consider and plan it. Macdonald (2007) makes a similar point.
It is a common experience that appropriate responses occur to us just after we have left a situation … It is in the nature of human interaction that we are affected by one another's emotions [and] when clients are anxious and unable to reflect, we will be affected by this … Leaving the room … allows us the cognitive space to think more clearly about their situation and about what comments will be most useful." (p. 25) Macdonald comments that, if in a situation where taking a break is impractical, he simply asks the clients to wait while he takes a few moments to think about everything they have said. This suggestion accords with Ratner, George, that you can tell us (me) next time, what happens in your (life, marriage, family, or relationship) that you want to continue to have happen." (p. 298) All of a sudden, the task was NOT the therapist prescribing behaviour but was asking clients to notice particular (already existing) positive behaviour. Lipchik et al. (2012) comment that, as the work of the Milwaukee team became more obviously Solution-Focused, there was more emphasis on what clients were doing that was working and the message from the team became "designed to reinforce the strengths and resources of clients discovered during the interview, as well as to stimulate more options for solutions between sessions" (p. 9). Lipchik (2002) comments specifically that the terms "intervention" and "task" seem incongruous with the cooperative stance of Solution-Focused Brief Therapy. She deliberately uses the terms "Summation message" and "suggestion", emphasizing that the summary allows clients to feel that they have been heard.
There seems to be some ongoing tension between those who see the task as instrumental and those who see it as primarily observational.
Macdonald (2007) talks explicitly about suggesting a task and his list of possible tasks includes "keep doing what's working", "do something different", a prediction task and a "pretend the miracle has happened" task. He goes further saying that sometimes a therapist might suggest the client carry out a specific piece of behaviour - either one that emerged from the client or one that occurs to the therapist (although he acknowledges that clients rarely act on the latter kind of suggestion!). He adds that such more direct suggestions may helpfully be framed as "an experiment". On the other hand, Ratner et al. (2012), who also prefer the term "suggestion", suggest that simple "noticing suggestions" are now most common.

The Primacy-Recency effect
This concept was originally observed by German psychologist, Hermann Ebbinghaus, in his 1913 experiments on memory (Crowder, 1976). When asked to recall a list of items in any order (free recall), people tend to begin their recall with the end of the list, recalling those items best (the recency effect). This is intuitively not surprising. However, this research shows that the NEXT most likely items to be recalled are the first few items. These are recalled more frequently than the middle items (the primacy effect). Put simply, if you are presented (verbally) a number of items, you are most likely to remember the last four or five items, and you are NEXT most likely to remember the first experiment" and it seems that "experiment" may sound more benign than "task". However, it is interesting that, before the name Solution-Focused Brief Therapy had been coined, de Shazer (1974) wrote about "interventions" and suggested, "Particularly useful is a kind of cross between paradoxical intervention and role-playing that might be called an 'experiment'." (p.22). As late as 2007, the founders of the approach and their colleagues (de Shazer & Dolan, 2007) still use the terms "experiments" and "homework assignments" apparently interchangeably. The term "homework assignment" seems to carry the connotation of a prescription.
Thus, the therapist's message following the break was initially seen quite instrumentally as an intervention. However, there was some indication from early in the development of the approach that some of the emphasis in thinking about the intervention was changing. Lipchik et al. (2012) report that the intervention message in early Solution-Focused Brief Therapy, but also in the Brief Family Therapy that preceded it, always began with compliments to the clients. They suggest that the use of compliments was probably a reflection of the influence of the Milan group's "positive connotation" (Palazzoli, Boscolo, Cecchin, & Prata, 1980); however, they note that it moved beyond that to a recognition that complimenting clients for what they had achieved encouraged cooperation and made it more likely that clients would return.
As mentioned above, in their delineation of the "defining characteristics" of Solution-Focused Brief Therapy, de Shazer and Berg (1997) specify that "the therapist will give the client some compliments" (p. 123). Campbell, Elder, Gallagher, Simon, and Taylor (1999) write extensively about crafting compliments to include in the end-of-session message. Ratner et al. (2012) see the major purpose of the message as building a clear expectation of beneficial change and that this is done by reminding the client about "qualities and capacities the client brings to his life that could be the basis of progress … and actions the client has taken in the direction of the 'best hopes'."(p. 142).
de Shazer and Berg (1997) specify that the end-of-session message (following the break) will include compliments, followed by a task, or "experiment". Tasks are still seen as something the therapist designs or prescribes. The use of the language "homework assignment" clearly reflects this view and, as mentioned above, as late as 2007 this nomenclature is still current (de Shazer & Dolan, 2007).
However, there are hints of something different much earlier. As early as 1984, de Shazer andMolnar (1984) describe what was then called the "Formula First-Session Task", Between now and next time we meet, we (I) want you to observe, so denly think, 'Oh … I should have said …!'". However, the question remains of how clients view their therapist leaving them for eight-to ten-minutes near the end of each session and then returning to give a summary and suggestion.
Therefore, the research question for this study was: How helpful do clients find the end-of-session break and the subsequent summary message in Solution-Focused Brief Therapy sessions and how do they describe its usefulness?
The participants in this primarily qualitative exploration were 33 adult clients attending therapy in a suburban area in Australia. This service is a "generalist adult counselling service" with clients referred by medical practitioners, community organisations and self-referral. Clients were approximately two-thirds women, between 20 and 50 years of age, and presenting with a range of concerns including depression, anxiety, post-natal depression, difficulty managing their children, relationship concerns, bereavement and effects of trauma.
Participants were recruited using non-random convenience sampling (Marshall, 1996). Every client who attended a third session during the period of data collection was asked if she or he would be willing to answer a brief question at the end of the session. The third session was chosen deliberately. The rationale for this choice was that, by the end of the third session, clients would have had three experiences of sessions in which the therapist took a break and returned to give a summary message. Therefore, it was assumed that they would have become "used to it" and would not be commenting purely on something they had experienced as novel.
Further, Solution-Focused therapists' experience is that the whole course of therapy tends to be brief. Iveson (2002) nominates three to five sessions as typical and Gingerich and Eisengart (2000) describe "usual" as fewer than six sessions. Thus, collecting this data at the end of the third session meant the study was less likely to miss contributions from any clients who terminated therapy after only a few sessions.
In fact, no client ended therapy prior to the third session during the period of data collection. No client refused the request to participate.
At the end of the third session, clients were asked, Is it okay if I ask you something as part of some research I am doing? (If they said "yes") You remember that, each time we have met, I have taken a short break towards the end of our session, thought about what you'd said to me and then come back and shared some ideas with you. You know that I find it helpful to take that break; however, I'm wondering how helpful my taking that break has been FOR YOU. So, on a scale of 0 to 10, where zero is "Not at all" and 10 is "Extremely four or five items. More recently, the phenomena of primacy and recency have been confirmed by psychologists exploring persuasion - the things a person hears LAST, then FIRST, are more likely to have an ongoing persuasive effect. Costabile and Klein (2005) demonstrated that evidence and arguments presented towards the end of a trial were more likely to be recalled by jurors and so were more likely to influence their decisions and Y. Li and Epley (2009) showed that items seen towards the end of a presentation were more likely to be recalled favourably. Panagopoulos (2011) demonstrated that the messages presented in the closing stages of a political campaign are more persuasive, but also that messages from the beginning of the campaign retain more persuasive value than those in the middle. C. Li (2010) found a significant primacy effect in examining the recall and impact of television commercials aired towards the beginning of a program and those aired towards the end.
The primacy and recency research suggests two things: how therapy begins is important and how therapy ends is important. The final few minutes of a therapy session are more likely to retain some impact (simply - they are more likely to be what the client remembers!). Therefore, if therapists want to make the most of this impact, it makes sense that they should devote some time and thought to what they are going to say at the end.
Taking a break and physically leaving the room punctuates the final five minutes or so of the session (Turnell & Hopwood, 1994). It becomes a discrete "phase" of the conversation and so is more likely to have greater impact on the client.

Survey of client's experience of the break and summary
The first author routinely takes a break towards the end of all her therapy sessions despite never working with a one-way mirror or a team. She finds it helpful to have the opportunity to reflect on what clients have said during the session and to have some space to plan her end-of-session summary message. Her end-of-session summary messages, following Lipchik (2002), simply consist of compliments "designed to reinforce the strengths and resources of clients discovered during the interview." (Lipchik, 2002 p. 9) and are frequently followed by a suggestion that the client "notice between this session and the next whatever it is that they do that moves them one step up the scale" (Cade, 2001, p. 205).
Her impression is that her summary messages are more comprehensive and seem more powerful because of the opportunity to stop and think. "No longer do I have the experience that they are half-way out the door and I sud-

The break has personal, practical benefit to the client (14 respondents)
A number of clients commented first on some personal, practical benefit of the break in the therapy process. Most frequent (11 respondents) was some appreciation of the opportunity to pause, clear their head and enjoy quiet. ♦ "It gives me time to check Facebook and check for messages".
♦ "When I've been emotional during the session, the break enables me to take a breath and collect myself".
♦ "The break gives me a chance to clear my head. I go outside and have a smoke".
♦ "I like having a moment without interruption at the end which lets me feel, have some space and collect my thoughts".

The break and summary allow the client to think/reflect (14 respondents)
The therapist often explains the break to clients by saying something like, "… I will take a break and think about what we have talked about …" (Korman, 2004). In this survey, a number of clients replied that the break offered them an opportunity to think about the session as well.
♦ "I like that it also means that we as clients can reflect about what we've said in the session".

♦ "I like the summary because it makes me think about what I said in the session".
helpful", how helpful TO YOU has it been that I have taken a break towards the end of each of our sessions?
Clients' responses between 0 and 10 were recorded. The initial numerical (scale) responses were simply averaged. While not a stringent statistical test, this provided an indication of the strength of subjective perception underlying the subsequent qualitative data. Then clients were asked, "So, what was it about my taking the break that makes it x on the scale (whatever number the client had said)? What has been helpful for you about me taking a break?" Client responses were recorded verbatim, but without identifying data, then a qualitative analysis of predominant themes was performed. The procedure described by Braun and Clarke (2006) for "thematic analysis" was adopted to analyse the data. Initial coding of the data was conducted, using an inductive process which sought to generate codes from the data itself rather than from any preconceived theory or system. Coding produced an initial list of tentative themes, which were then reduced through grouping. Braun and Clarke (2006) describe "thematic analysis" as a flexible approach that is "a relatively easy and quick method … to do" (p. 97). Given that this study had quite a limited research question and that the data from each respondent was only a few sentences, a more exhaustive qualitative analysis approach was not considered necessary. Nonetheless, "thematic analysis" provided a way of analysing the responses in a systematic manner.

A. Numerical scale
In answer to the question, "How helpful TO YOU has it been that I have taken a break towards the end of each of our sessions?", using a Likert-type 0-10 scale, the responses ranged between 5 and 10. The mean of the 33 responses was 8.6 and the mode was 10. In fact, one third of all respondents (11 out of 33) rated the usefulness to them of the therapist taking a break as being 10. The frequencies of the various responses are shown in Figure 1.

B. Qualitative data
Following the initial inductive process of identifying themes in the data and then combining or grouping these where it seemed they overlapped, six themes emerged for describing clients' experience of the helpfulness of the break and the ensuing summary. summary that was to come. ♦ "You taking the break, makes me more focused on the summary afterwards and that it gives me food for thought " ♦ "Having had the break helps me to think more about what you say when you come back ." ♦ "The break makes me consider what this all means and reflect on what you may be going to say during the summary."

The break and summary help extend impact of therapy (4 respondents)
Perhaps related to the idea that the break enhances the client's expectation of the end-of-session summary is the observation from a number of clients that the break and summary gave them something to take away. ♦ "It makes it easier for me to remember the session and to be able to tell my husband about it and that helps us to talk about it at home." ♦ "You having taken the break and then giving a summary makes me feel that I take something away from the session." ♦ "When you cement things through the summary, it makes me continue to reflect on things in the car on the way home".

Discussion
As mentioned above, anecdotal experience suggests that many (if not most) Solution-Focused therapists do not regularly take an end-of-session break. So, for example, Iveson describes himself as "someone who deliberately doesn't take a break (unless there are other people watching the session)". However, this research suggests that clients almost overwhelmingly describe the therapist taking a break and then returning to give the end-of-session summary as positive and helpful. As well as mentioning practical benefits (calming down, checking messages, etc.), a number of respondents commented on the usefulness of the break providing them with an opportunity to think and reflect. This supports Cade's (2001, p. 204) suggestion that "A break also gives the client time to think about the session … Also, clients often come to therapy expecting to be probed and exposed in areas of their greatest doubts or emotional sensitivity and/or to be 'told the error of their ways'. The break brings the realisation that this is not about to happen." ♦ "My husband and I were able to use that time to reflect together on what we'd said during the session".
3. The summary helps the client feel affirmed (12 respondents) A number of clients commented on the "positive" summaries and how they "felt good". There was also a sense that the compliments confirmed for the client that she/he was already heading in the right direction.
♦ "It's nice to hear someone say the positive things and affirm that I'm on the right track." ♦ "The positive comments in the summary forces me to reflect on things in a different way." ♦ "It affirms what I think or where I need to go from here." ♦ "You're good at saying what I did well." ♦ "I like that you reflect back what I maybe did not realise myself." ♦ "It makes me feel that I'm coping better than what I had thought."

The break increases the client's confidence in therapy (6 respondents)
There was a strong sense that the therapist "bothering" to take a break and think indicated to the client that the therapist was serious or was taking the client seriously. Rather than thinking that the therapist leaving the room for a "think break" was idiosyncratic or strange, clients felt affirmed in the knowledge that the therapist was actually taking time to think about what they had said.
♦ "I prefer this to my previous psychologist who did not take breaks and it made me feel that he gave less thought to the session or what I wanted or needed." ♦ "You taking a break and then giving the summary makes me feel that the counselling is going somewhere".
♦ "The break and the summary make me feel that you have really thought about what I said".

The break enhances the client's experience of the summary (5 respondents)
A number of clients reported that the break made them more focused on the discounted. A larger study, surveying clients from a number of therapists and with the data collected by independent researchers, would be required in order to remove this potentially confounding factor. Related to this is the question of generalisability and it might be argued that the results of a small study are not able to be generalised. Myers (2000) observes that a frequent criticism of qualitative research is that its reliance on small samples makes generalisation impossible. However, she asserts that generalisability (in terms of probabilities) is not a goal of qualitative research. Horsburgh (2003) suggests that qualitative research aims to add to understanding of a (subjective) phenomenon and that results may offer suggestions for understanding similar phenomena in other contexts. These findings that clients experience the break and summary as helpful and positive do not, of course, imply that therapists who, for whatever reason, do not take a break are giving their clients a negative experience. Knutsson, Norrsell, Johansson, Öhman, and Ericson (1998), in an evaluation of their clinic in Sweden, report that some clients appreciated the break as a chance to reflect and all their clients experienced the message as positive (however, they do not explore that further). In contrast, (Henfrey, 2010) reports that clients who had reported improvement following Solution-Focused Brief Therapy largely did not endorse the statement, "the therapist taking a break towards the end of the session was useful to me" (p. 30). Shennan and Iveson (2012) report that this latter finding prompted them to discontinue taking a break in their sessions.
In 1997, de Shazer and Berg named taking a break as one of only four defining characteristics of the approach. Almost twenty years later, it is clear that the approach has evolved (and is evolving), with different practitioners having different emphases. In the present study, the specificity of clients' accounts of how they found the break and the summary helpful offer some insight into how they experience a particular aspect of one flavour of Solution-Focused Brief Therapy in one particular place. What matters is not that this understanding can be generalized to all other practitioners of the approach. Rather, our hope is that this understanding both allows us to be clearer about what we actually do and also contributes to ongoing discussion within the community of Solution-Focused practitioners as the approach continues to evolve. Korman (personal communication, 2011) recounts an experience where a client commented, "I've seen lots of psychiatrists and therapists before, but none of them have really cared about me the way you do!" When asked what gave her this impression, the client replied, "None of them took a break to think about what I had said!" Compliments have been seen as a central aspect of the end-of-session message in Solution-Focused Brief Therapy (Campbell et al., 1999;de Shazer, 1988). Compliments are reflections from the therapist (or from the team) of strengths, resources, exceptions, and "things I can see you are already doing to move towards [your preferred solution]". That many clients commented that the end-of-session summary (based on compliments) helps them feel affirmed is perhaps not surprising. However, the frequent comments of clients about feeling heard or feeling valued simply by the experience of their therapist taking a break are illuminating. Believing that the therapist was taking them seriously and then feeling validated that they were already on the right track together seem to enhance the client's positive view about the whole therapy process. Cade (2001) suggests that the break "heightens [the client's] sense of anticipation about what the therapist's (and, where relevant, the team's) opinion and suggestion is going to be". This was confirmed by respondents who mentioned the break enhancing their experience of the summary.
It was suggested earlier that Ebbinghaus' concept of the recency effect (Crowder, 1976) might help explain the impact of the break and end-of-session summary. The very fact that the summary message is the last thing clients experience means it is most likely to be recalled. Those respondents who mentioned their experience that the message helped them reflect further after the session seem to support this. It might be argued that the move away from prescriptive tasks means that clients are not required to remember detailed instructions and so that improved recall explained by the recency effect is not relevant. However, if our end-of-session message reminds clients of successes and strengths they have already shown, then it could be argued that there is some benefit in their remembering this later. Further, broader research on the recency effect suggests that it is not just recall that is enhanced but also that those things presented last are more likely to have greater impact (Panagopoulos, 2011). In reality, the impact of the message is probably explained by a combination of factors (including those revealed by the themes that emerged from responses. Nonetheless, the recency effect is offered here as a reminder that the way we end our sessions matters.

Limitations and further debate
This qualitative study canvassed views from clients of only one therapist and that therapist conducted the survey and collected the data. Thus, the possibility that clients' ratings of the break and summary might have been inflated by their overall positive experience of this particular therapist cannot be