Heidi M. Levitt and Zenobia Morrill

Although clients’ silences have long been thought to be of interest to psychotherapists, the theories on what they denote have been widely disparate. In-session pauses in dialogue seem significant because they indicate that a process is unfolding that demands attention to the point of halting the discourse. Something important is going on. These moments have been thought to indicate positive dynamics, such as emotional attunement (Gendlin, 1996), closeness (Ferber, 2004), and trust and intimacy (Trad, 1993), as well as problematic dynamics, such as transferential struggle (Fliess, 1949; Sabbadini, 1991), regression (Winnicott, 1965), rage (Zeligs, 1961), cognitive burden (Perfetti & Bertuccelli-Papi, 1985), and resistance (Freud, 1912; Reik, 1926). For clinicians, this means that it can be difficult to know how to respond to a client during a pause. For researchers, it can be difficult to know how to identify pauses and their function.

The initial approach to the study of silences in therapy was to tabulate the total number of seconds per session, creating one variable for examination. This practice led to the production of a contradictory body of literature with conflicting interpretations and results. For instance, both Wepfer (1996) and Cook (1964) found more silence in therapy to be associated with insight and therapeutic success, however, Brähler & Overbeck (1976) found that smaller proportions of silence led therapists and clients to agree that sessions were positive. While silence was associated with lower client anxiety (Siegman, 1978), it also was associated with greater anxiety (Mahl, 1956). Researchers interpreted findings regarding silence as evidence of various processes, such as empathy (Hargrove, 1974) and discomfort (Becker, Harrow, Astrachan, Etre, & Miller, 1968). It was this conflicted state of the literature that led the first author to develop a system to differentiate pauses (Levitt, 2001a), rather than to see them as homogeneous.

The generation and credibility of the Pausing Inventory Categorization System (PICS)

The PICS has high internal validity because it was derived from a qualitative research study (Levitt, 2001a). Because the literature was rooted in differing theoretical perspectives, it was thought to be important to learn how clients experience these, often poignant, therapeutic moments. Understanding the intrapsychic experience of silences can assist therapists, researchers, or interlocutors in gaining insight into how a dialogue is influencing its participants.

Measure development

Seven clients from four psychotherapy orientations (emotion-focused therapy, interpersonal therapy, cognitive therapy, and client-centered therapy) were interviewed about their experiences of silences. In these 1- to 2-hour interpersonal process recall (IPR) interviews (Elliott, 1986; Kagan, 1975), the clients reviewed recordings of their sessions and then described what was happening during silences lasting three seconds or more in duration; shorter silences were seen as dysfluencies. Segments of the videotaped therapy session that were replayed in each interview included pauses of at least 3 seconds in duration. A total of 168 pauses were explored, just over half of the pauses (52%) in the sessions. The main question posed after each pausing event was “Can you describe your experience during this moment?” Using a grounded theory method of analysis, their descriptions of each silence experience were examined and clustered together based upon commonalities therein, revealing seven forms of silences (for further details on this method, see Levitt, 2001a).

Types of silences

Of the seven types of pauses identified, three were classified as productive, two as neutral, and two as obstructive. Still, all may be helpful in the session if addressed properly by a therapist who is sensitive to their functions.

Productive silences

There were three forms of silences that clients experienced as associated with progress in therapy: emotional, expressive, and reflective pauses. During these silences, clients developed new self-awareness and made gains.

Emotional pauses refer to the in-session pauses during which clients are accessing or experiencing their emotions. In these moments, emotional experiences range in valence and intensity and may include feelings such as fear, sadness, anger, and delight. Therapy discourse markers of emotional pauses include, but are not limited to, instances in which: clients expressed emotion words verbally (e.g., “I felt so sad. [Pause] I just feel so forlorn)”; emotion-laden context without accompaniment of emotion words (e.g., “I haven’t seen my daughter for five years now [pause] it has just been so very long)”; and voice intonations that tend to signify emotional experiences (e.g., shaky voice, tearful voice, shouting voice).

Expressive pauses are silences in which clients are engaged in a process of searching for accurate words, phrases, or labels to describe their current experience. In these moments, clients seek to articulate or symbolize their feelings or ideas. These moments could range from a quick search process with the objective of locating the expression that felt “right,” to a slower process that entailed a struggle to symbolize a complex or novel in-session experience.

Expressive pauses are associated with therapy discourse markers that include the awkward expression of experiences not previously articulated (e.g., “There’s no good way to put it … [Pause] I’m, I just not able to do that anymore”). In addition to this, when clients stutter or utter unusual or vague phrasing near an in-session pause, it could indicate that they are searching for a better, more accurate symbol of their experience (e.g., “I felt it had to do with all kinds of work things like … [Pause] oh, not being ambitious”).

During reflective pauses, clients are questioning ideas or generating connections and insights about an experience. Therapy session markers are mainly comprised of client expressions that indicate they are engaging in self-examination. The consideration of a new issue or idea could be initiated by the client or the therapist, and could result in client expressions of wondering, analyzing, judging, evaluating, or realizations of those behaviors. For instance, one client expressed: “Actually, it’s playing out probably a big thing … [Pause] now that I think about it. I was standing up to her.” Or, in the following instance, a client considered how to answer her therapist’s question about why she was being self-critical: “Well, just because [pause] I think I’m so used to being self-critical. It happens naturally.” This example demonstrates a reflective pause in which the client is engaging in developing novel insights and meaning.

Later research using the PICS led to a differentiation in the reflective pauses, enabling more attuned coding of those silent processes (Levitt & Frankel, 2004). In this revision, “low reflective” pauses indicated self-reflection that was mundane and not requiring a depth of thought. For instance, a client who was thinking about what to do after the session might say, “Hmm. [Pause.] I wonder if I go to see a movie with my husband or if we should just get dinner?” This type of silence contrasts with “high reflective” pauses in which clients are engaging in therapeutic self-exploration. To give a counter example, a client might consider patterns in her history: “I wonder why I never felt like I could be successful [pause] and whether this was tied to my Father’s hostility. [Pause.] I think he just chipped away at my self-esteem all the time.” This second type of pausing indicates that the client is not only considering established preferences but is engaging in the process of making new connections.

Neutral silences

These pauses signaled moments in which clients’ cognitive processing required some concentration. There are two forms of these silences. Mnemonic pauses are moments in which a client stops speaking in an attempt to recall an event or item. They also may entail remembering the order in which a happening unfolded, the use of a mnemonic phrase to stimulate memory, or a search for an event in a client’s history. For instance, these would be coded as mnemonic pauses: “My father did not come home often for dinner [pause] probably about every other week for Shabbat but really not during the workweek at all.” These pauses tended not to be associated with insights themselves, as they were accessing something that the clients already knew, but they could precede other productive pauses that were associated with moments of growth, for instance when remembering a childhood event, or could precede obstructive pauses if a resultant memory is too threatening.

Associational pauses indicate a style in which clients pause because they are switching from the description of one topic to another. Although these pauses are quite rare, some clients have a style of topic-switching and keeping the conversation at a more surface level, without any immediate threat that is causing them to veer the conversation away. Often, they seem to have a list of topics in their heads that they are wanting to review. An example of an associational pause is: “So, in the end we won the baseball game and all went to dinner. [Pause.] But, I wanted to tell you about what happened on my date last week.” In the silence, clients seem to have a sense that one topic is concluded or have some internal association that moves them to switch to the next topic. Because these pauses are rare, researchers may choose not to code for them if they are not relevant to the data in question. However, it can be good for therapists to be aware of these types of pauses to inquire about them in order to learn more about associations at play and draw clients’ attention to their topic-switching style, which might limit therapeutic activity.

Obstructive silences

Two forms of obstructive silences were identified as patterns in clients’ experiences. These pauses were labeled “obstructive” because they stopped the clients’ progress in the moment when they occurred. If the therapist was able to identify the process, however, and develop an attuned intervention to direct clients to consider these silences and how they influenced the exchange, these pauses could lead to central insights in the therapy.

Disengaged pauses signify clients’ emotional avoidance during the discussion of a particular session topic. Clients often reported discomfort during these silences. Disengaged pauses function as a way to deflect the topic at hand, detract from the therapist’s exploration of issues perceived as threatening, or simply withdraw from the interaction. Sometimes, client participants shared that this was a conscious process. In other instances, however, disengaged pauses represented an automatic response to heightened tension or emotional discomfort.

Disengaged pauses corresponded to numerous therapy session markers. They tend to occur alongside discussion of threatening emotions. Within such discussions, after pausing, clients typically signaled that they had stopped processing to the same depth as they had been prior to the pause. This was done by engaging in jokes, summarizing, dismissing, or distracting from the prior topic or emotional state. Clients used methods such as reassuring themselves, distracting themselves, and coaching themselves to “swallow” or avoid feelings. In the following quote from an interview about a session, one client described what it was like to use techniques to compose herself and avoid feelings.

C: I guess I feel it was like I was trying to choke something down, and I guess I was having a great deal of trouble with how I was doing at the time, how I was feeling. … Like trying to swallow a bone … Part of the problem is it’s just the, I’m at the point where I’m so well practiced at that thing that it’s very much an automatic process that kicks in, but I’m conscious of it happening. … I get the feeling there would be some kind of negative consequence if I was to allow emotional side to get out, I feel as though, I’ve got this idea that there’s some reason that I should be keeping it in like that … A few seconds later I’d say that it’s gone ’cause … I successfully was able to contain it.

In addition to avoiding feelings by resisting them, clients described related disengaged strategies. These could unfold in the form of clients attempting to lighten the conversation, making a joke, or switching to topics that are not as threatening to discuss.

Also, some clients shut down or withdrew from the dialogue. Clients might find themselves forgetting what was being discussed, feeling tired or confused. They might feel disconnected from the words they are speaking, as though they are watching the conversation from afar and engage in self-soothing until they can re-engage again. These processes could occur automatically and disrupt explorations that could be fruitful if they were engaged safely.

The second type of obstructive silence was interactional pauses. In these moments, clients shifted away from focusing on personal exploration to a focus on the therapist or the therapy relationship. During sessions, clients continually monitored therapists’ reactions and their own self-presentation. If clients felt negatively evaluated, confused by the therapist reaction or communication, or felt that they were not making a positive impression, they might engage in interactional pauses. In general, clients were invested in maintaining a positive alliance with the therapist and would stop to consider how to remedy any potential breech. Interactional pauses were found to be significantly longer in duration than the other types of silences (see Levitt 2001b) and they could feel awkward or stilted.

One client engaged in repeated interactional pauses during her session. While guiding her through a cognitive restructuring exercise, her therapist asked her what a friend might tell her about her negative thoughts. Although she knew that the therapist was seeking a response of affirmation, she paused and considered “all the disappointments I’ve had with friends not responding to my needs … It wasn’t really a good question for me.” Reluctant to admit that the friend would say nothing encouraging, she remained silent, and then said she didn’t know. The therapist then asked her what she might tell a friend who had these same thoughts, “but it just brought back [how] my friends let me down, and now I’m in a position where I have to tell them something that will be quite positive and I just wasn’t able to do it.” After a lengthy 34-second pause in which she wrestled with this dilemma, she replied to her therapist, “I suppose I’d tell a person in my shoes to have enough confidence in their proven ability to set limits” and then was quiet for another 16 seconds. When asked in the IPR interview about her experience of this last pause, she replied that it was due to feeling disbelief in what she had just said. Feeling as though she could not come up with an answer, she acted disingenuously to provide the correct response to the therapeutic task and hoped for time outside session to explore her true feelings. Throughout much of her session, this client engaged in a private evaluation of her interaction—none of which was shared with her therapist. She compared her experience of therapy to a game show, in which she was always waiting for a crowd to shout, “Good answer!” This example typifies the experience of interactional pauses as clients are trying to maintain their integrity and the alliance in the face of a potentially confusing or upsetting therapist reaction.

Although the last two categories of pauses are classified as obstructive, this characterizes their function if they are unaddressed. In contrast, these moments can lead to important insights when therapists notice these silences and invite clients to articulate their thoughts. By asking clients what is happening for them in the silence, they can initiate discussions about the therapy relationship and about the challenge of facing threatening topics that can be foundational.

As can be seen, the activities during these seven pauses are quite distinctive. The PICS can be useful in therapist training, as therapists can come to identify these processes by actively attending to the cues surrounding each silence. For instance, lengthy moments of awkward silence often are interactional pauses. Silences, when approaching threatening topics, may signal disengagement, especially if the client veers away from the issue. During moments of contemplation, clients may be reflecting. Therapists then can respond accordingly, to encourage clients to stay with or return to productive introspective pauses or to ask clients to put words to obstructive moments, increasing awareness of their in-session patterns.

For psychotherapy researchers or discourse analysts, the PICS allows for researchers to examine dynamics in dialogue that take into consideration the internal experiences that give meaning to these moments. The PICS manual (Levitt & Frankel, 2004) provides descriptions of cues using both transcripts and session recordings (available upon request from the first author). This categorization system provides an empirically based method for researchers to identify patterns that might otherwise go unnoticed.

Validity and credibility information

In addition to the strong internal validity based upon the qualitative analysis, there are a number of ways the rigor of the PICS has been established, including examinations of its reliability, cultural validations, and an empirically based sampling system.

Inter-rater and client-rater reliability

First, clients provided feedback upon the completion of their interview. They were encouraged to discuss anything that remained unsaid and to clarify experiences that may have been misinterpreted. In addition to this, the results of the study were mailed to clients for their review and any additional feedback. Most of the clients were no longer accessible, yet one client did respond. This client’s response endorsed the model presented and they shared that they had experienced insight from partaking in the interview. As a third check, the PICS manual was developed to describe and exemplify the seven pausing processes based upon this analysis. An independent rater, guided by the PICS manual, categorized both pauses from the therapy sessions and descriptions of pauses from the IPR research interviews. The rater achieved a Cohen’s kappa of .70 with the primary investigator on the rating of 40 randomly selected pauses from the session transcripts. When rating 35 randomly selected descriptions of pauses from the IPR interviews, their interrater agreement was .82. When rating recordings instead of transcripts, ratings were found to be higher. A recent study demonstrated a Kappa of .91 (Guzmán et al., 2018). The client-rater agreement was .83 when their descriptions of pauses in the IPR interviews were compared with the independent rater’s ratings of in-session pauses. Client-rater reliability ratings tend to be rare in process measures research. In addition, the study did reach saturation. It seemed that new forms of silences were not forthcoming in the final interview, suggesting the analysis was comprehensive (Glaser & Strauss, 1967).

Empirically-based sampling system

Coding using process measures is time-intensive. It requires raters to locate and evaluate each instance of the phenomenon under consideration, and to obtain inter-rater reliability and/or engage in a method of consensus with other raters. To substantially reduce the time that this takes, a study was conducted to see the amount of text that would need to be coded to reliably represent the therapy dynamic (Frankel, Levitt, Murray, Greenberg & Angus, 2006). Across 90 sessions, 1,503 silences were coded. Three sampling strategies were compared with the complete census of silence data: one sampled the first three, middle three, and last three sessions, whereas the other two randomly sampled either a quarter or a half of every session. The one-half random sampling strategy provided results closest to the complete census. This finding means that researchers can randomly select and rate half of the silences and still have reliable results–considerably reducing the time and costs required for coding.

Cross-cultural validation

This measure has been used by researchers internationally. These have included projects in Austria (Innsbruk University); Canada (OISE); Germany (International Psychoanalytic University in Berlin; University of Ulm; University of Witten); Denmark (University of Copenhagen); Chile (Pontifical Catholic University); Greece (Aristotle University of Thessaloniki); the UK (Anna Freud Center/University College London); Israel (Bar Ilan University); and in the US (Gannon University, The University of Memphis). This international research has suggested that the PICS is useful across diverse languages and cultures (e.g., Avdi, Verdenhalven & Acheson, 2018; Daniel et al., 2018; Holtmann, Seybert, & Huber, 2018).

Researchers in Chile embarked on a cross-cultural validation study of the PICS. Guzmán et al. (2018) conducted an ambitious dissertation in which she engaged in retrospective interviews of therapists and clients about their experiences of in-session silences. The findings of that study endorsed the categories of the PICS, with the exception of the associational pauses, which were found to be quite rare in other research and had been theorized to be mostly applicable with patients with highly avoidant personality styles (Levitt, 2001a). This version of the PICS manual contains cues related to the Spanish language and can be helpful in that context (Guzmán et al., 2018). In addition, the Spanish version includes a coding system for therapist silences (see Ladany, Hill, Thompson & O’Brien, 2004 for an English language version of a measure to code therapists’ pauses) and so offers the potential to consider the session from multiple perspectives. These findings support the use of the PICS for evaluating therapy in non-English languages. In contrast to most process measures that are developed theoretically, empirical evidence of the validity, reliability, sampling, and cultural use of the PICS has established it as unusually rigorous.

Findings

The PICS has been used in a number of research theses and dissertations, presentations, and published studies. It has been used in a variety of research designs, such as intensive case study, multi-case study, and randomized controlled trials comparing psychotherapy orientations.

Intensive case studies

In this section, two case studies are reviewed in which the PICS was used to examine the microprocesses that unfold within a therapy session.

Case study of client with learning disability

To better understand how silences function in successful therapy, Gindi (2002) examined 40 hour-long psychodynamic therapy sessions, with a young adult who had a learning disability, that were considered successful by both client and therapist. Out of every second session, a 15-minute excerpt was transcribed and analyzed according to the initial PICS manual (Levitt, 2001b). The results showed that the overall number of silences increased as the therapy progressed. Specifically, reflective silences were found to increase, thus supporting the hypothesis that increasing reflection is indicative of progress in therapy. Feeling silences also showed some evidence of an increasing trend. These findings suggested that, across their therapy, clients became more comfortable with therapeutic introspection or found it more valuable and so came to practice it more readily.

Case study of client recovery from trauma

In addition to the cross-cultural qualitative validation study, Guzmán et al. (2018) examined a three-year weekly psychodynamic psychotherapy that was 88 sessions in duration and focused on recovering from trauma related to family sexual abuse. Over this long-term therapy, they found the number of silences increased gradually during the therapy until about the midway point, when it gradually decreased, suggesting that long-term therapies might display a different trajectory. Clients engage first in concentrated work to gain insight and then shift toward integrating that insight. They examined 83 change episodes (Krause et al., 2007), indicating periods of new insight, and 79 rupture episodes (Safran & Muran, 1996), indicating relational problems from the same therapy. They coded 298 silences within those episodes. The evidence from this study confirmed the prior PICS findings, in that the numbers of reflective and expressive silences were significantly higher in the change events while the numbers of interactional and disengaged pauses were higher in the rupture events. Emotional pauses did not differ between these experiences, likely because strong emotions are characteristic of both change and relational rupture experiences. Together, these studies provide evidence that the amount of silence in therapy changes across its duration, supporting similar other findings (e.g., Frankel et al., 2006).

Comparisons of good and poor therapies

A question posed by the following research is whether differences exist in how silences function in therapies that are successful in comparison with ones in which change is minimal.

York Depression Project study

One of the first studies that used the PICS was a study of three good- and three poor-outcome emotion-focused psychotherapies from the York University Depression Project I (Frankel et al., 2006). In accordance with the hypotheses proposed by the PICS, the findings demonstrated that productive silences, namely reflective pauses, had a higher relative frequency in therapy sessions that resulted in positive outcomes. Conversely, obstructive silences, specifically disengaged pauses, were featured with a higher relative frequency than productive pauses in therapy dyads associated with poorer outcomes. In other words, the good-outcome dyads exhibited more emotional, expressive, and high-reflective silences and fewer disengaged and interactional silences than poor-outcome dyads.

Although power in this study was limited, large effect sizes that approached significance provided quantitative support for the PICS. The PICS demonstrated reliability in identifying pauses and subtly discerning their associated outcomes in psychotherapy. Frankel et al. (2006) also found that the frequency of silences changed across the duration of treatment. Emotional silences occurred earlier and more often in treatments associated with good outcomes. Poor outcome dyads were characterized by more emotional silences occurring later in treatment, possibly indicative of those clients taking longer to engage their feelings in psychotherapy.

Munich psychotherapy study

In Germany, Schlotheuber (2018) found similar findings using the PICS by looking at a good and a poor psychoanalytic therapy. She found significantly higher frequencies of productive pauses in the good-outcome case and significantly higher frequencies of obstructive pauses in the poor-outcome case. Although her coding reliability was only moderate, examining these data in conjunction with the other research supports prior findings. These data suggest clinicians should actively invite and sustain productive pauses in sessions but inquire directly about obstructive pauses to help clients work through blocks to their treatment.

Comparing silences across psychotherapy orientation

The following studies have examined the ways the silent processes of the PICS function across psychotherapies that were based in differing conceptual theories.

Psychoanalytic and CBT therapies for bulimia

A recent study examined the association of silences in session with client attachment, therapeutic alliance, and treatment outcome (Daniel, Folke, Lunn, Gondan, & Poulsen, 2018). Clients were provided with either psychoanalytic psychotherapy or cognitive behavioral therapy (CBT) for bulimia nervosa. Their primary aim was to determine whether or not the quality and quantity of in-session silences accounted for the relationship between client attachment and therapeutic alliance. Using multilevel linear regression analyses, they found a significant relationship between client attachment and frequency of pauses in session. Sessions with dismissive-avoidant clients had the highest total frequency of pauses followed by secure clients and then preoccupied clients with the fewest pauses. A major finding was that better therapeutic alliance and treatment outcome was predicted by lower silence frequency and lower relative frequency of obstructive silences.

Daniel et al. (2018) also found that all types of pauses, except for obstructive pauses, occurred more frequently in the psychoanalytic psychotherapy sessions than in the CBT sessions. In the CBT sessions, obstructive pauses occurred more, particularly in the “working phase,” the middle portion of treatment. During the working phase of CBT, not only did obstructive pauses increase but productive pauses decreased. This pattern was not observed in psychoanalytic psychotherapy. For CBT, frequent silences were thought to indicate that something is obstructing change whereas in psychoanalytic therapy, frequent silences were thought to be indicative of the therapy working. In concordance with the findings from other studies (e.g., Gindi, 2002; Guzmán et al., 2018), both types of therapy featured fewer silences initially than in the middle or late phases of treatment. Overall, this study suggested that attention to silence in session can aide therapists in tailoring to client attachment and bolster therapeutic relationship and treatment outcome.

Psychodynamic, psychoanalytic, and CBT therapies for depression

Using the PICS, Holtmann, Seybert, and Huber (2018) studied silences within a set of six therapies that were equally randomized between psychodynamic, psychoanalytic, and cognitive-behavioral therapy approaches. For each therapy, they examined a beginning, middle, and end session and coded, in total, 1,008 pauses of at least three seconds. Here, the findings from their fascinating research related to the comparison of therapies are summarized.

Across the therapies, CBT was found to show significantly fewer productive silences compared to the other treatments. For the psychodynamic therapy, there were significantly fewer obstructive pauses than for the psychoanalytic therapy. There were more obstructive pauses for the CBT approach, but this finding was not statistically significant. When examining high and low reflective pauses in the three treatments, the former were found to be significantly less frequent in CBT. These two studies together suggest that silences in CBT are more likely to be obstructive, while lengthy productive silences are more typical of insight-oriented therapies.

Examinations of silences and other processes

In addition to these studies, a number of innovative studies have examined silence processes in connection with other processes that unfold in psychotherapy to see how they relate.

Computer textual analysis of emotional tone and abstraction

To examine the ways in which these pauses related to emotional and reflective text, using a different coding system, the same six Process Experiential therapies from the York Depression Project were coded with both the PICS and the Therapeutic Cycle Model (Mergenthaler, 2008), that entails a computer analysis of verbal interactions. To examine the local context of each pause, the 75 words before and after pauses were classified for using this model. The sessions were rated as falling into one of the following four states which reflected different levels of affective experiencing (emotional tone: ET) and cognitive mastery (abstraction: AB): low ET and AB signaled a state of “Relaxing”; high ET and low AB indicated “Experiencing”; high AB and low ET suggested a process of “Reflecting”; and when both were high, the client was “Connecting.” In that model, therapeutic change is attributed to states of “Connecting” as well as “Experiencing.”

Findings from this study revealed a highly significant difference for types of reflective pauses. Connecting and Experiencing text was found to predict “high reflective pauses” and Relaxing and Reflecting text predicted “low” reflective pauses. This suggested that clients who were neither cognitively or emotionally engaged, or who were only cognitively engaged, produced more mundane reflective pauses. Emotional engagement was associated with high reflective pauses, indicating deeper connections, exploration, and insight. The findings were seen as mutual validation of two independent concepts (Mergenthaler & Levitt, 2005).

Adolescents in psychoanalytic therapy and therapists’ interventions

At the Anna Freud Center, Evrinomy Avdi has directed the research of both Nia Verdenhalven and Rachel Acheson (2018) in which they used the PICS to study 18 sessions drawn from three short-term psychoanalytic psychotherapies of adolescents with depressive diagnoses. They used a process of consensus to agree on the coding of 1,248 silences, finding that approximately three quarters of the silences were 3 to 10 seconds in duration, a little under a quarter were 11 to 60 seconds, and only 3.37 percent exceeded a minute in length—the majority of these being obstructive silences (86% disengaged). Looking closer at these longer silences, they found that the therapists most frequently broke the silence using an interpretation (33%), followed by emotion reflections (16.7%) and questions (16.7%). They found that productive silences were most common in the middle phase of therapy, which they saw as coherent with the model of therapy in use that expects clients to be most engaged during this period. Notably, the patients all spontaneously raised the experience of silence as a notable and multi-dimensional aspect of their therapy during their interviews on this experience—speaking to the importance of these moments in therapy.

A closer look at disengaged silences in an effectiveness study

Early sessions (second or third sessions) of 52 clients (Stringer, Levitt, Berman & Mathews, 2010) from across many psychotherapy orientations were examined to shed light upon the role of disengaged silences in therapy. These silences indicate withdrawing or distancing from a threatening topic. The clients were seen in a counseling center and were seeking therapy for diverse reasons. Results indicated that disengagement predicted poorer proximal (third session typically) and distal outcome (end of therapy) as measured by a short form of the Beck Depression Inventory (BDI; Beck, Guth, Steer, & Ball, 1997) and poorer proximal outcome on the Symptom Checklist-5 (Tambs & Moum, 1993). Inter-item analyses revealed that disengagement had a significant proximal effect on depressive mood and negative self-evaluative items for the BDI-PC, but across time these effects were sustained for only the negative self-evaluative items. These findings suggest that disengaged silences are associated with negative mood in the short term and with longer-term negative self-evaluation.

Conclusion

Overwhelmingly, the PICS has been shown to identify patterns in how people create new understandings within therapy dialogue. The body of research reviewed in this chapter speaks to the importance of examining therapy not only as a verbal endeavor, but also as one in which the discourse reflects the rich internal world of its participants. By distinguishing types of pauses, the PICS has not only enabled researchers to differentiate clients’ responses within their studies, but it has been helpful for developing therapists to act responsively to their clients’ silences in session. Repeatedly, across the studies reviewed, it has been found that the internal experiences within silences matter. These are the moments in which clients are at the peak of introspection, connection making and emotional experience. They represent the heights of interpersonal tension and intrapsychic threat. This system holds utility for psychotherapy process researchers or discourse analysts that wish their research to reflect the meaningful internal experience of actors within dialogue.

References

Avdi, E., Verdenhalven, N., & Acheson, R. (2018). Studying Silence in Psychoanalytic Psychotherapy with Adolescents with Depression [Unpublished manuscript]. London: Anna Freud Center.
Brähler, E., & Overbeck, G. (1976). Therapist’s and patient’s speech-pause behavior and psychotherapy session. Dynamische Psychiatrie, 9(4), 275–286.
Beck, A. T., Guth, D., Steer, R. A., & Ball, R. (1997). Screening for major depression disorders in medical inpatients with the beck depression inventory for primary care. Behaviour Research and Therapy, 35(8), 785–791.
Becker, R. E., Harrow, M., Astrachan, B. M., Detre, T., & Miller, J. C. (1968). Influence of the leader on the activity level of therapy groups. The Journal of Social Psychology, 74(1), 39–51.
Cook, J. J. (1964). Silence in psychotherapy. Journal of Counseling Psychology, 11(1), 42–46.
Daniel, S. F., Folke, S., Lunn, S., Gondan, M., & Poulsen, S. (2018). Mind the gap: In-session silences are associated with client attachment insecurity, therapeutic alliance, and treatment outcome. Psychotherapy Research, 28(2), 203–216.
Elliott, R. (1986). Interpersonal Process Recall (IPR) as a process research method. In: L. Greenberg & W. Pinsoff (Eds.), The Psychotherapeutic Process: A Research Handbook (pp. 503–528). New York: Guilford Press.
Fliess, R. (1949). Silence and verbalization: A supplement to the theory of the analytic rule. International Journal of Psycho-Analysis, 30, 21–30.
Frankel, Z., Levitt, H. M., Murray, David, Greenberg, L. S., & Angus, L. (2006). Assessing silent processes in psychotherapy: An empirically derived categorization system and sampling strategy. Psychotherapy Research, 16(5), 627–638.
Freud, S. (1912). The dynamics of transference. In: J. Strachey (Ed.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 12, 1911–1913, pp. 99–108). London: Hogarth Press, 1958.
Ferber, S. G. (2004). Some development facets of silence: A case study of a struggle to have a proximity figure. British Journal of Psychotherapy, 20(3), 315–332.
Gendlin, E. (1996). Focusing-Oriented Psychotherapy: A Manual of the Experiential Method. New York: Guilford Press.
Gindi, S. (2002). Patterns of Silences: A Case Study of a Young Adult with Learning Disabilities [Unpublished Master's Thesis]. Toronto, Canada: University of Toronto.
Glaser, B. G., & Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Theory. New Brunswick: Aldine Transaction.
Guzmán, M., Vidal, J., Soto, M., Jaime, D., Tomicic, A., Martínez, C., & Levitt, H. M. (2018). Coding System of Silences for Therapists and Patients (PICS-TP): A Manual for Training and Coding Silences of Therapists and Patients in Psychotherapy for a Spanish-Speaking Context. Santiago, Chile: University of Chile.
Hargrove, D. S. (1974). Verbal interaction analysis of empathic and nonempathic responses of therapists. Journal of Consulting and Clinical Psychology, 42(2), 305.
Holtmann, K. C., Seybert, C., & Huber, D. (2018). Sprachliche Charakteristika der Depression. Eine vergleichende Empirische Studie zu Sprechverhalten und Sprechpausen in Psychotherapie. [Verbal characteristics and speech-pause behavior in depressed patients. A compared process-analysis of psychoanalytic, psychodynamic and cognitive-behavioral therapy]. Berlin: International Psychoanalytic University.
Kagan, N. (1975). Interpersonal Process Recall: A Method of Influencing Human Interaction. Houston, TX: University of Houston.
Krause, M., De la Parra, G., Arístegui, R., Dagnino, P., Tomicic, A., & Valdés, N. (2007). The evolution of therapeutic change studied through generic change indicators. Psychotherapy Research, 17(6), 673–689.
Ladany, N., Hill, C. E., Thompson, B. J., & O'Brien, K. M. (2004). Therapist perspectives on using silence in therapy: A qualitative study. Counselling and Psychotherapy Research, 4(1), 80–89.
Levitt, H. M. (2001a). The sounds of silence in psychotherapy: The categorization of clients’ pauses. Psychotherapy Research, 11(3), 295–309.
Levitt, H. M. (2001b). Clients’ experiences of obstructive silence: Integrating conscious reports and analytic theories. Journal of Contemporary Psychotherapy, 31(4), 221–244.
Levitt, H. M. (2002). The unsaid in the psychotherapy narrative: Voicing the unvoiced. Counselling Psychology Quarterly, 15(4), 333–350.
Levitt, H. M., & Frankel, Z. (2004). Pausing Inventory Classification System (PICS) Manual, 2nd Revision. [Unpublished Manuscript]. Memphis, TN: University of Memphis.
Mahl, G. F. (1956). Disturbances and silences in the patient’s speech in psychotherapy. Journal of Abnormal and Social Psychology, 53(1), 1–15.
Mergenthaler, E. (2008). Resonating minds: A school-independent theoretical conception and its empirical application to psychotherapeutic processes. Psychotherapy Research, 18(2), 109–126.
Mergenthaler, E., & Levitt, H. M. (2005, June 24). Pausing experiences and the Therapeutic Cycles Model [Conference paper]. In: Micro-Processes in Therapeutic Dialogue: How Do Patient and Therapist do the Work of Therapy? Conference Conducted at the 36th International Meeting of the Society for Psychotherapy Research, Montreal, Canada.
Perfetti, C. C., & Bertuccelli-Papi, M. (1985). Towards a cognitive typology of pause phenomena. Communication and Cognition, 18, 339–351.
Reik, T. (1926). The psychological meaning of silence. Psychoanalytic Review, 55(2), 172–186.
Sabbadini, A. (1991). Listening to silence. British Journal of Psychotherapy, 7(4), 406–415.
Safran, J. D., & Muran, J. C. (1996). The resolution of ruptures in the therapeutic alliance. Journal of Consulting and Clinical Psychology, 64(3), 447–458.
Schlotheuber, R. E. (2018). Pausing Processes in Psychoanalytic Therapies [Unpublished Master’s Thesis]. Herdecke, Witten: University of Witten.
Siegman, A. W. (1978). The meaning of silent pauses in the initial interview. The Journal of Nervous and Mental Disease, 166(9), 642–654.
Stringer, J. V., Levitt, H. M., Berman, J. S., & Mathews, S. S. (2010). A study of silent disengagement and distressing emotion in psychotherapy. Psychotherapy Research : Journal of the Society for Psychotherapy Research, 20(5), 495–510.
Tambs, K., & Moum, T. (1993). How well can a few questionnaire items indicate anxiety and depression? Acta psychiatrica scandinavica, 87(5), 364–367.
Trad, P. V. (1993). Silence: The resounding experience. American Journal of Psychotherapy, 47(2), 167–170.
Wepfer, R. (1996). Silence in psychotherapy: A quantitative analysis [Conference paper]. In: Conference Conducted at the 27th International Meeting of the Society for Psychotherapy Research, Amelia Island, FL.
Winnicott, D. W. (1965). The Maturational Process and the Facilitating Environment. London: Hogarth Press.
Zeligs, M. A. (1961). The psychology of silence. Journal of the American Psychoanalytic Association, 9(1), 7–43.

Chapter 13

Measuring silence