The Role of Safety in Change-Promoting Therapeutic Relationships: An Integrative Relational Approach

Abstract

Objective

The significance of the psychotherapeutic relationship in promoting psychotherapeutic change is widely recognized. In this paper, we contribute to the relational orientation of psychotherapy through a transtheoretical exploration of safety. We aimed to identify and integrate those relational and change-promoting principles and aspects of safety that are school-independent.

Method

We conducted an overview and synthesis of the clinical-theoretical and empirical literature that we believe has significantly addressed the role of safety in regulating change-promoting therapeutic relationships.

Results

The relational and change-promoting aspects of safety form a dynamic system involving the therapist, the client, and the relationship. These interact, influence each other, and perform multiple homeostatic functions: they allow to resist change, assimilate small changes that do not disrupt the client’s way of functioning, regulate major changes that disrupt and alter the client’s way of functioning, and regulate adjustments in the way the therapist and client work together. From an integrative-relational perspective, a safe therapist is a precondition for co-creating a safe environment. This establishes trust and fosters an affective bond that provides additional sources of safety for the therapeutic relationship and the client. To promote change, however, the relational aspects of safety need to be fine-tuned (calibrated and personalized) for each therapy in terms of intensity, duration, timing, scope, and sources, accommodating developmental, individual, and situational differences. Crucially, the safety of the therapist, the client, and the relationship must be neither perfect, steady, or static, but rather safe enough, adaptive, and dynamic, leaving space not only for self-discovery and self-awareness but also for the co-regulation of tolerable frustrations, disappointments, and insecurities that facilitate the client’s resilience and adaptation.

Conclusions

Focusing on school-independent, safety-based relational principles and understanding how they evolve and adapt over time and across circumstances can make a significant contribution to the current relational orientation in psychotherapy. This has important implications for psychotherapy practice, training, and research.

Introduction

The therapeutic relationship is central to clinical expertise (APA Presidential Task Force on Evidence-Based Practice, 2006). A growing number of scholars and clinicians have been acknowledging that, regardless of therapeutic modality, the therapeutic relationship is the sine qua non of psychotherapy (Norcross & Lambert, 2014), “predicts outcome of psychotherapy over and above alliance” (Wampold & Imel, 2015, pg. 211), presents a decisive contribution to client improvement, and “serves as the active process or mechanism of change” itself (Norcross & Lambert, 2014, p. 398). In this respect, all therapeutic approaches “rest on the relationship as the foundation on which the treatment is built” (Markin, 2014, p. 328). Such a position is coherent with decades of research evidence converging with clinical expertise experience and summarized and discussed by the work of the American Psychological Association Division of Psychotherapy (29) task force (Norcross, 2002, 2011; Norcross & Lambert, 2019a, 2019b). This research evidence indicates a consistent and robust association between different aspects of the therapeutic relationship (e.g., warmth, acceptance, alliance, empathy, goal consensus, positive regard, congruence, genuineness, collecting client feedback, and repairing alliance ruptures) and client improvement across a variety of populations, treatment settings, psychotherapeutic orientations, and diagnoses (average correlation of about .25-.30, corresponding to an effect size of about .55; Norcross & Lambert, 2014; see also Wampold & Imel, 2015).

Based on these research findings, Norcross and Lambert (2014) elaborated two important take-home messages: (1) “the therapy relationship makes substantial and consistent contributions to psychotherapy outcome independent of the specific type of treatment” and (2) “efforts to promulgate best practices or evidence-based practices (EBPs) without including the relationship are seriously incomplete and potentially misleading.” (p. 399). In sum, we are facing “a gradual movement toward a relational orientation across theoretical orientations” (Norcross & Lambert, 2014, p. 402), with a massive body of research supporting “the therapeutic relationship as a foundational pan-theoretical change agent” (Markin, 2014, p. 327). These and analogous considerations have led Markin (2014) to state that “it seems time to move toward developing a more pan-theoretical relationship orientation to psychotherapy” with the aim of “establishing a common identity for relationally inclined clinicians across proscribed theoretical orientations” (pp. 327-328). Relational principles suggest that interventions are relational acts that promote the establishment of a safe relationship through personal, empathic, egalitarian, attuned, nonjudgmental, and authentic stances. Such a relationship facilitates the client’s affect regulation in interpersonal contexts and new relational experiences that can be transposed into their close relationships outside therapy (Markin, 2014).

We aim to contribute to the relational orientation in psychotherapy described so far by specifically focusing on safety. Scholarly literature indicates that safety is necessary not only for healthy brain development in early childhood but also for healthy relationships in adulthood (Allison & Rossouw, 2013; Bowlby, 1988; Gilbert, 2004, 2006; Porges, 2021; Schore, 2003) and successful outcomes of psychotherapy (Norcross & Lambert, 2019a, 2019b). On the contrary, a lack of safety (e.g., childhood neglect, trauma, unpredictable responses to attachment-seeking behaviors) significantly correlates with psychopathology and insecure relationships (Cassidy & Shaver, 2016; Gilbert, 2006; Schore, 2003). Moreover, it has been argued that the concept of safety permeates the theory and practice of different psychotherapeutic orientations and plays a critical developmental role in promoting change and adaptation within relational contexts, both in ontogenesis and clinical practice (Podolan & Gelo, 2023).

In this paper, we build on the foregoing to explore more specifically how safety can affect clinically productive therapeutic relationships and, in turn, the process of change. Our approach is integrative-relational since it focuses on the role of safety in regulating human relationships in general and the therapeutic relationship across therapeutic orientations in particular. Our main question was whether it is possible to identify from the existing literature on the role of safety in psychotherapy a set of school-independent principles that can help to understand its role in regulating change-promoting relationships. A systematic review was beyond the scope of this article. Rather, we referred to and discussed the clinical-theoretical and empirical literature that we consider to have significantly addressed the role of safety and safety-related processes in regulating change-promoting therapeutic relationships across a variety of therapeutic orientations. We first introduce the concept of safety and its relevance for psychotherapy. Then, we provide an overview of some school-independent principles regarding the role of safety in the psychotherapy change process with a specific focus on the therapeutic relationship. These principles are summarized from the existing literature on safety and psychotherapy within and outside the framework of attachment theory. Finally, we review some empirical findings supporting these principles and suggest future lines of research.

Safety and psychotherapy

The definition of safety in the psychotherapeutic literature is conceptually and terminologically fragmented. Bowlby (1973) referred to safety as an objective state of freedom from harm and danger, and security as a subjective feeling of being safe (i.e., free from anxiety, apprehension, and alarm), as reflected in the terms safe haven and secure base (see also Ainsworth, 2010; Blatz, 1966; Sullivan, 1955). However, over the years, this distinction between safety and security has faded for several influential authors, who have increasingly used the term safety to also refer to subjective experiences and not only to objectively given conditions (e.g., Beck, 2021; Bromberg, 2006; Cortina & Liotti, 2010a; Fonagy & Allison, 2014; Porges, 2021; Van der Kolk, 2015).

In the present paper, we use safety in line with these latter authors. Thus, safety is not simply the absence of danger, but encompasses multiple nuances that defy simple categorization and resist universal definition (for a discussion, see Podolan, 2020). At a very general level, safety in psychotherapy involves establishing a trusting, empathic, non-judgmental, and confidential therapeutic relationship that encourages clients to express themselves freely and promotes their well-being. In psychotherapy, however, the specific characteristics of safety are largely shaped by psychotherapeutic theories and constructs.

For example, psychodynamic approaches center safety on unconscious processes, using containment, holding, and mirroring along with attuned interactions to create a trusting and safe environment (Cabaniss, 2016). Safety is important primarily because it fosters insight into internal conflicts and unconscious drives and promotes their management and reflective functioning. On the other hand, attachment-based therapies focus on the possibility of a secure attachment with the therapist as the source of safety, providing a reliable and trustworthy relationship characterized by consistent emotional support (Bowlby, 1988). Here, safety is critical because it facilitates the exploration of emotions, memories, and novelties, as well as emotional regulation and healing. In contrast to focusing on the unconscious or attachment aspects of safety, cognitive-behavioral approaches establish safety through collaborative empiricism by implementing structured and predictable sessions, positive reinforcement, and evidence-based techniques (e.g., psychoeducation focused on understanding psychological processes) (Beck, 2021). Safety is essential because it facilitates the exploration of negative cognitive and behavioral patterns and the practice of new strategies and coping skills. Unlike other therapies, trauma-focused therapies prioritize the neurobiological aspects of safety through grounding techniques (e.g., visualizing a safe place), establishing clear boundaries, and providing trauma-informed educational responses about trauma reactions and triggers (Van der Kolk, 2015). For these therapies, safety is paramount to the regulation and reprocessing of challenging bodily sensations, emotions, and traumatic experiences. Finally, humanistic approaches view safety as dialectical, emphasizing the importance of the therapeutic relationship characterized by authentic, empathic presence and unconditional positive regard as the cornerstone of healing (Rogers, 1995). Safety is crucial because it fosters the exploration of the true self, enables authentic self-expression, and facilitates the tolerance of tension, thereby preventing stagnation and promoting personal growth.

While all these therapeutic modalities prioritize the creation of a safe environment and its role in shaping the therapeutic relationship, they differ in their specific theoretical orientations for achieving this goal (for a review, see Podolan & Gelo, 2023). Even within these main approaches, sub-theories may emphasize different aspects of safety. Some may prioritize establishing firm therapeutic boundaries or conflict resolution, while others may focus on the development of self-cohesion, empathy, mentalization, mindfulness, repairing attachment wounds, increasing emotional regulation, interpersonal effectiveness, psychoeducation, or internalization of adaptive relational experiences (e.g., Beck, 2021; Fonagy & Allison, 2014; Kohut, 1971; Schore, 2012; Winnicott, 1971). These diverse perspectives illustrate the complexity of conceptualizing safety. As such, safety in psychotherapy appears to exhibit a combination of uniform elements and diverse contextual variations, reflecting its homogeneous foundation and heterogeneous application.

Interestingly, all the different psychotherapeutic approaches share the idea, though to a different extent, that safety is a fundamental prerequisite for our experiences to be effectively processed and organized (e.g., Schore, 2003). Podolan and Gelo (2023) have recently argued that different psychotherapeutic orientations tend to converge around five basic school-independent developmental functions that safety plays in ontogenesis and clinical practice: securing survival, facilitating restoration, promoting exploration, sustaining risk-taking, and enabling integration. The basic idea is that client change (i.e., successful psychotherapy) is homologous to functional child development (i.e., adaptive ontogenesis), with safety playing a pivotal role in the latter as well as in the former. However, even these safety functions do not yet provide a sufficiently comprehensive definition of the concept of safety in psychotherapy.

Safety as a multidimensional concept

Drawing from the psychotherapeutic literature, safety in psychotherapy can be organized along several overlapping and interactive dimensions and polarities, with the primary emphasis on a) function (survival vs. adaptive), followed by b) locus (external vs. internal), c) domain (affective vs. cognitive), d) state (static vs. dynamic), and e) perspective (individual vs. relational).¹

Survival vs. Adaptive. Safety encompasses vital functions that are critical for survival (defense) and adaptation, with a distinct focus on survival and adaptive aspects. In dangerous, unpredictable, and unknown environments, individuals are driven by the survival function of safety, which aims to alleviate distress by seeking safety: the possibility of being sufficiently protected from danger, injury, or loss (see also the concept of safe haven [Bowlby, 1988] or safeness [Gilbert 1995, 2004]) and the related inner experience of relief and comfort. In the context of danger and life threat, safety is sought through survival instincts mediated by sympathetic and dorsal vagal parasympathetic responses (e.g., fight, flight, freeze; Porges, 2021), mature and immature defense mechanisms (Cramer, 2015), preventive and restorative safety behaviors (Bowlby, 1988; Helbig-Lang & Petermann; Porges, 2021), and/or secure or insecure attachment (Bowlby, 1988). Prior to reaching safety, the individual’s experience may range from acute fear and distress to manageable levels of anxiety or vigilance. During the activation of the survival function, these experiences are primarily shaped by external factors such as environmental threats and, to a lesser extent, by supportive individuals.

In contrast, the adaptive function of safety involves a transition in which the ventral vagal parasympathetic nervous system and the pro-social systems – such as social engagement, attachment, and cooperation (Cassidy & Shaver, 2016; Cortina & Liotti, 2010b) – become dominant over the survival and defense responses, facilitating experiences ranging from immediate relief and relaxation to a progressive and deeper internalization of safety. As individuals move from seeking safety to feeling safety, they become more ‘satisfied’, and their focus shifts from survival to adaptation. Domination of the ventral vagal parasympathetic and pro-social systems allows the prefrontal cortex, responsible for higher-order thinking, to function more adaptively, facilitating the ability to navigate and cope with life's challenges (Porges, 2021). Importantly, the consistent experience of a calming, nurturing, familiar, and predictable environment allows individuals to increasingly internalize a feeling of safety: a sense of stability and predictability of the environment as well as trust and confidence in one’s resources for coping and navigating environmental challenges (see the concept of secure base; Bowlby, 1988). This personalized feeling develops gradually over time, in contrast to the more immediate relief experienced upon reaching safety. This distinction between survival-oriented safety, in which defense systems predominate, and adaptive safety, in which pro-social systems predominate, underscores the evolution from external reliance to internalized confidence in coping with life's challenges. When we talk later about the safe enough therapist and client, we will show how therapists play a central role in guiding individuals through this transition.

External vs. Internal. External safety is represented by environmental conditions (e.g., extra-therapeutic sources of safety, therapeutic setting, boundaries, and the overall background and interpersonal skills of the therapist) that reduce the risk of danger, injury, or loss and increase the likelihood of an individual’s internal safety. Internal safety, on the other hand, refers to the individual's subjective experience of being sufficiently free from danger, injury, or loss (e.g., the overall experience of comfort, trust, and security) in a given situation. In the early stages of development (and treatment), internal safety may be experienced primarily when external safety is sufficiently provided. Over the course of development, however, internal safety may become relatively independent of external safety to the extent that the latter can be experienced sufficiently often in relationships (i.e., internalized secure base [Bowlby, 1988], regulatory coping skills [Beebe & Lachmann, 2005, Fosha et al., 2009; Schore, 2009]).

Affective vs. Cognitive: The subjective experience of safety can be described in terms of its affective vs. cognitive domains. Affective safety consists of the bodily-based sense of being safe in a given situation (e.g., the immediate/intuitive sense of safety, trust, and comfort we may feel in a relationship). It is mostly associated with automatic, implicit, procedural, and presymbolic regulatory strategies (Lyons-Ruth, 1998; Schore, 2009). In contrast, cognitive safety refers to the reflexive assessment of how safe we are in a given situation (e.g., the conscious evaluation and differentiation of our experience of safety in that situation, the explicit meaning we give to it, the identification of its causes, the prediction of its consequences, and the ability to regulate our behavior accordingly). This type of safety is most often associated with more controlled, explicit, declarative, and symbolic regulatory strategies.

The affective domain of safety is the first to be experienced in a given situation, and it shapes cognitive safety through bottom-up processing. However, once the cognitive domain of safety is involved, it modulates affective safety through top-down processing. In essence, the more we experience safety at the affective level, the more our cognitive safety can be flexible and problem-solving oriented.

Stable vs. Dynamic. A stable state of safety refers to the establishment of a “safety baseline” that should remain consistent throughout therapy. The extent to which this is possible depends primarily on the trait-like dispositions of the participants (e.g., attachment style) and relies on the therapist’s establishment of an appropriate therapeutic setting by the therapist (e.g., firm but flexible therapeutic boundaries, confidentiality agreement, informed consent, crisis management protocols). In contrast, dynamic safety refers to how safety may increase or decrease (from its baseline) over the course of treatment due to the nature of the therapeutic relationship and process. The therapist is responsible for a moment-by-moment assessment and management of these state-like variations of safety during the treatment. These variations are related to, among other things, the client’s risk-taking exploration, rupture-repair patterns, and the resulting corrective emotional experiences. The therapist must ensure that the relational atmosphere and therapeutic interventions are aligned with the client's current level of safety (Safran & Kraus, 2014; Safran et al., 2011).

Individual vs. Relational. Safety can be viewed from an individual or relational perspective. The individual perspective refers to the experience of safety perceived by an individual at a subjective level (e.g., client and therapist considered alone) as a result of their independent self-regulatory strategies (Hane et al., 2008). In contrast, the relational perspective refers to the experience of safety shared by an interacting dyad at an intersubjective level (e.g., client and therapist considered in the context of their dynamic interaction) (see S. Salvatore et al., 2009). Relational safety is based on the individuals' interactive regulatory strategies (e.g., Tronick, 2007). It is important to emphasize that individual and relational safety interact continuously and reciprocally. The client’s and therapist's individual experience of safety depends on and simultaneously shapes their shared relational experience of safety. It follows that the experience of safety for both the client and therapist is continuously co-constructed through the mutual influence of their self and interactive regulatory strategies.

With regard to the above dimensions, it should be noted that more classical psychotherapy approaches may have emphasized certain polarities of the safety dimensions (e.g., cognitive-behavioral approaches focus on cognitive aspects of safety at the individual level, whereas psychodynamic and humanistic approaches emphasize affective aspects at the relational level). However, more contemporary approaches converge on the idea that safety is a complex phenomenon resulting from the dialectical interplay of both polarities of the different dimensions described above (for a review, see Podolan & Gelo, 2023).

From what we have discussed, it is clear that defining safety in psychotherapy is challenging because it is a multifaceted concept that interacts with numerous psychological, emotional, and relational factors. It involves both affective and cognitive aspects, self- and interactive regulatory strategies (including reflective functioning about the self and others), and has both an internal and external locus. In addition, safety is both an individual and relational experience that can be stable over time while also exhibiting contextual variations that aim to both provide protection and promote exploration. Thus, effective safety in psychotherapy requires striking a delicate and dialectic balance between these multiple dimensions and their polarities. This complexity highlights the dynamic and evolving nature of safety within the therapeutic process and underscores the importance of attending to a wide range of factors to ensure a clinically productive therapeutic relationship.

Safety in change-promoting relationships

Building on the multidimensional definition of safety described above, we elaborate on the concept of the safety zone (Freeman & Dolan, 2001) to better understand the nuanced role of safety in change-promoting relationships. We use the term safety zone refer to the actual range of self- and interactive regulatory strategies (individual-relational perspectives) that enable individuals to maintain sufficient homeostasis in their biopsychosocial transactions with the internal and external environment (internal-external loci). It encompasses a variety of states that are not experienced as perfectly safe but rather range from being almost perfectly safe to being safe enough (see Mitchell, 2003). From a neurobiological perspective, the safety zone can be said to correspond to an individual’s range of optimal arousal (also called window of tolerance) between parasympathetic hypoarousal and sympathetic hyperarousal, within which affects and emotions can be experienced and processed effectively so that higher cognitive processes may develop and operate (affective-cognitive domains) (Siegel, 2012; Slade, 1999; see also Fosha et al., 2009). As such, the safety zone influences the individual’s balance between safety and novelty-seeking behaviors (survival-adaptive functions). In addition, the safety zone is responsible for both an individual’s trait-like tendency to experience safety (e.g., how we tend to feel safe in a given situation) and its state-like variations over time (e.g., the transitions from stability to chaotic variability to new adaptive stability during treatment) (stable-dynamic states). In psychodynamic psychotherapy, this concept has been used most prominently by Schore (2003) in his theory of affect regulation in development, psychopathogenesis, and psychotherapeutic change processes. From a cognitive-behavioral perspective, the safety zone may refer to a self-regulatory capacity (Carver & Scheier, 1998) within which we can control and regulate our attention, thoughts, emotions, and behaviors (Baumeister et al., 2007). From a humanistic perspective, the safety zone may include existential trust in the continuity of being, relationships with significant others (Erikson, 1993), and the material, social, and technological environment – trust that future events will be sufficiently favorable to one’s interests (Edmondson, 1999).

Throughout ontogenesis, each individual develops their safety zone within which they can feel, think, and behave as they are used to. The difference is that, in the case of maladaptive patterns of functioning, the safety zone is too narrow and rigid. The more an individual has been exposed, especially early in development, to stressful, neglectful, or traumatic experiences not sufficiently repaired by a good enough caregiver, the more often they will have experienced excessive organismic hyper (e.g., fight or flight) or hypoactivation (e.g., freeze) (Schore, 2009). As a consequence of this, we will observe the development of a safety zone that is increasingly rigid and narrow (e.g., defensive); that is, the range of regulatory strategies available to the individual will be too restricted, with the consequence of lower levels of adaptation. In other words, the lack of enough parental sensitivity and responsiveness contributes, to different extents, to the development of self- and interactive regulation strategies that narrow and stiffen the scope of the safety zone (Duros & Crowley, 2014). In these cases, an individual’s safety zone may turn into a sort of “safety enclave” clinically known as defensive organization (O’Shaughnessy, 1981), psychic retreat (Steiner, 1993), or safe zone (Barbanell, 2006) that preserves homeostasis in less effective ways.

From this perspective, the goal of the process of change in supportive therapy might be seen as making the affective, cognitive, and behavioral regulatory patterns within a client’s safety zone more effective (e.g., improving the coping skills and reducing distress). On the contrary, the process of change in expressive therapy can be seen as making the above patterns more flexible and/or expanding the safety zone and achieving more structural changes (facilitating long-term changes in personality and self-awareness).

Safety relationships: and change-some promoting school-independent therapeutic relational principles

In this section, we propose some school-independent principles that address the calibration and personalization of safety in change-promoting therapeutic relationships. These principles are summarized in table 1. Our approach draws on commonalities across therapeutic schools and empirical evidence (see Podolan, 2020; Podolan & Gelo, 2023). We focus on the principles of safety in allowing a clinically productive therapeutic relationship and how this relates to therapeutic change. In doing this, we concentrate on how a safe relationship facilitated by a safe therapist can help the client feel safe and thus foster therapeutic change.

Table 1.

School-independent principles that deal with the role of safety in the psychotherapeutic change process with a focus on relationship

Safe enough therapist <– – – – –> Safe enough relationship <– – – – –> Safe enough client
Therapist’s effects exceed treatment effects. Therapist’s developmental, healing, and training processes have a more significant impact on treatment than therapeutic techniques. ↕ Personality safety features have a decisive impact on the psychotherapeutic change process. Treatment is more effective if the therapist has or has gained a secure attachment style, secure self, secure identity, and secure self-esteem. ↕ The therapist’s actual safety is a precondition for effective psychotherapy. To cope with environmental transactions, accept, understand, and process their own uncomfortable feelings, respect the client’s suffering, express compassion and positive emotions, and maintain therapeutic boundaries, the therapist must be safe in the present moment of the clinical setting, both from a legal, financial, and personal perspective.	Securing the client’s survival and establishing trust requires the therapist to utilize prosodic voice and warm eye contact, supported by holding, attunement, containment, enduring defenses, promptly repairing mismatches, and maintaining therapeutic boundaries. Additionally, the therapist should anticipate the client’s stage of precontemplation or contemplation and employ non-defensive and supportive interventions. ↕ Facilitating healing and restoration is primarily achieved through the therapist’s utilization of empathy, unconditional positive regard, responsive caregiving, mentalizing, and sourcing safety measures. The therapist employs predominantly non-interpretative and invitational interventions. ↕ Passing the client’s safety tests is crucial in order to preserve safety during therapy. To achieve this, the therapist adapts his or her approach to ensure the client feels secure. The therapist utilizes open, non-interpretative, and non-defensive approaches, carefully considering the pros and cons of interpreting any defenses displayed by the client. ↕ Forming an alliance and reaching an agreement on goals and tasks is essential for a safe relationship. The therapist considers the client’s therapeutic zone of proximal development and takes into account external factors that may impact the therapeutic process. He or she assists in increasing the client’s self-awareness and engages in a co-assessment of the client’s problems, examining the costs and benefits of their lifestyles. Additionally, the therapist informs the client about the potential variations in therapeutic difficulties and outcomes. The ultimate aim is to establish a shared understanding and commitment to the agreed-upon goals and tasks.	Promoting exploration. Utilizing exploratory interventions. Employing work-enhancing strategies. Encouraging curiosity and deeper exploration of vulnerable material. Acknowledging successes and preparing for the action stage. Moving towards the boundaries of clients’ safety zones. ↕ Tolerating and sustaining risk-taking. Identifying realistic goals/tasks and the scope of exploration while fostering cooperation during challenging moments. Encouraging clients to take risks while maintaining a sufficient level of safety. Repairing ruptures through complex empathy, intersubjectively attuned containment, and cooperative adjustment of goals and tasks. ↕ Managing the risks and navigating the balance between safety and danger. Implementing restructuring interventions that are curiosity-invoking, playful, creative, and sometimes humorous. Moving predictably and rhythmically within the zone of proximal development, at an appropriate pace, and in affectively tolerable doses. Destabilizing and co-reorganizing habitual processes into more adaptive ones. Emphasizing recycling as part of the process. Viewing ruptures and relapses as natural occurrences that are co-regulated with empathy, mentalization, acknowledgment of failures, and reparations through attunement or the creation of additional safety resources. ↕ Consolidating new experiences. Maintaining adequate safety while navigating novelty and danger. Reconsolidating experiences. Rearranging relational understanding in a creative and playful manner. Expanding dyadic consciousness and fostering synchronous meeting of minds. Assisting clients in affectively re-experiencing certain past or present relational events in a more adaptive manner. ↕ Maintaining new patterns. Adopting a complementary stance aimed at synthesizing the change process across stages. “Pushing where it moves” – working around and slightly beyond the client’s assimilation level to facilitate the re-emergence of more functional relational patterns. Emphasizing new opportunities to overcome resistances. Implementing working-through, counterconditioning, stimulus control, or contingency management techniques. Developing coping strategies to prevent relapses. Implementing rewards for avoiding relapse. Ensuring that clients and their significant others value the changes in newer and more adaptive self-awareness/narratives.
Consequences: The therapist’s broad safety zone facilitates holding, containing, and fosters the development of a safe environment and a trusting affective bond.	Consequences: The client’s safety zone internalizes new relational aspects of safety established by the therapist’s safety zone, such as the holding environment, safe haven, and secure base. As a result, clients feel increasingly safer, cared for, and capable of deactivating their defenses. This leads to a mutual agreement between the therapist and client regarding goals and tasks.	Consequences: Exploring and facing novelty, as well as tolerable danger, insecurity, and frustration, while experiencing enough safety has transformative effects. New self-regulatory strategies emerge and facilitate the assimilation of novelties into one’s safety zones, leading to corrective emotional experiences and innovative moments. Reconsolidating and recycling these experiences in an environment that is safe enough, adaptive, and dynamic helps clients affectively re-experience certain events more adaptively, thereby increasing the range, resiliency, and flexibility of their safety zones.

Note. The bidirectional arrows in the table depict the complementary and interdependent relationships between the various aspects of safety described.

Safe enough therapist

Therapists should provide a safe environment to clients. To do this aim, they should be safe enough themselves. Neuroscientific research suggests that humans detect safety cues mainly unconsciously through neuroception, which automatically scans the environment for danger and safety (Porges, 2021; Schore, 2012). Clients perceive therapists’ safety cues mainly through their voice and eye contact (Porges, 2021). A safe voice is characterized by authentic, soothing intonation with prosodic, playful, resonant, low-pitched, timbral, rhythmic, and melodic patterns. The therapist’s warm and engaging eye contact (Hymer, 1986), with occasional head nods and adjustments regulated according to the client’s needs, is perceived as safe by the client (Curtis, 1981). In order for this to happen, the therapist should be characterized by a safety zone that is broad and flexible enough (depending, for example, on their attachment style, psychotherapy training and personal therapy, professional experience, and current life circumstances) to facilitate the client’s experience of safety and to better regulate the therapeutic relationship.

Importantly, there is the idea that some therapists achieve better outcomes than others, regardless of the modality. Wampold’s extensive research across therapeutic modalities has concluded that therapist effects “generally exceed treatment effects” in terms of reaching therapeutic alliance and therapeutic outcomes (Wampold & Imel, 2015, p. 176). Such therapists’ effects come from their personality, relational qualities (e.g., warmth, empathy, self-disclosure without crossing therapeutic boundaries), and training (e.g., insights into one’s feelings, thoughts and behaviors, self-discovery, self-awareness, knowledge), which develop through their own’s developmental and training process before becoming a psychotherapist (Messina, Gelo, Gullo, et al., 2018; Messina, Gelo, Sambin, et al., 2018b; Rønnestad et al., 2018; Zerubavel & Wright, 2012). Yet, licensed psychotherapists may still have problematic personality traits, insecure attachment styles, or defense mechanisms that may make therapy less effective or ineffective (Schmidbauer, 2018; Victor et al., 2022).

Therapist’s safety is a precondition for therapy

If the therapists are not safe enough (i.e., if their safety zone is not sufficiently broad enough and flexible enough), they are unsuitable to conduct psychotherapy because they cannot cope effectively with environmental transactions (Porges, 2021). However a sufficient inner state of safety enables therapists to accept their own uncomfortable feelings, respect the client’s suffering and experience, and express compassion and positive feelings towards others. Kernberg (2014) radically postulated that “if the safety of the therapist cannot be established and maintained, the treatment is not possible” (p. 95). Other scholars concluded that it is the therapist’s own safety that is essential for the progress and effectiveness of therapy (Yeomans et al., 2015) and for a deep and effective engagement in one’s own and the client’s unconscious processes (Davies, 1999). In this regard, if the therapist is not safe enough, the therapist may need supervision or may need to interrupt or terminate the therapeutic process (G. Salvatore et al., 2024). Such insecurity of the therapist may arise, for example, from the therapist’s personal self (e.g., the therapist is overwhelmed by current physical, economic, or legal threats posed by others; the therapist’s countertransference or beliefs affect the therapist’s ability to practice competently) and or from the professional self (e.g., the therapist lacks the needed skills or expertise; the therapist cannot maintain therapeutic boundaries).

Therapist’s safety improves therapy significantly

attachment research has provided increasing evidence that securely attached therapists may be crucial for effective psychotherapy. Unlike insecurely attached therapists, secure therapists seem to form stronger alliances, have better outcomes, respond and repair ruptures more empathically, use countertransference more effectively, and are able to intervene with more compassion (Degnan et al., 2016; Dozier et al., 1994; Rubino et al., 2000). As opposed to insecurely attached therapists, secure therapists seem to be more effective in treating clients with personality disorders (Bruck et al., 2006), more sensible in dealing with their overt demands or needs (Slade, 1999), more able to act as a secure base (Mikulincer et al., 2013; Slade, 1999), and better equipped to manage countertransference reactions (Gilbert & Stickley, 2012). Securely attached therapists are also better at creating therapeutic alliances (Bucci et al., 2016), exploring the depths of the client’s world more thoroughly (Romano et al., 2008), and making narratives more coherent (Talia et al., 2019). It seems that a movement toward greater attachment security is central to creating a safe relationship and achieving favorable therapeutic outcomes (Mikulincer et al., 2013).

Scholarly literature characterizes therapists’ safety also beyond strictly attachment-related concepts such as secure self (Kohut, 1971; Steele, 2007), secure identity (Erikson, 1993; Lichtenberg, 2018), secure interpersonal boundaries (Epstein, 1994), secure sense of self-efficacy (Maddux, 2016), and secure self-esteem (Kernis, 2003). These therapists maintain higher levels of internal safety, tend to be more effective in dealing with internal or external difficulties (Epstein, 1994), and exhibit greater resilience compared to those with characteristics of fragility, insecurity, or defensiveness in the self (Kernis, 2003; Kernis et al., 2008).

The above points suggest that the therapist’s individual safety may enhance the client’s individual safety by co-creating a safe enough relationship so that both the participants’ needs for protection and exploration can be met at different moments in treatment.

Safe enough relationship

Securing survival and establishing trust

When the client first comes into contact with the therapist, there are various factors that may cause insecurity in the client (e.g., pre-existing distress, past trauma, readiness to change, threats posed by the therapist’s gender, beliefs, values, office, etc.). Therefore, one of the most critical functions of the therapist is to secure the client’s psychological survival through a safe environment (Podolan & Gelo, 2023).² If this is successful, a shared relational safety zone can be established between the client and the therapist.

The psychotherapeutic literature contains numerous relational concepts considered to contribute to the establishment of a safe environment that promotes a shared relational safety zone. These include, for example, the holding environment (Winnicott, 1971), safe heaven (Bowlby, 1988), affective attunement/resonance (Stern, 1985), mirroring (Brussoni et al., 2012; Kohut, 1971), containment (Bion, 1962), synchronous interaction (J. Levy & Feldman, 2019), rhythm of safety (Tustin, 1986), moments of meeting (Sander, 1992), directional fittedness (Bruschweiler-Stern et al., 2010), and presence (Geller & Porges, 2014). Within this context, therapists sufficiently accept their clients, empathize with their suffering, maintain confidentiality, try to understand them, act in their interest, and remain in “nurturing” roles. They must endure clients’ defenses (e.g., devaluation, denial, and ignorance of their problems), anticipate the precontemplation or contemplation stage (Prochaska et al., 2013), and tolerate their own uncomfortable experiences.

In order to build trust through safety, therapists consider any interventions that might in any way compromise the clients’ safety in the early stages of therapy (Messer & Gurman, 2011). From this perspective, therapists tend to prioritize non-defensive, warm, and supportive (e.g., empathic, nonjudgmental) interventions while refraining from explanatory, expressive, interpretative, or restructuring interventions (Ackerman & Hilsenroth, 2003; Locati et al., 2020). Recognizing that unavoidable relational ruptures, mismatches, miscoordinations, or misapprehensions may occur (Tronick, 2007), therapists aim to address these issues promptly, adequately, and compassionately (Newhill et al., 2003; Safran & Kraus, 2014; Safran et al., 2011). These therapeutic qualities promote the establishment of a good enough relationship also through sufficiently flexible, permeable, consistent, predictable, and reliable therapeutic boundaries (Gabbard, 2016; Langs, 1992).

Facilitating Healing and Restoration

Safety also facilitates healing and restoration, for example through empathy, unconditional positive regard, and authenticity (Rogers, 1995), responsive caregiving (Ainsworth, 2010), security priming (Gillath & Karantzas, 2019), co-regulation of inner processes (Beebe & Lachmann, 2005; Porges, 2021), mentalizing (Fonagy & Allison, 2014), or sourcing of safe places and installing or internalizing safety sources (Shapiro, 2017). This points to the interaction that exists between trait-like aspects of safety (e.g., therapist’s and client’s attachment styles and interpersonal skills) and the state-like variations that are possible in the context of the former (e.g., client’s alliance ruptures and therapist’s repairs).

To renew and restore clients’ stability and control over their lives, therapists should avoid disrupting existing patterns that activate anxiety, defenses, or resistance. There is no need to “push” clients through any techniques or exploration until they feel safe enough to do so. In this stage, therapists should use invitational, supportive, and largely non-interpretative interventions, such as self-state conjectures, empathic validations, and joinings (Talia et al., 2020), or other techniques aimed at stabilizing the clients and facilitating their safety (e.g., through mindfulness of bodily sensations), and making them feel cared for.

Testing the safety of the therapist

Before feeling safe, some clients test the therapist’s safety through covert or overt tests. These tests aim to determine if the therapist is sufficiently authentic, genuine, trustworthy, competent, or strong enough to hold the client’s inner states (Rappoport, 1997; Siegel & Hilsenroth, 2013). In order to pass the tests, a relatively open, non-interpretative, and non-defensive approach by the therapist is generally advisable (Rappoport, 1997; Siegel & Hilsenroth, 2013). Yet, therapists must tailor their approach appropriately to the patient’s level of fragility. It is important for therapists to avoid focusing on clients’ defenses as this can be experienced as excessively frustrating and thus counterproductive at this stage. Instead, therapists should co-create safety by being sufficiently empathic and attuned to clients’ wishes, fears, or origins of their defenses (Frederickson, 2020; Rappoport, 1997). They may simultaneously help clients gain insight into their conditions, identify the barriers to change, and weigh the pros and cons of behavioral change (Prochaska et al., 2013). However, some clients may use defensive strategies (e.g., projecting their will to heal or punishing superego onto the therapist), leading to a feeling of insecurity. This can create a sense of danger associated with forming an alliance, regardless of the therapist’s openness, supportiveness, non-interpretative approach, or nurturing nature. In these cases, the therapists may need to sensitively and empathically interpret the client’s behavior (i.e., help the client see the defenses and their price) in order to start loosening their pathogenic behavior.

Forming tasks alliance and agreeing on goals and

As clients move from precontemplation to contemplation (towards the formation of alliance), they may feel safer when the therapist helps raise their self-awareness and co-assesses how they feel and think about themselves and their problems (Prochaska et al., 2013). The therapist should inform the client that psychotherapy may lead to better outcomes when both parties have a shared understanding of and commitment to the goals and tasks. The parties should find meaningful ways of dealing with clients’ problems, know their roles and expectations in therapy, and trust that the way they work on the problems is correct and appropriate (Horvath et al., 2011). To preserve transparency, the therapist should explain to the client that, although good cooperation typically leads to better outcomes, the speed, extent, or precise way of reaching such outcomes may vary from case to case. To outline therapeutic goals and prepare the client for the active phase of therapy, the therapist should adopt a solid and experienced role encouraging deeper exploration or action (Hill, 2020; Prochaska et al., 2013). By joining the clients in their (self)analysis of the costs and benefits of their lifestyles, both parties may collect more information in order to form an alliance and reach an agreement on therapeutic goals (Levitt & Williams, 2010; Williams & Levitt, 2007). While updating agreements, both parties must carefully, collaboratively, and continuously assess the client’s therapeutic zone of proximal development (Leiman & Stiles, 2001), as well as the client’s extra-therapeutic factors that significantly contribute to the therapeutic outcome (Lambert, 1992; Messer & Gurman, 2011). Once the parties form a bond, alliance, or confident collaboration (Siegel & Hilsenroth, 2013), the client’s sense of safety and potential for agency will significantly rise (Siegel & Hilsenroth, 2013).

Safe enough Client

Engaging in exploration

At this point, the client has begun to consolidate their safety zone through positive and safe enough experiences (e.g., holding environment). The therapist has passed the client’s safety tests. The client has internalized these therapeutic experiences and began to perceive therapy as a safe haven (providing soothing and reassuring) and secure base (promoting curiosity and encouraging deeper exploration). The parties also started forming an affective bond and reaching an agreement on goals and tasks. The platform of a safe relationship (Kirsha, 2019; Levitt & Williams, 2010) enables now the client to move toward the boundaries of their safety zone. Trying to approach what has been defined as the road less traveled (Peck, 1997), “fault lines [of self-experience] where interactive negotiations have failed, goals remain aborted, negative effects are unresolved, and conflict is experienced” (Lyons‐Ruth, 1999, p. 607), “the edges of the regulatory boundaries of affect tolerance” (Schore, 2009, p. 131), and “edges of what [is] bearable” (Wilkinson, 2010, p. 141). The client became less prone to activate maladaptive defenses and more prepared to dive into a deeper exploration of unknown or vulnerable material.

When the client’s will to engage is activated, the therapist may start using – slowly, sensitively, and gradually – more exploratory interventions (e.g., direct questions, clarifications) or work-enhancing strategies (e.g., explaining the value of therapy, encouragement to speak openly). They cannot be excessively threatening or too inconsistent with the client’s existent patterns of functioning. This allows the client to accept and assimilate new information effectively. In Prochaska’s terms, the parties progress from the preparation to the action stage, working towards the agreed goals. In achieving this, the therapist may encourage the client to seek social support, identify safe and constructive alternatives, and tap into their own resources (Prochaska et al., 2013).

Tolerating and sustaining risk taking

At this stage, the client begins to liberate the self and is willing to take risks to achieve change. In order for the client to tolerate risk-taking, the therapist must first help the client learn how to navigate the dialectic between safety and danger (Segalla, 2018). The therapist must help the client to accept that, for things to get better, they may first experience painful yet sufficiently safe turbulences (Gelo & Salvatore, 2016; Pascual-Leone et al., 2016). Excessive reliance on safety (e.g., supportive interventions) might lead to missing new opportunities to revise maladaptive patterns; on the contrary, excessive emphasis on risk-taking (e.g., expressive interventions) might stimulate instability and trigger resistance. For the client to tolerate risk-taking in therapy, clients and therapists should agree first to explore maladaptive patterns and then to expand the client’s safety zone through more structural changes. They should also make a plan, clarify the goals, assess the potential risks, and agree on how to cooperate during the plan’s realization. The unavoidable risks should be balanced with the engagement of already existing resources and the promotion of new ones, together with the increasing fostering of the client’s identity. Coping strategies for anxiety, activations of defenses, alliance ruptures, or dyselaborations (e.g., filling, hedging, or fleeing from topic) should be discussed in a cooperative way. While encouraging the client to take risks, the therapist must repair any ruptures by complex empathy, intersubjectively attuned containment, or cooperative rearrangement of treatment goals and tasks. At all times, the therapist must prioritize continuous re-co-creation of the client’s safety to protect both parties amidst instabilities, insecurities, and dangers. The client’s right to slow down, stop, or withdraw from the plan shall also be guaranteed.

Managing safety and the danger risks and working between

The co-created safer and consequently more resilient relationship enables both parties in the therapy to be in a position of “safe uncertainty” (Mason, 2015), which is “safe but not too safe” (Bromberg, 2006, p. 289). Thus, clients and therapists may increasingly work “in the gap between perceived security and perceived novelty” (Greenberg, 1991, p. 130). The paramount importance of therapeutic work rests now in “establishing safety while confronting danger” (Eldridge, 2018, p. 602). Since the client now trusts that the “danger is safe” (Eldridge, 2018, p. 596), they will be more willing and able to engage in the “necessary danger” involved in any structural change (Carr & Sandmeyer, 2018; p. 557) by making a constructive use of what Bromberg (2006) defines “safe surprise” (p. 95).

The goal of this phase is to destabilize clients’ maladaptive patterns and reorganize and accommodate them into more adaptive ones (see Gelo & Salvatore, 2016). The therapist challenges clients’ maladaptive processes using restructuring interventions and expressive interventions such as transference interpretations, role-playing, cognitive restructuring, desensitization techniques, interpretations and deactivations of defenses, safety enhancement techniques, exposure techniques, and memory reprocessing. The goal is to co-reorganize these processes and collaboratively create new and more adaptive ones (Frederickson, 2020). While doing so, both parties must work with realistic treatment tasks/goals that do not go beyond the therapeutic zone of proximal development (Stiles et al., 2016). The therapist should acknowledge that the change process can cause discomfort while reassuring the client that it can ultimately lead to a positive outcome. It is important for the client to learn to cope with and tolerate this discomfort. This is coherent with the observation that, sometimes, “it gets worse before it gets better” (Pascual-Leone et al., 2016, p. 338). Thus, worsening of the client’s state may be a precondition for the subsequent increase of their self-efficacy, self-regulation, resilience, ability to tolerate danger, cope with stress, and engage in risk-taking explorations.

This process may produce turbulent emotional insights, “aha moments,” retrieval of encoded implicit memories, or surprises that challenge dominant self-narratives. The client may experience new relational experiences by recognizing dysfunctional patterns as familiar or functional patterns as unfamiliar.

These activities become safer if they are co-conducted – like on a child swing – through synchronous, predictable, rhythmic, and regular oscillations between safety and danger (Koole & Tschacher, 2016; J. Levy & Feldman, 2019). Curiosity-invoking, playful, creative, or humorous interventions may further enhance the client’s safety during such a process (Johnson et al., 2015). Importantly, the disrupting or reorganizing interventions must be made collaboratively at the right pace in affective tolerable doses within a zone of safe separation where the disruption is tolerable and desired change is realistic (e.g., between the client’s safety zone and the therapeutic zone of proximal development).

To preserve safety, both parties should anticipate that clients’ progress may not be linear or that there will be a need to recycle several times through various stages (e.g., precontemplation, contemplation, preparation, action) before achieving long-term changes (Prochaska et al., 2013). Relational ruptures and or symptom relapses may be viewed as natural. In this context, therapists should co-regulate the client’s anxiety with empathy, mentalize the relapse or rupture, avoid restructuring interventions, admit their own contributions to failures, and repair them through intersubjective attunement or creation of additional safety resources (Davis et al., 2013).

Consolidating new experiences

The client’s ability to experience sufficient safety while facing novelty and danger (e.g., expressive intervention, eye movement desensitizing, empty chair work) may have transformative effects. It allows new regulatory strategies to emerge and facilitates the assimilation of novelties into one's own mental functioning through corrective emotional experiences (Sharpless & Barber, 2012), now moments (Stern, 1985), and innovative moments (Gonçalves et al., 2009). Such synchronous meeting of the minds on the platform of a safe relationship may creatively and playfully rearrange the parties’ implicit relational knowing (Lyons-Ruth, 1998), expanding their dyadic consciousness (Tronick, 1998). These reconsolidating experiences help the client to affectively (re)experience certain past or present relational events more adaptively (Castonguay & Hill, 2012). At the same time, they also provide a neurological basis for a long-lasting or permanent change (Ecker, 2015) of a client’s safety zone, which is most pliable to alterations and reconsolidation during a window from ten minutes up to six hours (Alberini & LeDoux, 2013; Ecker, 2015).

Maintaining new patterns

The client’s old patterns and dominant “community of internal voices” (Stiles, 2001, 2011) may resist assimilating the new non-dominant voices (i.e., the newly emerged patterns) into the client’s therapeutic zone of proximal development (i.e., the expanded safety zone) because they are more familiar to the client. Although clients may resist assimilating novel regulatory patterns that may have already emerged, the latter will continue to offer new opportunities for alternative functioning. To promote their assimilation and maintenance, it may be necessary to repeat the therapeutic work. To these aims, the therapist must continue to “push where it moves” (Leiman & Stiles, 2001, p. 315) and work around and slightly beyond the client’s assimilation level to facilitate the re-emergence of more functional patterns already experienced by the client. Their consolidation may be enhanced when the client experiences the previously dominant voices as less functional and the new voices (which were previously projected, repressed, or isolated) as more adaptive and functional (Ribeiro et al.., 2013). The maintenance of such newly emerged patterns of functioning may be characterized by working through counterconditioning, stimulus control, or contingency management. These become more effective when the therapist helps the client to see that maintaining the change supports such aspects of the self that are valued by the client or significant others (Prochaska et al., 2013).

Therapists should view the desired change as an ongoing process, supporting clients in maintaining insight, developing coping strategies against relapse (e.g., identifying obstacles to change and triggers that lead to old ways of functioning), and rewarding themselves for avoiding relapses. The use of complementary therapeutic stances can facilitate the client's assimilation and consolidation of changes experienced throughout the therapeutic process. Through repeated cycles of such experiences in a safe enough and increasingly familiar environment, clients can transform their narrower and excessively defensive safety zones into broader and more flexible ones with newer and more adaptive self-awareness/narratives.

Empirical processes research on safety and change

The above school-independent relational principles on the role of safety in the therapeutic change process align with existing quantitative and quantitative empirical findings in psychotherapy research (see Gelo, Pritz, et al., 2015a, 2015b). In qualitative research, for example, Timulak and colleagues concluded that safety, along with reassurance and support, can be an important factor clients find helpful in psychotherapy (Timulak, 2007; Timulak et al., 2010). Kirsha (2019) examined corrective experiences in psychoanalytic psychotherapy and found that clients perceive a safe environment as the most beneficial experience. They viewed it as a platform for self-discovery facilitated by the therapeutic relationship, which allows clients to explore, change, and develop (for analogous results in cognitive-behavioral and systemic therapy, see Dourdouma et al., 2020; Gelo et al., 2016). The role of safety within the therapeutic relationship in a sample of eminent psychotherapists was explored by Levitt and Williams (2010); they report that “safety within the psychotherapeutic relationship was identified as a central element in creating client change to the extent that in-session risk-taking was important in that orientation” (p. 337). Levitt et al. (2016) found that professional boundaries and reliability in contact, as well as therapists’ authentic care and acceptance, help clients feel safe to explore vulnerable and threatening themes and, consequently, constitute key elements for the change process. Similar studies found that the client’s feeling of safety rises through agreement on therapeutic goals, the therapist’s ability to join the client, and the therapist’s self-disclosures that do not cross boundaries, while the client’s safety falls when therapy lacks mutual goals or when the therapist fails to understand the client (Levitt & Williams, 2010; Williams & Levitt, 2007).

Quantitative studies within attachment theory have consistently demonstrated that secure attachment in both clients and therapists leads to stronger alliances and better outcomes compared to insecurely attached individuals (Degnan et al., 2016; Diener & Monroe, 2011; K. N. Levy et al., 2018). Securely attached therapists are more empathic when responding to clients and repairing alliance ruptures than insecurely attached therapists (Rubino et al., 2000). Unlike insecurely attached therapists, securely attached ones use countertransferences more effectively and are more willing to intervene effectively in ways that are uncomfortable for themselves (Dozier et al., 1994). Talia and colleagues showed that, unlike insecure therapists, therapists with a secure attachment behavior use more often intersubjective and engaging interventions (i.e., self-state conjectures, empathic validations, joinings, and disclosures of own inner states) and less often detaching or coercive interventions (Talia et al., 2020). Similarly, unlike unsecure clients, clients with secure attachment behavior present higher levels of rupture repairs (Miller-Bottome et al., 2019).

Some other studies have been conducted on the role of safety in psychotherapy without specific reference to attachment theory. For example, Siegel and Hilsenroth (2013) focused on clients’ in-session reports and showed that safety significantly correlates with the bond between the client and therapist and their confident collaboration. Friedlander’s studies of family systems found that clients’ safety is associated with a place where one expects something good (e.g., new learning and experiences), senses comfort and emotional connection when taking risks, is open and vulnerable, can manage conflict without harm, does not need to rely on defenses, and can share a sense of purpose (Friedlander, Escudero, & Eatherington, 2006; Friedlander, Escudero, Horvath, et al., 2006). Similarly, studies by Escudero and colleagues showed that a client’s safety is associated with such repairs of alliance ruptures, where the therapist first ensures safety, then enhances emotional connection and promotes a shared sense of purpose (Escudero et al., 2008, 2012).

Conclusions

Safety is crucial for effectively processing and organizing experiences within relational contexts. The present paper intended to contribute to the relational orientation of psychotherapy by focusing on the concept of safety from a transtheoretical perspective. We discussed some school-independent principles concerning the role of safety within change-promoting therapeutic relationships. The relational aspects of safety work adaptively and dynamically to facilitate change. To establish a safe and effective therapeutic relationship, the therapist must prioritize their own safety first. This means that the therapist must have a sufficiently broad safety zone and be able to adapt their regulatory strategies to meet the client’s needs. The safety of the therapist is a precondition for the latter to be increasingly perceived by the client as a safe haven. The therapist must be safe enough to “meet” the client within their safety zone by accepting, attending, holding, and containing all the client’s habitual ways of thinking, behaving, and feeling, including their dysfunctional coping strategies and related distress. This fosters the development of a trusting affective bond between the client and the therapist. In this way, the therapeutic relationship slowly becomes a safe environment characterized by attunement and intersubjective empathic validation on the background of the therapist’s mentalizing attitude. This creates additional safety sources for the client that they can incorporate into their safety zone and leads to an alliance increase. The internalization of such experiences in the therapeutic relationship allows the therapist to be perceived as a secure base. Clients can then explore novel strategies and take risks at the boundaries of their safety zones.

If the therapist and the client are able to disrupt the client’s usual patterns of functioning in a collaborative and tolerable way, transformative safety may emerge, which implies an expansive accommodation of the client's existing regulatory strategies. Within this process, if the resulting experiences are co-regulated, meanings are co-constructed, and alliance ruptures are sufficiently co-repaired, the client may not only renew their psychological homeostasis but also experience new ways of feeling, thinking, and behaving and assimilate them into more adaptive ways of functioning within their safety zone. Recycling such experiences enables the client to better tolerate challenges and engage in biopsychosocial learning processes. This also allows the client to develop more resilience and consolidate the novel patterns of functioning, ultimately improving or enlarging their safety zone and adaptive capacity.

In essence, the relationship does not need to be perfectly safe (generating steady forms of homeostasis) but rather safe enough (generating dynamic forms of homeostasis that facilitate the client’s self-discovery, self-awareness, and adaptation). Thus, safety seems to be change-promoting within relational contexts if it is well-modulated in its intensity and scope and well-timed with regard to specific types of clients, their limitations, and specific phases of therapy.

Focusing on such school-independent principles of safety within change-promoting therapeutic relationships may contribute to the current relational orientation in psychotherapy, with relevant implications for practitioners. Awareness of such principles and understanding how they evolve and adapt over the course of treatment can help therapists of different orientations develop a common identity and language about what makes therapeutic relationships therapeutic (Markin, 2014). As a result, therapists who typically focus more on the role of school-specific treatment methods may be encouraged to acknowledge more fully the importance of the school-independent relational aspects of safety in promoting change. Conversely, therapists who tend to emphasize the role of the therapeutic relationship in promoting change may find in such safety-based relational principles a school-independent theoretical background that legitimizes their actions beyond what is offered to them by reference to their specific orientation alone. We believe this could be useful in promoting a constructive dialogue between advocates of empirically supported treatments and empirically supported relationships to benefit our understanding and practice of psychotherapy. Moreover, this may help to better understand the extent to which both psychopathology and psychotherapy are socially constructed processes (see Gelo, Vilei, et al., 2015) in which safety plays a central role. Similarly, psychotherapy training programs could be encouraged to better integrate the teaching of empirically supported treatment and relationship principles (Markin, 2014; Norcross & Lambert, 2014).

The main limitation of this article is that the definition of safety (along with its dimensions and polarities) and the identified change-promoting relational principles are not exhaustive. Although we trust that we have analyzed and covered the topic sufficiently and based on the current state of knowledge in psychotherapy, we acknowledge that the nature of safety may continue to maintain its heterogeneous application and elude a more universal definition and understanding of its nuances. Safety and the process of change in psychotherapy remain highly complex and diverse. Individual clients and therapists bring unique experiences, backgrounds, and needs to the therapeutic setting. Therefore, safety and its change-promoting aspects (including the interplay among them) may include other aspects that are specific to each client's circumstances, personality, and therapeutic goals. Failure to acknowledge this complexity could limit our understanding of safety and the change process in psychotherapy and hinder our ability to effectively address the diverse needs of clients in therapy.

Future research should focus on better articulating the extent to which safety affects change-promoting therapeutic relationships within and across theoretical orientations through both theoretical and empirical research (see Gelo et al., 2009, 2020a, 2020b). Safety-related dimensions and variables should be identified and assessed during therapy sessions so that both the client and therapist can better understand their roles in shaping beneficial therapeutic relationships. To this end, procedures of both quantitative and qualitative text analysis should be employed (see Gelo et al., 2012). Particularly interesting would be the within-session identification and assessment of neurobiological markers of safety of both clients and therapists to explore their role in shaping specific safety-related dynamics. To better understand how therapists can strategically shape the therapeutic relationship for effective change, examining therapist responsiveness (Norcross & Lambert, 2019b) is crucial. Longitudinal models and multilevel data should be used to analyze the complex interactions of these variables over time and their relationship with both post-session and final outcomes (see Gelo & Manzo, 2015; Schiepek et al., 2020). Additionally, differentiating between therapeutic and non-therapeutic safety-related interpersonal dynamics, specifically for different cultures, mental disorders, and relevant safety-related (inter)personal attributes (e.g., attachment styles), will enhance our understanding of how safety can benefit individuals within therapeutic relationships.