Counselling and psychotherapy researchers have not settled the question of what produces therapeutic change. This disagreement permeates training, funding, regulation, and the everyday decisions practitioners make in the consulting room.

There are two paradigms in tension. One paradigm is often called the medical model. It holds that psychological problems have specific components, that particular techniques target those components, and that outcome depends on delivering the right technique with fidelity. This logic underwrites the movement towards empirically supported treatments and the manuals that carry them. Cognitive behavioural therapy has developed a substantial research program around specified procedures, while also pursuing personalised interventions and the mechanisms through which change occurs (McMain et al., 2015).

The competing paradigm is the contextual model. It holds that the benefits of counselling come mainly from elements shared across approaches, including the therapeutic relationship, the client’s expectation of help, and a believable rationale for the work. Wampold (2015) set out the contextual account and marshalled the meta-analytic case for common factors such as alliance, empathy, and expectation. The fuller argument, with its comparison of contextual and medical assumptions, was developed at length by Wampold and Imel (2015).

These are not merely academic positions. They imply different approaches to counsellor training, service commissioning, and quality evaluation. Before proposing a process-based account, it is worth being precise about where the evidence genuinely supports each side and where it does not. This article is a selective conceptual synthesis. It draws on comparative-treatment evidence, research on alliance and responsiveness, causal-mediation reviews, routine outcome monitoring, and Australian decolonising scholarship because these literatures bear directly on whether relational and technical elements can be understood as components of a feedback-informed therapeutic process. It is not a systematic review of all comparative or mechanism evidence.

Where the Evidence is Genuinely Contested

The Equivalence Finding and its Limits

The empirical anchor for the contextual model is the finding that bona fide therapies produce broadly similar outcomes. Wampold et al. (1997) meta-analysed direct comparisons of treatments intended to be therapeutic and found the distribution of effect sizes consistent with a true difference of zero; under their most liberal assumptions, the upper bound of the true difference was about d = .20. Echoing the Dodo in Lewis Carroll’s Alice’s Adventures in Wonderland, they characterised this result as “all must have prizes” (p. 203). Luborsky et al. (2002) revisited the so-called Dodo bird verdict—the claim that all major forms of psychotherapy are of equal benefit—and concluded that it remained broadly supported, although differences between treatments became smaller after correction for researcher allegiance.

The equivalence finding has been used to support the contextual model, in which common factors are treated as central to therapeutic benefit (Wampold, 2015; Wampold & Imel, 2015). That inference requires care. Equivalence at the level of averaged outcomes is consistent with several explanations, including the possibility that different treatments work through different specific pathways that happen to converge on similar aggregate effects. Equivalence constrains theory. It does not by itself establish the contextual model.

The Unresolved Mechanism Problem

Common-factors literature tends to disregard the unresolved mechanism problem. Cuijpers et al. (2019) examined the mediation evidence for both common and specific factors and applied standard causal criteria, including temporal precedence, dose-response, the ruling out of third variables, experimental support, and a coherent theoretical framework. Their conclusion is uncomfortable for both camps. Their review concluded that neither common nor specific factors had met sufficient criteria to be established as working mechanisms. This finding reframes the debate.

Counselling scholarship provides strong evidence that counselling works but weak evidence as to why. The evidence base for the therapeutic relationship exemplifies this. The therapist–client alliance is one of the most reliable process-outcome associations in the literature, with a modest but robust correlation across hundreds of studies (Flückiger et al., 2018). Yet a correlation between alliance and outcome does not prove that the alliance causes the outcome, since early symptom change could drive both. Bordin (1979) gave the alliance its enduring three-part structure of bond, goals, and tasks; and Rogers (1957) argued much earlier that certain relational conditions were necessary and sufficient for change. These are foundational contributions. They are not, on their own, causal proof.

Manuals, Fidelity, and Responsiveness

The medical model’s preferred vehicle is the treatment manual. Manuals improve replicability and training, and they make research tractable. They also provoke a durable clinical objection. Addis and Krasnow (2000) surveyed practising psychologists and found varied attitudes toward treatment manuals, related to what practitioners understood a manual to contain. Manuals therefore create a practical tension between standardisation and responsive clinical judgement. Even with guidance for adaptation, their use requires judgement about what a particular client needs at a particular time.

At this point, the concept of skilled responsiveness becomes essential. Good practice is not the rigid application of a protocol and it is not the abandonment of technique in favour of warmth. It is the adjustment of what a counsellor does in light of how the client is responding, within a changing interaction that influences both participants (Stiles et al., 1998). The manualisation debate is, at bottom, a debate about how much room the model leaves for that adjustment.

Culture as Frame, not Moderator

Most counselling theory was built on Western, individualist assumptions about the self, autonomy, and what a good outcome looks like. Australian scholarship has been direct about the cost of treating those assumptions as universal. Dudgeon and Walker (2015) discussed how mainstream psychological theory and practice have at times been actively harmful to Aboriginal and Torres Strait Islander peoples, and proposed decolonising strategies to interrupt that harm. Dudgeon et al. (2017) explored the seven interconnected domains of Aboriginal and Torres Strait Islander social and emotional wellbeing: body; mind and emotions; family and kinship; community; culture; Country; and spirituality.

The implication is stronger than the usual call for cultural sensitivity. If wellbeing is defined differently across cultural frames, then the goals of therapy, the meaning of a good alliance, and the criteria for improvement are themselves culturally situated. Culture is not a variable that moderates an otherwise universal account of therapeutic change. It is part of what therapeutic work is aiming at. A model that adds culture as an afterthought misunderstands the problem.

The Neurobiological Turn and its Overreach

Trauma-focused counselling has increasingly reached for neurobiological language to explain distress and justify intervention. Polyvagal theory is one prominent example, offering practitioners an account of safety, threat, and autonomic regulation (Porges, 2022). Its clinical appeal is obvious, since it gives counsellors a physiological vocabulary for what they observe.

However, the scientific standing of that vocabulary is contested. Grossman (2023) presented fundamental challenges to the theory’s core premises and argued that several are likely to be refuted on physiological and evolutionary grounds. The purpose of this discussion is not to adjudicate a debate in psychophysiology, but to identify a recurring evidentiary pattern: the field sometimes adopts mechanistic explanations faster than the evidence supporting them accumulates. A responsible model should distinguish the possible therapeutic value of a compelling rationale from evidence that its proposed physiological mechanism is correct.

Feedback as Contested Practice

If the mechanism question remains open, one might expect the field to rely heavily on measurement to find out what is working, client by client. Routine outcome monitoring does exactly this by tracking progress in each session and alerting clinicians when a case is not on track. Lambert et al. (2018) meta-analysed two established feedback systems and found benefits including reduced deterioration and improved outcomes among clients predicted to have poor outcomes.

Yet adoption of routine outcome monitoring remains patchy and the practice is genuinely contested. Barkham et al. (2024) addressed practitioner and service-lead reservations about routine outcome monitoring, including measurement burden, possible disruption of the therapeutic process, and consistency with therapeutic theory. These rational objections point to a real gap between collecting data and using it effectively.

A Culturally Situated Responsive Calibration Model

Core Proposition

The preceding review motivates a process-based reframing of the common-versus-specific-factors debate in counselling. I call this process responsive calibration and define it as the therapist’s ongoing appraisal and adjustment of relational and technical elements in response to emerging, client-specific feedback, guided by the client’s expressed framework of wellbeing. Feedback includes direct comments from the client; formal outcome or session measures where used; and possible in-session cues such as changes in pace, affect, silence, or engagement. Such cues are provisional information to explore with the client, rather than reliable readouts of an internal state.

The central hypothesis is that the quality of this calibration contributes proximally to change and may explain something beyond the average level of common factors or fidelity to a treatment model. Whether it is a primary mechanism remains an empirical question. Common and specific factors are the materials a counsellor works with. Calibration describes how the counsellor selects, delivers, and reviews the use of these materials with the client.

Responsiveness has a formal theoretical and research history. Stiles et al. (1998) described mutual influence across levels of complexity and time, from treatment selection and overall strategy to momentary interaction and the delivery of an intervention. Their account includes feedback and the context-sensitive use of cultural knowledge. The present model therefore cannot claim these elements as additions absent from Stiles et al.'s account.

For the purposes of this model, attunement means sustained attention to the client’s expressed experience and unfolding responses, with a willingness to have one’s interpretation corrected. Responsiveness concerns how the interaction changes in light of its emerging context. Calibration draws attention to a particular judgement within that interaction: whether the form, intensity, timing, and goal of the current response fit what is being learned about this client. These are overlapping descriptions of practice, not separate mechanisms established by evidence.

The proposed contribution is a focused organisation of this judgement into an episode that can be examined in supervision and research: the client’s stated priority, the feedback available, the counsellor’s interpretation and action, and the subsequent check of fit. Kramer and Stiles (2015) distinguished appropriate responsiveness and reviewed ways of studying it. Naming calibration does not establish a new construct. Its added value would depend on whether this explicit link between feedback, clinical action, and the client’s frame of wellbeing improves explanation or practice beyond existing responsiveness and alliance accounts.

The Calibration Loop

Calibration can be described as a loop with four elements. First, the counsellor forms a working hypothesis about what this client needs, drawing on presenting concerns, relational signals, and any structured measures in use. Second, the counsellor acts, deploying relational and technical elements consistent with that hypothesis. Third, the counsellor explores the response with the client, using direct comments, possible in-session cues, and, where available, routine outcome data (Lambert et al., 2018). Fourth, the counsellor reviews the hypothesis and next action. This may mean changing course, or continuing an approach when the available feedback supports it. The frequency of change is therefore not a measure of calibration quality.

Within this proposed loop, the alliance (Bordin, 1979; Flückiger et al., 2018) provides a medium through which feedback is exchanged and repair becomes possible. Technique (McMain et al., 2015) supplies possible actions. These proposed roles do not exclude other relational or technical pathways to change. Routine outcome monitoring (Barkham et al., 2024; Lambert et al., 2018) supplies information for reviewing the work, whose usefulness depends on how it is understood and used. Expectation and rationale, central to the contextual model (Wampold, 2015; Wampold & Imel, 2015), may shape engagement with each cycle. Whether a neurobiological rationale helps through credibility, physiological change, or another pathway requires testing; the theoretical dispute itself does not decide that question (Grossman, 2023; Porges, 2022).

Situating Calibration in Culture

Calibration requires a target, and the target proposed here is the client’s expressed frame of wellbeing. This emphasis is consistent with broader responsiveness theory, while making the basis for judging fit explicit. A counsellor may alter an intervention skilfully yet pursue a goal that the client does not value. Cultural situatedness therefore concerns whose understanding of a good life guides the work and how that understanding is negotiated.

The social and emotional wellbeing framework makes the importance of this question concrete (Dudgeon et al., 2017; Dudgeon & Walker, 2015). Connections to family, community, culture, Country, and spirituality may be central to how a client defines wellbeing. Those priorities must be explored with the particular client rather than inferred from group membership. An individually focused or a relationally focused goal can fit or misfit, depending on what the client values. Culturally situated calibration requires the counsellor to establish, and keep checking, whose definition of improvement is guiding the loop. This is consistent with evidence-based practice that integrates research evidence, clinical expertise, and client values, characteristics, and culture (American Psychological Association Presidential Task Force on Evidence-Based Practice, 2006; Norcross & Wampold, 2011).

In the context of this calibration process, culture is understood inclusively. It includes the meanings and commitments associated with family, community, language, religion, gender, disability, and social position, as relevant to the person. It also includes the counsellor’s assumptions and the service’s preferred goals, forms of communication, and outcome measures. This broader application should not erase the specific histories of colonisation and collective rights addressed by Australian decolonising scholarship (Dudgeon & Walker, 2015). Individual adjustment cannot substitute for changes to discriminatory or inaccessible services.

In a hypothetical example, a counsellor proposes firmer family boundaries while a client wants relief from conflict without losing valued family connection. Attunement includes noticing the client’s hesitation and asking what the suggestion means to them. Calibration may involve revising the goal and rehearsing a way of expressing needs that preserves connection, then checking whether the next action fits. For another client, greater independence may be the desired outcome. Neither position is the default. The same enquiry applies when a standardised score suggests progress but the client reports no improvement in an important relationship: the discrepancy needs discussion before the score is treated as evidence that the goal has been met.

Testable Propositions

A conceptual model earns its place only if it can be wrong. The following propositions are advanced as falsifiable predictions. They test calibration against established measures and alternative explanations, while recognising that several predictions may also follow from existing responsiveness and contextual accounts.

Proposition One

Calibration quality will explain additional variance in later client outcomes after adjustment for prior symptom change, alliance ratings, treatment model, and treatment fidelity. Quality would be assessed through independently coded episodes linking available feedback to a reasoned decision to revise or continue an intervention, with subsequent outcome withheld from coders. The model predicts that this process carries additional information; simply counting changes would not test that claim. Comparison with established ratings of appropriate responsiveness is necessary to determine whether calibration adds a useful distinction (Kramer & Stiles, 2015).

Proposition Two

Where feedback identifies an unmet client priority or a problem in the therapeutic work, its benefit will be partly mediated by a subsequent response that addresses that information. Measurement without such use is predicted to show less benefit than measurement that informs a clinically justified response. Continued treatment may be appropriate when the feedback supports it. This specifies a possible pathway within the feedback literature (Lambert et al., 2018) and addresses the gap between collecting scores and using them discussed by Barkham et al. (2024). Testing it would require recording feedback, clinical action, and later outcome separately.

Proposition Three

When there is a mismatch between a client’s expressed framework of wellbeing and the goals or responses offered in therapy, calibration that improves this fit will predict better engagement and later client-defined outcomes. The relevant comparison is the degree of fit within a particular therapeutic relationship, not distance from an assumed individualistic norm. Client accounts of fit should be collected separately from subsequent outcome, and the analysis should allow for both individual and relational priorities without assigning either by cultural group. This operationalises the concern, emphasised in Australian decolonising scholarship, that care should not impose a presumed framework of wellbeing on the client (Dudgeon et al., 2017; Dudgeon & Walker, 2015).

Proposition Four

Where counsellors add neurobiological rationales, any additional outcome benefit is predicted to be explained in part by changes in client expectation and the credibility of the rationale. This is a contextual prediction (Wampold, 2015), rather than evidence against all physiological pathways. A discriminating test would compare rationales while holding therapeutic procedures as constant as possible, assess expectation and credibility before later outcome, and measure any claimed physiological mediator independently. The disagreement between Porges (2022) and Grossman (2023), outlined above, motivates testing these explanations rather than deciding between them on the basis of a persuasive rationale alone.

Implications for Practice and Training

If the proposed calibration process contributes to change, training should develop counsellors’ ability to enact it in particular therapeutic relationships. The calibration view brings together relationship skills and technique delivery through deliberate practice in reading feedback and revising action, including the more difficult skill of noticing when one’s own hypothesis is wrong. This approach reframes the manuals debate (Addis & Krasnow, 2000) as a question of how much structured guidance supports calibration without freezing it.

For services, the model offers a more defensible rationale for routine outcome monitoring than compliance or benchmarking. Measurement has a clinical purpose when it informs a reasoned decision about the work, which gives service leads a principled response to the reservations their staff raise (Barkham et al., 2024). For supervision, calibration provides a shared language that neither privileges one school nor collapses into eclecticism.

Limitations

First, this is a conceptual article. The model is a proposal, and the propositions above are untested predictions rather than results. Second, calibration is difficult to measure, and the model’s value depends on developing indices of feedback-contingent adjustment that are more than repackaged process ratings. Third, the mediation problem identified by Cuijpers et al. (2019) applies to this model as much as to any other. Demonstrating that calibration correlates with outcome would not establish that it causes outcome, and the model will require experimental and dismantling designs, not only naturalistic association, before it can claim mechanistic status. A further limit is conceptual overlap. If calibration cannot be distinguished reliably from appropriate responsiveness, or adds no explanatory or practical value, a separate model would not be justified (Kramer & Stiles, 2015).

Conclusion

The long argument between common and specific factors in counselling effectiveness remains unresolved. Counselling and psychotherapy are associated with beneficial outcomes, yet the field cannot explain the mechanisms of change with the confidence that either philosophical camp asserts (Cuijpers et al., 2019; Wampold & Imel, 2015). In this article, I have argued that framing the debate around competing classes of factors can obscure how relational and technical elements are used within an unfolding therapeutic process. The proposed alternative is to investigate the contribution of a process: the counsellor’s culturally situated, feedback-informed calibration of relational and technical elements to a particular client. That reframing does not discard the existing evidence; it organises the evidence around a proposed unit of clinical judgement. Whether that organisation adds to established responsiveness theory and helps explain change remains to be tested.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflicts of interest

The author declares no conflicts of interest.

Ethics approval

Not applicable. This is a conceptual article and did not involve human participants or data.