In the first two issues of this journal various articles opened a debate revolving more or less around the question: “what type of research for clinical psychology?” Instead, this issue, significantly, is entirely devoted to research into psychotherapy, with the aim of stimulating reflection on the conceptual and methodological issues involved in the need to “do research” into a clinical practice such as psychotherapy.
First of all, we naturally need to state that the term “psychotherapy” is not at all clear-cut, since the types of intervention it includes are quite different from each other (cfr. Roth and Fonagy, 1996), though all refer to a more general form of “contract” between one or more “clients” and one or more “professionals in the psychological health field”, who agree to work together for the purpose of improving the psychological well-being of the clients using the technical rules that the specialists choose to adopt.
Many clinicians actually prefer to underline that the practice of psychotherapy involves a “craft-like” rather than a strictly “technical” competence, and almost all agree on the need to adapt the technical principles from case to case to suit the patient’s specific characteristics,  both as regards his/her form of suffering (ie the psychopathological diagnosis: see for example, Horwitz et al., 1996), and as regards his/her peculiar ways of relating to the world (ie his/her personality traits, which also determine the way of constructing a therapeutic alliance with a specific therapist: cfr. Safran and Muran, 2000), and as regards the conditions established in the specific context of that psychotherapeutic relationship (in other words, the setting, in the broad sense of the term, including the formal conditions of the contract, but also the mental state of therapist and patient: cfr. Nissim Momigliano, 2001).
Obviously, the principles and practical rules that make up the theory of technique in the psychotherapy intervention are inspired by certain theoretical hypotheses related to what is "healthy" or "ill", or, in other words, to what is “dysfunctional” in a specific context, in that it causes individual suffering and/or psychosocial maladjustment. Such a definition, taken literally, would not seem to belong to clinical psychology, if the latter – as Pazzagli and Benvenuti wrote in the second issue of this journal – “takes man in a general sense as its object, not looking for signs or symptoms of illness. Instead, it tries to understand certain aspects of the person, of his suffering, his malaise, and also of his disorders, independently of whether the person is ill or not,  in an in-depth study of individual cases, examining aspects of the personality, conflicts, anxieties and defence mechanisms” (Pazzagli & Benvenuti, 2006, p.1).
In actual fact, in the folk psychology, that is, in the way each person thinks of his/her own mental and affective functioning, and above all of his own interpersonal behaviour, experience is inevitably classed as dysfunctional versus functional, translating the antonymy in terms of subjective malaise/well-being; and, if  psychotherapy is a psychological treatment, ie an intervention based on psychological knowledge, it is equally inevitable for psychology (and in the case in point, clinical psychology) to define its goals by using categories that identify what is experienced as being, at least subjectively, problematic and “dysfunctional”. Moreover, if psychology as a “scientific” discipline wants to go beyond the mere phenomenological description of lived experience, it will have to establish – liaising also with the basic research – which hypotheses about psychological dysfunctioning are more explicative or more grounded than others and which enable the construction of more effective intervention techniques. Furthermore, if what it wants to offer with psychotherapy is a professional intervention supported by scientific knowledge, it must be possible to develop empirically-based tools to assess its efficacy, both in absolute terms and compared to its different forms and technical choices.
Psychotherapy research, which has a history of more than half a century, in our opinion has, among other things, produced an acceleration in a historically interesting phenomenon: on the basis of research data, in the past 30-40 years we have witnessed a process of reciprocal fertilization between the various theoretical models that emerged from the dynamic and the clinical over the last century, so that today we find fewer differences, for instance, between dynamic psychotherapy in the strict sense (that is, therapies in the psychoanalytic area), cognitive psychotherapy and systemic-relational psychotherapy, both as regards psychopathological explanations and hypotheses on therapeutic change and on the factors responsible. Therefore, while the goals that psychotherapy research set itself in the initial stages were identified above all in the validation of specific techniques for the treatment of specific disorders (What works for whom? was the title in the ‘90s of the wide-ranging overview by Roth and Fonagy, translated in Italian as Psicoterapie e prove di efficacia), the results obtained thanks to advances in research in this sector have helped to orient clinicians towards the choice of integrated models of intervention and to push researchers to focus on shared factors in therapeutic change.
For instance, in the area of dynamic psychotherapy – which is more specifically the field of our research  – we now find concepts, common to the various psychoanalytical orientations in the broad sense, which also have clear links with the other models of psychotherapeutic intervention. In particular, the formulations of evolutive psychopathology tend to converge on a view of the mind as “relational”, in the sense that psychopathology, like normal functioning, is explained as being the result of a construction, starting from the major affective experiences of childhood, of “operative models” or cognitive-affective-behavioural patterns acting at the unconscious level as motivational programmes, which orient the behaviour of the adult person according to expectations, motivations, behavioural reactions etc, derived from the interpersonal relations of the past.  These models however continually undergo revisions and reorganizations in the growth of a whole lifetime, based on new experiences of affectively significant relations: the psychotherapeutic relationship offers, according to this hypothesis, a new and unique opportunity to actively change – thanks to the joint effort of patient and therapist – the established models which have proven dysfunctional because they tend to apply the rules of functioning in a rigid way, insensitive to the changing nature of interpersonal contexts (cfr. Sandler and Sandler, 1993; Fonagy and Target, 2003).
The concept of “structural” change, too, traditionally defined by psychoanalysis as the acquisition of capacity for integration and synthesis of the different psychic demands, is translated today in terms that follow the lines, at least in part, of cognitivist studies on the development of “metacognition”, connected in turn to the formation of a theory of the mind in the age of development. It is thought for instance that psychotherapy should facilitate and promote the development of mentalization – in the sense of the capacity to control the affects using cognitive competences of reflection on the mental states of oneself and of others (Fonagy et al., 2002). Or it is highlighted that changes in the language used by the patient  to tell his stories and talk about his emotions are an indicator that is sensitive to the  development of a function of connection between emotion and thought, called “referential activiity” and considered to be a significant goal in the psychotherapy process (Bucci, 1997; 2005).
Also the hypotheses concerning the factors responsible for change – despite the difficulty of responding empirically to such a complex question - focus more and more on the quality of the therapeutic relationship: more precisely, research into the microprocesses that seem to work together to achieve a satisfactory outcome reveals that the quality and efficacy of interventions, in whatever technical form they may take (eg as clarifications or as interpretations or as communications of mirroring, etc), is to be assessed both in relation to the therapist’s capacity to capture the thematic focus and to further discuss it  (Colli & Gazzillo, 2006), and to the function of “repairing” the therapeutic alliance, under constant negotiation between  patient and therapist during the process(Safran & Muran, 2000). In a sense, the studies underlining the importance of an activity of reciprocal resolution of the continual breakdowns that take place in therapeutic conversations compared to an ideal optimal harmonization implicitly suggest taking into consideration the vital interconnection between therapeutic factors that have long been seen as fundamental by psychoanalytic technique: the expanding of awareness and the emotional experience of the therapeutic relationship. Or rather, from the patient’s point of view, the experience of an empathic relationship is not enough if this empathy is not actively demonstrated by the therapist through interventions specifically aimed at maintaining and restoring the mental and affective harmony between the two participants.
To introduce this issue on empirical research in psychotherapy, we would now like to make a brief survey of some fundamental methodological questions. More in general, in psychology, empirical research is understood as the systematic study of phenomena using an adequately standardized methodology; this study aims to increase the scientific knowledge of the phenomena observed, by means of the identification of specific relations between dependent and independent variables that  make it possible to empirically confirm or confute the validity of the initial hypothesis. Methodologists distinguish different levels of empirical research, defined respectively as descriptive, correlational and experimental enquiry: descriptive enquiry aims to offer a picture, that is as accurate and complete as possible, of what happens in the situation under observation; correlational enquiry aims to describe whether and how what is observed is related to other factors and dimensions; lastly, experimental enquiry aims to put forward the existence of causal links between the phenomena observed and, to do so, it must operate by using a rigorously controlled methodology that envisages the manipulation of the independent variable (Ercolani et al., 1990).
It is obvious that applying a third level experimental methodology to research into psychotherapy risks obtaining misleading results, since it causes “unnatural” distortions in the very phenomenon it is trying to examine: for a more thorough treatment of this problem, see the paper presented by Westen in Rome in June 2006, criticising psychotherapies based on empirical evidence, summed up in this issue by Gazzillo and Lingiardi. On the other hand, the problem of the effects of data collection arises for all the procedures of observing  situations in progress: the observer must use techniques that are as non-invasive as possible, in order to reduce the inevitable impact that the observation itself has on the relationship being observed (eg the mother-child relationship, that of the therapist-patient or teacher-pupil etc.), yet also take into account the fact that the act of research in some way entails an intervention on the phenomena being studied.
On the other hand it cannot be said that for the advancement of knowledge in the area of the clinical intervention it is pointless to attempt to systematize the study of psychotherapy – in the many psychological constructs related to its results and to the therapeutic process – hiding behind the absolute ineffability of the clinical experience. Freud himself, known for his hostility to experimental research, hoped for the possibility of “proving” the value of his theoretical hypotheses by assessing the patient’s improvement: the methodological naϊvety of his suggestion lay, not just in the difficulty of operationalizing the theoretical constructs used, but mainly in the fact that the same clinician involved in the therapeutic relationship would be the only one entrusted with the task of making this assessment. Today most researchers support the wisdom of distinguishing the functions: the clinician does his/her work as therapist and the researcher studies the progress of clinical work (cfr. Gill, 1994).
However, to produce clinically significant results, empirical research into psychotherapy must certainly manage to obtain relatively reliable results from research protocols that respect both the need for a clinical practice heedful of professional ethics, and the “relational” aspects characterising any form of clinical psychology intervention (cfr. Carli, 2006). So for example the comparison between the outcomes of different treatment conditions should be approached by considering the effects of the specific reciprocal interactions between modes of treatment, therapist and patient (Elkin, 1999), with the research picture becoming exponentially more complex.
As Orlinsky (2006) underlines, in the tradition of psychotherapy research, the patterns of enquiry that have actually prevailed are those based on the study of certain abstracted qualities: the “patient” component is therefore studied as an individual removed from his life context; the “mental disorder” component or “problem” to solve is studied in manualistic terms – where the diagnosis categories strictly exclude each other; psychotherapy is defined as a set of “manualized treatment skills” in which the therapist is espected to be trained in order to apply them with competence (independently of his personal characteristics); finally, the results are interpreted as the outcome of an impact of psychotherapeutic techniques on the disorder being treated or, at the most, on the conditions of social life (eg holding down a job,  resorting to medical therapies, etc.) in the period immediately after the end of the treatment.
More in general,  research into psychotherapy that gives meaning to the daily work of clinicians should satisfy numerous requisites, including, first of all,  setting it as the goal of one’s study to find out what happens in psychotherapy as it is practised in natural clinical situations. Orlinsky (2006) identifies the following characteristics of the “reality of psychotherapy” which should be addressed by the studies of psychotherapeutic process and outcome: (a) an intentionally-formed and culturally-defined social relationship through which a potentially healing intersubjective connection is established (b) between persons who interact with one another in the roles of client and therapist (c) for a delimited time during which their life-course trajectories intersect, (d) with the therapist acting on behalf of the community that certified her (e) to engage with the patient in ways that aim to influence the patient’s life-course in directions that should be beneficial for the patient and to treat the patient as an individual and not as a specific “disorder” diagnosed. Secondly, to be considered effective, the intervention – carried out with specific relational strategies by a specific therapist in a specific institutional context – should subsequently promote in the patient a growth towards more satisfying conditions of life through the subject’s sense of personal integrity and social recognition of his self expression. It is necessary, for instance, to use descriptions of the much more specifically “psychological” characteristics of the patient, using highly complex psychodynamic variables and not to merely describe the pathology or the socio-anagraphic variables that are usually indicated in the selection of subjects for research (Gruppo OPD, 2000; Orlinsky, 2006).
In fact, international research into psychotherapy – not only in the psychodynamic area – seems to be moving towards a more thorough study of microprocesses. Emblematic is the double issue of the journal Psychotherapy Research of 2005 (1-2), dedicated to the therapeutic relationship, where most of the studies deal with the therapeutic alliance, evaluated using different measurements and in different clinical contexts, such as cognitive-behavioural psychotherapy (Watson & Geller, 2005), in brief solution-focused psychotherapy (Wettersten, Lichtenberg, & Mallinkrodt, 2005), in dynamic psychotherapy of personality disorders (Lingiardi, Filippucci, & Baiocco, 2005); some contributions also show particular aspects of the dyadic relationship, like the divergence of viewpoints between therapist and patient (Fitzpatrick, Iwakabe, & Stalikas, 2005), the interactive regulation of emotional exchanges, studied through the coding of video-recordings (Benecke, Peham, & Bänninger-Huber, 2005), or the changes in motivation for psychotherapy triggered by a prolonged consultation phase (Rumpold et al., 2005).
In the second issue of 2006, the same journal published six articles discussing the influence of the therapist on the outcome of psychotherapy. The central question is: how to reduce the effect of the “therapist” variable in research on outcomes, in order to obtain studies that are “controlled” enough to document the effects of psychotherapy on the patient? The various authors debate the possibility of constructing models at multiple levels that can enable the numerous factors to be held together, and the debate ends by underlining that attempts to construct data analysis models that reduce the effect of the therapist are pointless, since all the research shows the importance of the therapist for the outcomes of psychotherapy and, as soon as the effects of the therapist are made to disappear, “all the rest will disappear”! (Wampold & Bolt, 2006, p.186).
Let us also cite, from the fourth issue of 2006, two quite original articles: one by Canadian authors that uses a qualitative methodology to explore what were, from the patient’s point of view, the critical events which in the initial phase of treatment promoted confidence and increased the patient’s willingness to cooperate in the psychotherapy (Fitzpatrick, Janzen, Chamodraka, & Park, 2006); and a study from Columbia University on the temporal sequences of interactive events characterising the processes of “revelation” of the patient in the course of psychotherapy (Farber, Berano, & Capobianco, 2006).
As can be glimpsed from this very quick survey, the interest in psychotherapy research has lately been moving towards a reciprocal matching up of clinical activity and research; in other words, researchers’ interest is coming closer to the typical questions that occupy clinicians: what are the prognostic indicators of a fruitful relationship? When and how is it possible to help the patient to really place him/herself in the therapist’s hands and to cooperate in the exploration of his inner world? Which therapist characteristics affected that relationship with that specific patient? To answer these and many other similar questions, recent empirical research into psychotherapy has generally been oriented towards naturalistic investigations and also qualitative methodologies, with a preference for those formalized methods that however require a clinical evaluation on the part of experienced judges, with the goal of making a more thorough study of the dynamic and interactive micro-processes that characterize psychotherapeutic work and influence its results.
Starting from the classic contrast between a holistic “subjective” clinical method and experimental “objectivizing” procedures that compartmentalize reality, empirical research into psychotherapy seems today to be approaching a fertile hybridization of techniques and theoretical assumptions which can set an example in dealing with the methodological problems of psychology research in the complex, elusive area of clinical psychology.
More specifically, in the psychoanalytic field, it is of some interest to look more closely at the critical discussion sparked off by Kernberg’s recent contribution to the International Journal of Psychoanalysis with the significant title “The pressing need to increase research in and on psychoanalysis” (2006a). The article sets out to survey the present state of research, focusing both on strengths and weaknesses. In particular, the following result is highlighted: the increase in the reflective function in a group of borderline patients who have undergone a specific treatment, TFP (Transference-Focused-Psychotherapy), compared to two other groups subjected to other kinds of treatment. There are a series of recommendations to enable a form of research to be developed that is acceptable within the psychoanalytic community. Kernberg obviously believes in the future integration of therapy and research, although he admits that so far research has not been a great help for psychoanalytic practice (“so far it has not helped psychoanalytic practice”, p. 922). Although the critical comments of the French psychoanalyst Roger Perron (2006) constantly repeat that research is crucial for psychoanalysis, they actually reveal a radically different and in many ways incompatible approach. It should be noticed that most of the criticisms can be accepted and concern questions recently handled by psychoanalysts who are closer to research, such as Drew Westen: the problem of the quantification, the repeatability of observations, the falsification of hypotheses, evidence-based methodology that is inadequate for the complexity of psychoanalytic therapy, etc. But it is Perron’s quotes (or non quotes!) that betray his distance, if not his hostility, to the matter: reference to French research (INSERM, 2004) that has been highly criticised and therefore of little significance, reference to texts like that of Prigogine and Stengers (La nouvelle alliance: métamorphose de la science, 1979), now clearly showing their age, reference to the scientific models common in research (according to Perron, those of physics and chemistry).
This is, if we may use the expression, an overtly essayistic-speculative approach, which takes absolutely no account not only of the problems and of the now substantial growth of research into psychotherapy (not to mention research into psychoanalysis: see, besides  Open Door Review edited by Fonagy in 2002 and cited by Kernberg, more recent contributions by De La Parra and Del Rio (2005), Kächele and others (2006), Leuzinger-Bohleben and Fischmann (2006), Westenberger-Breuer (2007), published in International, and lastly the paper by Leuzinger-Bohleber and Target (2002), translated into Italian in 2006 with the title The results of psychoanalysis I risultati della psicoanalisi). Kernberg obviously found it easy to respond to Perron (Kernberg, 2006b), moreover citing some interesting new research, such as the empirical demonstration of the mechanism of projective identification done by Krause and Howard (1999) and an analysis of defence mechanisms emerging from an examination of the language used in the stories of trauma patients.
It is disappointing to see that today, despite the institutional initiatives taken by the International Psychoanalytic Association in the field of psychoanalytical research, there is still what could be called a “conversation falling on deaf ears”. A few steps forward may be made if there are some concrete examples of ways in which research has helped clinical practice. However this will only be possible if we can go beyond the present situation in which it is impossible to compare paradigms.

References

Arbeitsgruppe OPD (2001). Operationalisierte Psychodynamische Diagnostik. Bern: Hans Huber Vlg.

Benecke, C., Peham, D., & Bänninger-Huber, E. (2005). Nonverbal relationship regulation in psychotherapy. Psychotherapy Research, 15, 81-90.

Bucci, W. (1997). Psychoanalysis and cognitive science. New York/London: The Guilford Press.

Bucci, W. (2005). Process research. In E.S. Person, A.M. Cooper & G.O. Gabbard (Eds.), Textbook of psychoanalysis (pp. 317-333). Washington, DC: American Psychiatric Publishing.

Carli, R. (2006). Psicologia clinica: professione e ricerca [Clinical psychology: profession and research]. Rivista di Psicologia Clinica, 1 (1).

Colli, A., & Gazzillo, F. (2006). Due strumenti per la valutazione del processo terapeutico: le Analytic Process Scales e lo Psychotherapy Process Q-set [Two tools for the assessment of the therapeutic process: the Analytic Process Scales and the Psychotherapy Process Q-set]. In N. Dazzi, V. Lingiardi & A. Colli (Eds.), La ricerca in psicoterapia. Modelli e strumenti. Milano: Raffaello Cortina.

Dazzi, N. (2006). Il dibattito contemporaneo sulla ricerca in psicoterapia [The current debate on the research in psychotherapy]. In N. Dazzi, V. Lingiardi & A. Colli (Eds.), La ricerca in psicoterapia. Modelli e strumenti. Milano: Raffaello Cortina.

De La Parra G. & Del Rio M. (2005). Can psychoanalysis and systematic research work productively together? International Journal of Psychoanalysis, 86,1, 151-154.

Elkin, I.E. (1999). A major dilemma in psychotherapy outcome research: Disentangling therapists from therapies. Clinical Psychology: Science and Practice, 6, 10-32.

Ercolani A.P., Areni A., Mannetti L. (1990), La ricerca in psicologia [The research in psychology]. Roma: NIS.

Farber, B.A., Berano, K.C., & Capobianco, J.A. (2006). A temporal model of patient disclosure in psychotherapy. Psychotherapy Research, 16, 463-469.

Fitzpatrick, M.R., Iwakabe, S., & Stalikas, A. (2005). Perspective divergence in the working alliance. Psychotherapy Research, 15, 69-80.

Fitzpatrick, M.R., Janzen, J., Chamodraka, M., & Park, J. (2006). Client critical incidents in the process of early alliance development: a positive emotion-exploration spiral. Psychotherapy Research, 16, 486-498.

Fonagy, P. (Ed.). (2002). An open door review of outcome studies in psychoanalysis. London: International Psychoanalytic Association.

Fonagy, P., & Target, M. (2003). Psychoanalytic theories. Perspectives from developmental psychopathology. London: Whurr.

Fonagy, P., Gergely, G., Jurist, E.L., & Target, M. (2002). Affect regulation, mentalization, and the development of the self. New York: Other Press.

Gill, M.M. (1994). Psychoanalysis in transition. Hillsdale, NJ: The Analytic Press.

Horwitz, L., Gabbard, G.O., Allen, J.G., Frieswyk, S.H., Colson, D.B., Newson, G.E., & Coyne, L. (1996). Borderline personality disorder. Tailoring the psychotherapy to the patient. Washington, DC: American Psychiatric Press.

INSERM (2004). Psychothérapie. Trois approches évaluéees [Psychotherapy. Three approaches evaluated]. Paris: INSERM.

Kächele H., Albani C., Bucheim A., Hölzer M. et al. (2006). The German specimen case, Amalia X: Empirical studies. International Journal of Psychoanalysis, 87,3, 809-826.

Kernberg, O.F. (2006a). The pressing need to increase research in and on psychoanalysis. International Journal of Psychoanalysis, 87, 919-926.

Kernberg, O.F. (2006b). Research anxiety. A response to Roger Perron’s comments. International Journal of Psychoanalysis, 87, 933-937.

Krause, M.S. & Howard K.I. (1999). Between group psychotherapy outcome research and basic science. Journal of Clinical Psychology, 55, 159-169.

Leuzinger-Bohleben, M. & Fischmann T. (2006). What is conceptual research in psychoanalysis? International Journal of Psychoanalysis, 87, 5, 1355-1386.

Leuzinger-Bohleber, M., & Target, M. (2002). Outcomes of psychoanalytic treatment. London/Philadelphia: Whurr.

Lingiardi, V., Filippucci, L., & Baiocco, R. (2005). Therapeutic alliance evaluation in personality disorders psychotherapy. Psychotherapy Research, 15, 45-54.

Nissim Momigliano, L. (2001). L’ascolto rispettoso. Scritti psicoanalitici [The respectful listening. Psychoanalytic papers]. Milano: Raffaello Cortina.

Orlinsky, D. (2006). Comments on the state of psychotherapy research. North American SPR Newsletter, June 2006.

Pazzagli, A., & Benvenuti, P. (2006). Per una psicologia del patologico [For a psychology of the pathological]. Rivista di Psicologia Clinica, 1 (2-3).

Perron, R. (2006). How to do research? Reply to Otto Kernberg. International Journal of Psychoanalysis, 87, 927-932.

Prigogine, I., & Stengers, I. (1979). La nouvelle alliance. Métamorphose de la science [The new alliance. The metamorphosis of science]. Paris: Gallimard.

Roth, A., & Fonagy, P. (1996). What works for whom? A critical review of psychotherapy research. New York: The Guilford Press.

Rumpold, G., Döring, S., Smrekar, U., Schubert, C., Koza, R., Schatz, D.S., Bert-Schüssler, A., Janecke, N., Lampe, A., & Schüssler, G. (2005). Changes in motivation and therapeutic alliance during a pre-therapy diagnostic and motivation-enhancing phase among psychotherapy outpatients. Psychotherapy Research, 15, 117-128.

Safran, J.D., & Muran, J.C. (2000). Negotiating the therapeutic alliance. New York: The Guilford Press.

Sandler J., & Sandler A. M. (1998). Internal Objects Revisited. Monograph Series of Psychoanalysis Unit. London: University College.

Wampold, B.E., & Bolt, D.M. (2006). Therapist effects. Clever ways to make them (and everything else) disappear. Psychotherapy Research, 16, 184-187.

Watson, G.C., & Geller, S.M. (2005). The relation among the relationship conditions, the working alliance and outcome in both process-experimental and cognitive-behavioral psychotherapy. Psychotherapy Research, 15, 25-34.

Westenberger-Breuer H. (2007). The goals of psychoanalytic treatment: Conceptual considerations and follow-up interview evaluation with a former analysand. International Journal of Psychoanalysis, 88, 2, 475-488.

Notes

* Full professor of Dynamic Psychology, “La Sapienza” University of Roma. Top

** Full professor of Dynamic Psychology, “La Sapienza” University of Roma. Top