Psychological skills training: issues and controversies.
Subject: Psychotherapy (Training)
Authors: O'Donohue, William
Ferguson, Kyle C.
Pasquale, Michele
Pub Date: 06/22/2003
Publication: Name: The Behavior Analyst Today Publisher: Behavior Analyst Online Audience: Academic Format: Magazine/Journal Subject: Psychology and mental health Copyright: COPYRIGHT 2003 Behavior Analyst Online ISSN: 1539-4352
Issue: Date: Summer, 2003 Source Volume: 4 Source Issue: 3
Topic: Event Code: 280 Personnel administration
Organization: Organization: University of Nevada
Accession Number: 170020742
Full Text: Implicit in the notion of skills training is the concept that a particular skill, or skillful behavior, may be instructed, acquired, and displayed in important situations. This emphasis on the delineation and acquisition of overt, effective (skillful) behaviors, clearly puts psychoeducational skills training within the behavioral model of therapy. This compatibility of behavior therapy and skills training can be easily understood when the theoretical bases of behavior therapy are made explicit. Masters, Burish, Hollon, and Rimm (1987) present the following eight primary postulates, or assumptions, of behavior therapy:

1. As compared to traditional psychotherapy, the focus of behavior therapy is on behavior itself rather than a presumed underlying cause of that behavior.

2. Any behavior may be learned, and as such, both adaptive and maladaptive behaviors are acquired through learning. Behavior therapy specifies that the mechanisms of this behavior acquisition are the established principles of leaning theory, such as classical conditioning, operant conditioning, and modeling.

3. Psychological principles can be extremely effective in modifying maladaptive behaviors.

4. Behavior therapy sets specific, overt, well-defined treatment goals. Instead of targeting a global problem such as general unhappiness, a behavioral therapist works with the patient/client to target the specific problems that are interfering with the client's functioning.

5. Behavior therapy rejects classical trait theory, which posits that a person possesses specific traits that predispose him or her to behave similarly in any situation. Rather, behavior therapy focuses on behavior-behavior relations and the mutability of a behavior relative to its antecedents and consequences.

6. Behavior therapists adapt their treatment in response to the client's specific problems, creating and modifying treatment plans in response to their effectiveness, always guided by the principles of learning.

7. Behavior therapy concentrates on the present, focusing on a client's current circumstances and problems more than on "formative" experiences from childhood.

8. Behavior therapists look to empirical support to judge the effectiveness of treatment.

Treatment success is measured by actual, significant improvement in the client's presenting problems and maladaptive behaviors. Skills training is certainly compatible with each of these assumptions. Skills training interventions are concerned with directly altering maladaptive behavior, without attempting to discern any underlying cause of such behavior. For instance, while a client experiencing trouble asserting him- or herself at work might be encouraged to discuss the thoughts that interfere with effective action, it would be in the service of altering such maladaptive thoughts rather than exploring early experiences responsible for these thoughts. Skills training rests on the fundamental assumption that a new, more effective behavior, or skill, can be learned, and typically utilizes methods such as modeling, rehearsal, and operant conditioning to instruct skillful behavior. Assessment of skill deficits is a necessary precursor to remediation of skills, and behavior therapists often utilize tools such as functional assessment and chain analysis to determine the specific problems a client is encountering.

The Skills Training Rationale

O'Donohue & Krasner (1995) have suggested that the skills training approach is based on the notion that individuals in pursuing their lives are confronted with a wide variety of problems and tasks (e.g., communicating with others, resolving conflict, problem solving, relaxing). They need responses in their repertoires to effectively and efficiently achieve their goals in these situations. In general the skills training approach to psychotherapy is based on the following propositions:

1. Situations and problems arise regularly, in order to achieve some end (e.g., solve the problem, realize some personal goal) an individual must be able to respond in a competent, skilled manner.

2. Situations and problems create diverse demands (e.g., need to communicate, need to solve some problem, need to relax, need to interact successfully with others).

3. These diverse demands require diverse skills and capacities (e.g., communication skills, social skills, relaxation skills, problem-solving skills) for their resolution. Life has its "hidden curriculum" (Chan & Rueda, 1979).

4. Individuals vary in their abilities to execute various skills. All individuals have a range of potential abilities, although, as a result of certain conditions (e.g., genetic, physiological, or environmental/ learning problems), individuals can have restricted potentials or restricted levels of achievement within a given potentiality.

5. Some individuals, at certain times and in certain situations, are deficient in skills necessary to meet some demand or achieve some end. The qualifiers in the previous sentence are there to indicate the situational specificity of performance deficits.

6. When situational demands arise that exceed the individual's skills, states of affairs may arise that may be variously described as lack of success, frustration, or even depression, psychophysiological illness, and the like. The manner in which these consequences are described has important implications concerning what appears to be a reasonable way to improve these states of affairs.

7. These individuals can often profit from an educational (psychoeducational) experience in which skill and performance deficits are directly addressed and remediated. Moreover, as McFall (1982) has stated,

Psychological skills training also can be clarified in part by a more precise definition of a "skill." The Oxford English Dictionary (1982) provides a relevant definition of skill: "to have discrimination or knowledge, esp. in a specified manner" (p. 2847). Competent performance in some skill may require declarative knowledge and/or procedural knowledge. Declarative knowledge concerns knowing that certain relevant proportions are true--for example, "Occasional eye contact with the speaker will increase the probability that the speaker will know that I am listening." Procedural knowledge is knowing how to do something--knowing how to relax striated muscles is one such example. Egan and Cowan (1979, p. 8) define skills in a more molar and functional manner: "the competencies that are necessary for effective living." McFall (1982) defines "skills" as "the specific abilities required to perform competently at a task (pp. 1213).

Goldstein (1982) has provided a succinct definition of psychological skills training:


1. Psychological skills training relies on the mechanism of "learning" that is relatively well researched, clear, and understood instead of on less well researched, clear, and understood mechanisms such as cathartic insight or authentic living.

2. Psychological skills training relies on the notion of continua of skill abilities and competencies, as well as of situational demands determining what abilities are necessary to produce what ends.

3. Psychological skills training potentially decreases power differentials (and thereby potential abuses) between trainer and client in several ways.

4. Psychological skills training directly implies a course of remediation, unlike problems encountered when conceptualizing problems along the lines of the diagnoses found in the Diagnostic and Statistical Manual of Mental Disorders--IV (American Psychiatric Association, 1994), which have notorious problems in predicting what treatment will be recommended.

5. A psychological skills training model provides clear, testable hypotheses concerning the origin of psychological problems (e.g., deficiencies in exposure to skilled models at certain developmental periods).

6. Skill training may be an important method for the prevention of problems.

7. Problems are seen as arising from a discrepancy between an individual's capabilities and environmental demands.

8. Psychological skills training avoids what may be called the "irrationalist's inconsistency" of certain models in which therapists are rational scientist-practitioners who weigh evidence in coming to their conclusions, but whose conclusions are that human behavior is controlled by irrational (e.g., unconscious sex and aggressive) forces.

9. Psychological skills training in providing clearly specified and focused topics and training goals is more amenable to scientific evaluation than more artful forms of psychotherapy that rely on less replicable and more idiosyncratic modes of therapeutic interaction.

10. Though consistent with a deficit model of intervention, psychological skills training also is consistent with a personal growth model in which individuals who are performing relatively competently strive for further improvement in various skill areas.

11. Goldstein, Gershaw, and Sprafkin (1985) have argued that traditional psychotherapy, with its emphasis on verbal abilities, insight, and middle-class values, is often inappropriate with lower SES clients, and that psychological skills training has advantages (e.g. it is shorter term, more concrete, and more directive) that are particularly useful for this population.

12. To the extent that similar component skills are necessary for topographically dissimilar tasks (e.g. pain management is necessary for both maintaining an exercise regimen and coping with chronic headaches), an independent measure of this component still should allow more accurate prediction of behavior across similar (e.g. different episodes of headache coping) as well as dissimilar tasks (e.g., headache coping, maintaining a jogging regimen).

13. Psychological skills training, to the extent that it teaches general skills such as problem solving, may enable the client to be in a better position to solve diverse problems and not only the problem that may have precipitated professional contact.

14. Larson (1984) has suggested that "the replicability, accessibility, portability, brevity, and efficiency of skills training approaches make them ideal vehicles for extending training in helping skills beyond the circumscribed traditional population of mental health workers" (p. 9).

How to Develop a Skills Curriculum

Social skills programs have proliferated within the last several decades, many of which are based on commonsense notions of what behavior ought to be relevant in a particular social context. For example, based on commonsense, a program developed for school-aged children might involve protocols for sharing, turn taking, and dealing with conflict.

These targets seem reasonable as these behaviors would undoubtedly occur at least several times each day, for most children. However, if one were to ask children or their teachers or parents directly, what behaviors are most relevant in a given social exchange, a different set of targets might emerge. The above protocols may be deemed irrelevant. This is especially the case when, say, comparing inner city schools with private schools, Christian with secular, and so on. With nonempirically derived programs, therefore, the behavior necessary for successful social performance may or may not be targeted for intervention. One would never know for certain without testing these hypotheses. By contrast, empirically derived programs attempt to target the behavior deemed appropriate by those interested parties who are most affected by discrepancies in performance. For example, in developing an appropriate skill set for on-the-job behavior, naturally the program designer would turn to the individual's employer or fellow employees for their input.

Some of the most comprehensive empirically derived social skills programs are those developed by McFall and colleagues (e.g., Freedman et al., 1978; Gaffney & McFall, 1981; Goldsmith & McFall, 1975). The development of these typically involved four distinct phases: Identifying patient-relevant problem situations; analyzing effective responses for those situations; deriving principles governing effective responses, and developing explicit scoring criteria for such behavior (as a means of evaluating performance).

1) Identify patient-relevant problem situations

This phase would entail meeting with the targeted population and asking members via structured interviews to describe common interpersonal problems. These transcripts would then be given to a similar group and their task would be to rate these items, regarding the extent to which these identified problems are relevant to their own lives, cause marked distress, and the like. In many cases, the program designer would employ the use of a Likert scale in rating these items. A Likert scale might range from "not important" at all or "not relevant," to "extremely important" to "highly relevant," respectively. At the discretion of the treatment developer(s), for sake of parsimony, certain items could then be discarded if there is significant overlap or they are otherwise deemed trivial.

2) Analyze effective responses for the situations

These items would then be given to a cross-section of competent performers (Gaffney & McFall, 1981). Specifically, there would be equal representation along the continuum ranging from upper, middle, and lower performers. As a case in point, the target might involve "fitting in". In sampling effective versus ineffective repertoires, the program developer would first talk to children and their teachers asking them to identify the most popular individuals, the least popular, and those who fall somewhere in between both extremes. Once identified, this sample, comprised of individuals who differ in popularity, would then be given the transcripts of the situations identified in phase one and asked to generate responses with respect to what they would do in those situations.

Responses would then be given to individuals in a position to rate the relative merits of these (e.g., which responses were effective, ineffective, or neither). Ideally, judges would entail those individuals in a position to observe numerous interactions between children as they interact. Accordingly, judges might include teachers, the principal, and teachers' aides, as these individuals have ample opportunities to observe peer interactions.

3) Derive principles governing effective responses

As judges rate responses, they are also asked to provide rationales behind their ratings (Goldsmith & McFall, 1975). Namely, why was one response rated as effective and another ineffective? Guiding principles would then be extracted from these rationales. One guiding principle might read as follows: Children who respond aggressively when their needs are not met tend to be avoided by other children. Specifically, such individuals tend to be excluded from group activities taking place on the playground.

4) Develop explicit scoring criteria for target behavior

Scoring criteria for target behavior is derived statistically. For example, in Gaffney and McFall's (1981) study, judges' ratings were analyzed using coefficient alphas. Those coefficient alphas below a certain level were discarded (in their study .70). The remaining ratings were used in a rater's manual. Thus, for each scenario (identified in the first phase of development), a 5point answer suggested the best response in the situation; 1-point answer suggested the worst response. Cohen's (1960) kappa coefficients, among other rater reliability indices can also be employed (interested readers should consult the following reviews: Cordes (1994) and Kelly (1977).

How to Best Effectively Teach, Especially to Solve Generalization Problems

Once skills are acquired in the therapeutic setting, the question of generalization becomes relevant. A client learning new skills such that he or she can act more skillfully during session, in the presence of the therapist, is certainly important, but for skills training to be truly effective, these skills must extend to all aspects of the client's life. The process by which this occurs is generalization, and altering skills training so that the client can use newly acquired skills in multiple conditions is known as programmed generalization.

Behaviorists speaking of generalization are most often referring to stimulus generalization, which is defined as an "increase in responding to a novel stimulus as a result of training with a different stimulus" (Michael, 1993, p. 83). For example, stimulus generalization is required for a client who has learned relaxation skills in the presence of the therapist (training with an initial stimulus) to then perform these skills at a stressful office meeting (learned response performed under a novel stimulus). To achieve generalization, the therapist must actively work with the client to promote such change, always guided by principles of behavior therapy. For a response to generate to a novel situation, the therapist must first ensure that the skill has been acquired at strength, as a skill that is newly acquired is less likely to generalize to novel stimuli as will a response that has been acquired, practiced, and become part of the client's skill repertoire. The behavior therapist must then actively program for generalization to the new stimulus by using techniques such as modeling and rehearsal, guiding the client through role playing or visualization to encourage new repertoires to be utilized in different situations. Finally, the therapist must reinforce attempts by the client to use newly acquired skills in other environments, and conduct functional assessments of these attempts to determine both the effectiveness of the skill implementation and ways the client may become even more successful in their use.

While stimulus generalization is certainly the most common, and arguably most important type of generalization, it is still important for the behavior therapist to bear in mind two other types of generalization when implementing skills training--response generalization and temporal generalization. Where stimulus generalization is the generalization of a skill to a novel stimulus, response generalization is generalization of a response to another response, often acquired concurrently. An example is a client who is directly instructed in the skill of muscle relaxation, and incidentally learns the skill of deep breathing at the same time. Of final importance is temporal generalization, which can also be viewed as the maintenance of a skill over time. Ideally, when a client successfully uses new skills in his or her life, the effect will be to produce positive changes that will in themselves reinforce the use of that skill, allowing the environment to maintain the skill and promote temporal generalization. However, the therapist must be aware of the actual effects of skill implementation in the client's life for this to be ensured, and treatment to be considered successful.

Rule vs. Contingency Analysis

Social skills training assumes that the components of a given skill set can be described as a series of graduated steps. As such, many social behaviors lend themselves well to instructional design. We can easily teach a person what fork to use with salad, what to say after a person sneezes, and how to politely queue up behind someone standing in line. However, other types of social behavior are elusive, if not inexplicable. Take dating as a case in point. While there are generalities to successful dating, mostly in the form of admonitions (e.g., do not flirt with your food server in front of your date), other more subtle behavior are not as easily taught or perhaps cannot be directly trained. For example, how can an individual be taught how to discern when his or her date is truly interested or simply trying not to hurt his or her feelings? Under what circumstances is sustained eye contact welcomed and when does it become "creepy"? How many unreturned phone calls does it take before a person stops calling? The long and short of answering these is that it depends on the individual and given situation, with respect to both parties.

The skill set concerning this highly complex discriminative repertoire is by and large contingency-shaped versus rule-governed. That is, there are no hard and fast rules that can be invoked when say, it comes to knowing whether or not the person's date is interested or just being nice. Rather, the individual has to contact the naturally occurring contingencies and hopefully, the environment will select effective behavior. While these contingencies cannot be produced in the therapist's office, the therapist can increase the likelihood that this discriminative repertoire will be selected by natural conditions by teaching the client to become more sensitive to social cues, particularly in response to body posture, facial expressions, proximity, and the like. For example, the therapist and client might role-play, whereby the client identifies the emotion conveyed nonverbally, and the therapist provides corrective feedback.

From a theoretical standpoint (1), contingency-shaped behavior differs fundamentally from what is called rule-governed behavior, in that the former entails behavior shaped by way of direct contact with contingencies and the latter is shaped in accordance with verbal stimuli (Skinner, 1989). Verbal stimuli in the form of rules serve a discriminative stimulus function, that is, they affect the differential probability of a given response. Simply, rule-following behavior that often eventuates in positive reinforcement and avoids aversive consequences is likely to be evoked by those verbal antecedents that participate in the contingency. In particular verbal antecedents or rules are said to specify contingencies of reinforcement (Baum, 1994; Pierce & Epling, 1995). Accurate rules, for example, specify what behavior is necessary for effective action, when the response(s) should be emitted or withheld, and when to anticipate the consequence (as many consequences are deferred).

Because certain types of social skills cannot be directly taught, therapists teach clients how to extract rules for themselves from the situation, so-called "metarules" (Poppen, 1989, p. 335). Similar to teaching clients how to discriminate among nonverbal cues, clients also be taught to generate rules of their own. Again, the therapist might role-play with the client; after which, the client specifies the prevailing contingencies and, in so doing, the therapist provides corrective feedback.

Controversies in Skills Training

We will briefly mention five issues that are unsettled regarding skills training. These are:

1) What are the relative strengths of two different skills training strategies: a) the compensatory strategy of teaching the client how to generalize skill sets in which they already display significant strengths to domains in which they demonstrate deficiencies; b) the corrective strategy, i.e., directly teaching skills in these deficit areas.

2) Can it be important, at least in some situations, to understand why the client has failed to learn skills that tend to be learned by others? For example, might a child's poor social skills be secondary to their problematic attention span found in ADHD?

3) To what extent is it important to better understand the process by which skills are learned, maintained, and displayed in order to better improve therapeutic effectiveness, as well as better understand the contingencies related to treatment failures? Currently, we have very few models of these phenomena, although McFall's (1990s) information processing model is an exception to this rule.

4) What are the contextual factors that influence skills acquisition, retention, and display (such as motivation, decreased anxiety, self-efficacy, decreased ambivalence, among other possible candidates) that need to be better identified and understood, particularly intervention strategies to optimize these.

5) How does one design an optimal skills training curriculum for a particular client? Subproblems include: a) How does one prioritize and order when there are multiple skills deficits? b) How do we identify the particular skills needed to remedy the problems that the client is experiencing? At times do our constructs appear more informative than they actually are? Phrases such as social skills and "assertion skills" actually cover a lot of possible ground. c) How do we know that the distress associated with our client will be actually remedied by skill acquisition? d) What are the best skill acquisition methods--how much didactic information vs. how many models, vs. how much actual supervised practice? When are skills best taught in group vs. individual therapy? e) How does one best program generalization? f) How does one structure homework? g) How does one handle therapy interfering behaviors that may arise during the curriculum? h) To what extent are booster session relevant? i) to what extent does one design the curriculum with a prior knowledge of the contingencies that will maintain the class of behaviors in the client's natural environment?


Skills training is both an intervention strategy that seems reasonable in principle (i.e., it seems to be consonant with our experience in life--we learned to ride a bike), and one in which there is an impressive body of literature that attests to its general effectiveness (O'Donohue & Krasner, 1995). However, the actual practice of skills training is not as simple and straightforward as it might appear. In this paper we have outlined some of the unresolved research questions as well as some of the choice points that clinicians face.

Keywords: Skills training, Behavior Therapy

(1) In actuality, contingency-shaped and rule-governed behavior are not easily discernible, as verbally-abled individuals are capable of generating self-rules every moment they contact the contingencies in the natural world.


American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: Author.

Baum, W. M. (1994). Understanding behaviorism. New York: Harper Collins College Publishers.

Chan, K. S., & Rueda, R. (1979). Poverty and culture in education: Separate but equal. Exceptional Children, 45, 422-428.

Cohen, J. (1960). A coefficient of agreement of nominal scales. Educational and Psychological Measurement, 20, 37-46.

Cordes, A. K. (1994). The reliability of observational data: I. Theories and methods for speech language pathology. Journal of Speech and Hearing Research, 37, 264-278.

Egan, G., & Cowan, M. A. (1979). People in systems: A model for development in the human-service professions and education. Monterey, C.A.: Brooks/Cole.

Freedman, B. J., Rosenthal, L., Donahoe, C. P., Jr., Schlundt, D. G., & McFall, R. M. (1978). A social-behavioral analysis of skill deficits in delinquent and nondelinquent adolescent boys. Journal of Consulting and Clinical Psychology, 46, 1448-1462.

Gaffney, L. R., & McFall, R. M. (1981). A comparison of social skills in delinquent and nondelinquent adolescent girls using a behavioral role-playing inventory. Journal of Consulting and Clinical Psychology, 49, 959-967.

Goldsmith, J. B., & Mcfall, R. M. (1975). Development and evaluation of an interpersonal skill-training program for psychiatric inpatients. Journal of Abnormal Psychology, 84, 51-58.

Goldstein, A. P. (1982). Psychological skills training: The structured learning technique. New York: Pergamon Press.

Goldstein, A. P., Gershaw, N. J., & Sprafkin, R. P. (1975). Structured learning therapy--Skill training for schizophrenics. Schizophrenia Bulletin, 14, 83-86.

Kelly, M. B. (1977). A review of the observational data-collection and reliability procedures reported in the Journal of Applied Behavior Analysis. Journal of Applied Behavior Analysis, 10, 97-101.

Larson, D. (1984). Teaching psychological skills: Models for giving psychology away. Monterey, CA: Brooks/Cole.

Masters, J. C., Burish, T. G., Hollon, S. D., & Rimm, D. R. (1987). Behavior therapy: Techniques and empirical findings (3rd ed.). Fort Worth, TX: Harcourt, Brace, Jovanovich.

McFall, R. M. (1982). A review and reformulation of the concept of social skills. Behavioral Assessment, 4, 1-33.

McFall, R. M. (1990). The enhancement of social skills: An information-processing analysis. In W. L. Marshall & D. R. Laws (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 311-330). New York: Plenum Press.

Michael, J. L. (1993). Concepts and principles of behavior analysis. Kalamazoo, MI: Association for Behavior Analysis.

O'Donohue, W., & Krasner, L. (1995). Psychological skills training. In W. O'Donohue & L. Krasner (Eds.), Handbook of psychological skills training: Clinical techniques and applications (pp. 1-19). Boston: Allyn and Bacon.

Oxford University Press. (1982). The compact edition of the Oxford English dictionary. Oxford: Oxford University Press.

Pierce, W. D., & Epling, W. F. (1995). Behavior analysis and learning. Englewood Cliffs, New Jersey: Prentice Hall.

Poppen, R. (1989). Some clinical implications of rule-governed behavior. In S. C. Hayes (Ed.), Rule-governed behavior: Cognitions, contingencies, and instructional control (pp. 325-357). New York: Plenum.

Skinner, B. F. (1989). The behavior of the listener. In S. C. Hayes (Ed.), Rule-governed behavior: Cognition, contingencies, and instructional control (pp. 85-96). New York: Plenum.

William O'Donohue, Kyle C. Ferguson & Michele Pasquale

University of Nevada, Reno

Correspondence should be sent to:

Nicholas Cummings Professor of Organized Healthcare Delivery

Department of Psychology/297

University of Nevada, Reno

Reno, Nevada 89557

Phone: (775) 784-8072



University of Nevada, Reno

Department of Psychology/296

University of Nevada, Reno

Reno, Nevada 89557

Phone: (775) 784-1869


University of Nevada, Reno

Department of Psychology/296

University of Nevada, Reno

Reno, Nevada 89557

Incompetence can be seen as the product of a mismatch between
   a person's performance abilities and the task demands imposed
   on the person. This discrepancy can be described, alternatively,
   as being due either to a deficit in skills, or to excessive
   performance demands. Psychological problems grow out of this
   imbalance between abilities and demands in the person-environment
   system; therefore, the reduction of psychological problems,
   which involves establishing a balance in the system, can be
   achieved either through increasing the person's abilities or
   through decreasing the environmental task demands
   imposed on the person. (p. 22)

The planned, systematic teaching of the specific behaviors needed
   and consciously desired by the individual in order to function
   in an effective and satisfying manner, over an extended period
   of time, in a broad array of positive, negative and neutral
   interpersonal contexts. The specific teaching methods which
   constitute social skills training directly and jointly reflect
   psychology's modern social learning theory and education's
   contemporary pedagogic principles and procedures. (p. 3)
Gale Copyright: Copyright 2003 Gale, Cengage Learning. All rights reserved.