Behavioural marital therapy (BMT) has developed from the application of social exchange and social learning theories to the understanding and modification of marital interaction. Encouraging high rates of exchange of positive behaviours, and training in communication skills, have been the major focuses of BMT till recently. While these strategies produce therapeutic change, there are still many couples whose marriages do not improve with this sort of therapy. Recent developments have focused on the role of cognitive, affective and situational variables in marital interaction. Outcome research has shown the value of cognitive and affective interventions in their own right, but it is unclear if the addition of these interventions will enhance the efficacy of BMT.

Our aim in this paper is to provide researchers and practitioners with a brief overview of current approaches to the assessment and treatment of dysfunctional communication in marriage. In our discussion of assessment, we examine four basic types of procedures: questionnaires, self-monitoring techniques, interviews and techniques involving the direct observation of interaction. We conclude that both insider (spouses' perceptions and interpretations gathered through self-report and self-monitoring) and outsider (direct observation of interaction) provide different but equally valid information about the relationship. Therapists and researchers are encouraged to use both types of information. Our discussion of therapeutic techniques includes the three major components of Behavioural Marital Therapy (BMT): Behaviour Exchange, Communication Training and Problem-Solving Training. We also focus on the importance of considering spouses' cognitions and emotional reactions, particularly those related to sex roles, when designing therapeutic approaches. Issues related to assessing the efficacy of therapy outcome are also discussed.

Cognitive models of marital distress have received increasing attention. There are four major goals of this paper: (1) to review the prominent cognitive theories of marital distress and corresponding empirical data; (2) to provide theoretical integration; (3) to discuss the adjunctive use of cognitive interventions with behavioural marital therapy and outcome studies assessing the efficacy of cognitive clinical applications with couples; (4) to suggest future clinical and research directions.

This paper discusses current issues in the literature on marital violence. Issues in three areas are discussed from a research and clinical perspective: conceptions of marital abuse, assessment of marital aggression and issues in treatment. Although there are specific cognitive-behavioural treatments reported in the literature, a position of caution is advocated here. Before intervening there are issues that the therapist should consider carefully. (1) Assessment must first answer whether conjoint marital therapy is appropriate at all. (2) Therapists must define their own position and estimate availability of supporting resources (e.g., legal, medical) when working with abuse. (3) Although all couples should be screened routinely for the nature and frequency of their aggressive patterns, neither the sensitivity nor specificity of currently available screens is yet established. Examples of current approaches to assessing spouse abuse are presented.

There is a well established relationship between marital discord and individual psychopathology in family members. This paper reviews evidence relating to patterns of marital interaction related to the co-occurrence of marital discord and three common psychological disorders: depression, alcohol abuse, and childhood conduct disorders. The effects of behavioural marital therapy (BMT) on dysfunctional marital interaction in such cases, and the impact of BMT on individuals' psychological disorders, are evaluated. It is concluded that BMT is often a useful component of treatment in each disorder reviewed. Further it is argued that it is important routinely to assess the relationship context in which these disorders occur. Finally, some of the difficulties and limitations of the application of BMT in cases where the presenting problem is an individual psychopathology are considered.

Four married couples were recruited from an outpatient alcohol counselling centre. Four husbands and one wife had recently received therapy for alcoholism and all but one was abstinent during the study. All scored as distressed on the Marital Adjustment Test. They participated in ten sessions of behavioural marital therapy emphasising positive reinforcement, communication, expression of feelings and problem solving. Problem solving discussions were audio-recorded at each session and analysed using a behaviour code. Measures of marital satisfaction, conflict and depression showed modest gains which were generally maintained at follow-up. Communication skills improved most with training, problem solving less so and feelings expression declined below baseline levels. These changes did not appear to be very durable. The non-abstinent client reduced his drinking to a level satisfactory to his wife during therapy. These findings support other recent work, draw attention to the need to find ways of promoting the maintenance and transfer of therapeutic changes and suggest that behavioural marital therapy may usefully be employed in outpatient alcohol counselling settings.

The strength of behavioural marital therapy is seen to lie in its close links between theory, assessment and therapy. Issues raised by contributions to this special issue are reviewed as they bear on each of these domains and the implications they have for the next decade of research and practice are discussed. Some problems with the prevailing cognitive-behavioural models of marital functioning are noted, the difficulties of establishing effective clinical use of research innovations in assessment are lamented, and the need for innovation in therapy to meet the demands of diverse client groups is considered. An appeal is made for more attention to be given to maintenance and transfer (generalisation) issues. Finally, it is suggested that the decade of the 1990s become the decade of prevention research.