FACTORS WHICH PREDICT THE EFFECTIVENESS OF DIFFERENT TREATMENTS OF MILD DEPRESION OR DYSPHORIA

either researchers nor clinicians agree on the meaning of the term "depression" (Brewin, 1985), and it is used to refer to a wide variety of deficiencies. As with any other psychological disorder, there is a need for a precise distinction between the different types of depression. The factors on which we can most usefully base this differentiation are the symptoms or reactions associated with depression, their evolution and their tre atment (Feinberg, 1992; Garber, Miller and Seaman, 1979). Etiological factors may also be very useful for the clinical prediction of the onset and, where it occurs, the worsening of depressive reactions.
The principal models that have been constructed to explain depression can be grouped in two blocks accor ding to the role they assign to a series of factors in the development and persistence of the disorder: those that emphasise factors of cognitive vulnerability, and those that focus on the behavioural deficiencies involved in depression. Below, we present the main findings of each approach, with the aim of clarifying the possibility that there exist different types of depression.

Findings of the cognitive models
In the first place are the cognitive models of Ellis and Beck, the authors whose starting point was the clinical observation of depressive symptomatology. The approach of Ellis (1962, 1987; Ellis and Grieger, 1977) has become practically obsolete in research on depressive disorders, possibly because no reliable and valid measu res of its cognitive constructs have been developed (Kendall and Korgeski, 1979; Smith, 1989). Moreover, research on its therapeutic assumptions is virtually non existent. Beck's model (1967; Beck, Rush, Shaw and Emery, 1979), on the other hand, offers the approach that has been most influential, offering as it does a series of theoretical principles of great heuristic value, which continue to inspire numerous studies on cognitive fac tors in depressive disorders. Furthermore, it is on this approach that the most important current form of treating depression is based.
Most research based on the revised theory of learned helplessness (see Buceta and Polaino-Lorente, 1982), and the more recent theory of hopelessness (Abramson, Metalsky and Alloy, 1989; Alloy, Abramson, Metalsky and Hartlage, 1988), indicates that there is an important relationship between negative attributions and depression. The findings from Seligman's (1975) reformulated learned helplessness approach already tended to confirm that the style of causal attribution, in interaction with stressful life events, is a factor of cognitive vulnerability for the development of depression (Alloy et al., 1988; Alloy, Lipman and Abramson, 1992).
Nevertheless, in many other studies attributional style is not found to be related to depressive responses (Barnett and Gotlib, 1988; Brewin, 1985; Lewinsohn, Steinmetz, Larson and Franklin, 1981), even when a stressful element is present. The contradictory results in this field can perhaps partly be explained from the position of Alloy et al. (1992). These authors maintain that a depressive attributional style may contribute to the development of depressive symptoms and may increase the probability of depression, but that it is not a necessary condition for the generation of depressive reactions. Thus, further factors, cognitive or other, may also predispose subjects to experience depressive reactions.

Clinical investigation on deficiencies of social behaviour in depressive subjects
In the behavioural model of depression it is postulated that the decrease in the proportion of positive reinforce ment contingent on behaviour plays a fundamental role as a triggering stimulus for depressive behaviours (Lewinsohn, Youngren and Grosscup, 1979). In accordance with this assumption, general difficulties as regards social skills for obtaining the available social reinforcement have been associated with depression (Lewinsohn, Mischel, Chaplin and Barton, 1980), as have deficiencies in assertive behaviour (Raich, Carregal, Hernández and Sánchez, 1987; Sánchez v Lewinsohn, 1980). This approach has given rise to the therapeutic proposal presented in the book Control your depression (Lewinsohn, Muñoz, Youngren and Zeiss, 1986).
Although this perspective on depression is less develo ped than the cognitive approach, there is a wealth of material on the social behaviour of depressive people (Coyne, 1976; Lewinsohn, 1975; Lewinsohn et al., 1979; Rosenblatt and Greenberg, 1991; Sanz and Graña, 1991). It has even been found that measures of social skills predict subsequent depression (Wierzbicki, 1984). Other studies show that depressive subjects tend to be more sensitive to aversive social contingencies than con trol groups of normal and psychiatric subjects (Lewinsohn and Amenson, 1978; Lewinsohn, Lobitz and Wilson, 1973; Lewinsohn et al., 1980). This means that these subjects perceive the consequences of negative social interaction (for example, being criticised or disagreed with) in a more aversive way than normal and psychiatric subjects.
According to Lewinsohn and his co-workers, once the subject fails to emit an adaptive pattern of social behaviour, a decrease is generated in the reinforcement contingent on this behaviour which, in the event of being perpetuated, may intensify the depressive state, giving rise to a vicious circle that is difficult to break. In this sense, in the interactive model of depression it is proposed that depressive behaviours generate negative reac tions in others, which serve as negative feedback for maintaining the depressive state (Coyne, 1976). Coyne has shown that problematic social behaviour of depressive subjects (or with predisposition to depression) generates rejection by the social environment. However, the rejection is subtle: the depressive person (probably through his/her passive complaints) elicits feelings of guilt in others who, consequently, instead of responding in an openly hostile manner, offer nongenuine verbal support (Rosenblatt and Greenberg, 1991). The depressive person realises that he/she is not truly accepted, and consequently tries to control others' behaviour, generating more symptoms and expressing more affliction. The inconsistent and partial reinforcement generated by these behaviours makes them difficult to extinguish. Thus, reinforcement is very probably taking place, but in order to increase nonadaptive patterns of behaviour. Nevertheless, contradictory results with this approach have also been found in studies that do not find the phenomenon of negative feedback in depressive subjects (King and Heller, 1984; MacNeil, Arkowitz and Pritchard, 1987).

Implications of the exposed models for the treatment of depression
The models described are those that have generated most research on depressive reactions, being mainly focused on the study of the factors that facilitate or favour this type of reaction. It may occur, and the results reviewed are consistent with this, that different subjects present equally depressive behaviour controlled by different factors (Staats and Heiby, 1985). While the phenomenological nature of the depressive disorder may be similar (distorted thoughts, feelings of inferiority, low level of activity), both the distal environmental factors responsible for the onset of these reactions and, in turn, the proximal factors res ponsible for their taking root and continuing (for example, negative attributional style or deficiencies in social skills) may be different, and thus require the distinction of different types of depression. These types may predict the response to different types of treatment. It is on this basis that we can interpret the findings indicating that different treatments may be dealing with different levels, or dimensions, of the disorder (Maldonado, 1984; Polaino, Barcelo and Maldonado, 1991).
The classification of depressive behaviours should be made precisely on the basis of the factors that predict the effectiveness of therapy. In sum, we feel it necessary to resolve the existing contradictions between the theoreti calmethodological assumptions of clinical psychology (determining the treatment according to the particular characteristics of each disorder) and its practice (the increasingly widespread design and use of multicompo nent programmes).
Within this conceptual framework, the study presented here is aimed at testing whether the results of treating mild depression or dysphoria depend on whether or not the treatment is matched with the type of depression suffered by the subject. The results are considered on the basis of effects on depression measures and on anxiety level of subjects. Since we postulate that the differential effectiveness of treatments depends on their modifying the variables that lead to the development of depressive episodes, we consider it advisable to test our hypothesis on subjects with mild depression or dysphoria, in whom it is less probable that there is a conjunction of factors which, in continuous interaction, require a more com plex type of treatment. We leave, therefore, for future research the study of those variables on which we should take action to avoid the intensification of depressive behaviours and relapses.

METHOD
Subjects
The total sample of participants in this study was made up of 236 adolescents in secondary/further education, 97 of whom were male and 139 female. Mean age was 15 years 5 months, the age range being 14-23 years.

Materials and procedure Assessment instruments
(a) Depression. To evaluate the depressive state, we used the Centre for the Epidemiological Study of Depression Scale (CESD) constructed by Radloff (1977), which includes the following factors: (1) Somatic disorders, (2) Depressed affect, (3) Positive affect, and (4) Interpersonal problems. Subjects are requested to indicate the frequency with which they experienced each symptom during the previous week. Responses are assessed on a 4-point scale that ranges from "rarely or not at all (less than once a day)": 0, to "most or all of the time (5-7 days)": 3. This scale of 20 items has been used successfully to assess depressive symptoms in adolescents (Compass, Ey and Grant, 1993; Gotlib, Lewinsohn and Seeley, 1995). Its alpha coefficients range from .84 to .90, and its four-factor structure has been sufficiently endorsed (Joseph and Lewis, 1995). The total scores have a range of 0 to 60 points, with 17 or more being taken to indicate "possible" depression, and 23 or more "probable" depression. However, some authors recommend reducing the cut-off scores in order to decrease the frequency of false nega tives (Myers and Weissman, 1980). Thus, in order to select dysphoric subjects, we used the mean in depression of the sample under study (Mean: 14.10 points; S.D.: 8.39).
We also used Beck's Depression Inventory (BDI), designed by Beck et al. (1979) to assess the seriousness of the depressive symptoms and their response to treatment, specifically, its Spanish adaptation, composed of 19 items (Conde, Esteban and Useros, 1976).
(b) Attributional style. Attributional style was assessed by means of the Attributional Style Questionnaire (ASQ) of Seligman, Abramson, Semmell and Von Baeyer (1979) and Peterson, Semmell, Von Baeyer, Abramson, Metalsky and Seligman (1982). We used the version of Vázquez, Avia, Alonso and Fernández (1989), in which to the original dimensions (internal-external, stable-unstable and global-specific) they add the dimensions controllable-uncontrollable and personal-universal. These attributional dimensions were assessed only for negative events, since these correlate more strongly with depressive affect than positive events (Raps, Peterson, Reinhard, Abramson and Seligman. 1982; Seligman et al., 1979; Sweeney, Anderson and Bailey, 1986). To adapt the situations of the scale to our population of adolescents we carried out a pilot study on 198 adolescent students, with similar characteristics to the sampled subjects, who were asked to write down all those events that had given them bad feelings over the last two years. We found the following events to be those most frequently mentioned, and con sequently it was these that made up the final scale:
1. You spend some time worrying about school marks.
2. You fail a subject.
3. Your father punishes you and scolds you for somet hing that has happened.
4. You have some problem at home (such as an argu ment with a member of your family).
5. You have to repeat a school year.
6. You have an accident.
7. You have an argument with a friend.
8. You have a disappointment in love.
9. You have a problem with a teacher.
10. You lose someone's friendship.
11. Your pet dies.
The final score of the scale is obtained by including only the responses about those events that subjects consider important for themselves, operationalised in a score of 5 points on a scale that ranges from "no impor tance" (1 point) to "very important" (7 points).
(c) Social skills. To assess this variable we used the College SelfExpression Scale (CSES; Galassi, Deleo, Galassi and Bastien, 1974), in its Spanish adaptation by Caballo and Carrobles (1987). This scale was chosen as it concentrated specifically on subjects' self-assertion. It assesses three types of behaviour: positive expression, negative expression and negative consideration of one self. The peoplestimuli to which the scale refers are: strangers, figures of authority, family and relatives, and peers of both sexes.
(d) Anxiety. The Stimulus-Response Inventory of Anxiousness (SR) of Endler, Hunt and Rosenstein (1962) and Endler and Okada (1975) consists of 11 potentially anxietyinducing situations (involving inter personal threat, physical harm and ambiguous or novel situations), and 14 questions asking about the anxiety reactions of the subject faced with each of the situations presented. Specifically, three modalities of answer were considered: (1) fear, hindrance and avoidance; (2) optimism, joy and approach; and (3) autonomic reactions.
(e) Questionnaire for the assessment of various disor ders according to the criteria of the DSM III-R. A questionnaire was developed from the DSM III-R (APA, 1987) for the assessment of anxiety disorder, severe depression, obsessive-compulsive reactions and addic tion (current or previous) to drugs or alcohol.

Procedure used for the selection of subjects to be involved in the treatment conditions
The subjects to be employed were chosen in the follo wing way. A first selection was made on the basis of those who obtained scores of 14 points or more on the CESD. From this group we selected those who fulfilled the criteria for the formation of the two groups cogniti ve dysphoric and behavioural dysphoric. In the case of the cognitive dysphoric group, these criteria were the following: (a) showing a negative style of causal expla nation (scores higher than 0.25 standard deviations above the mean of the sample in the ASQ); and (b) absence of deficiencies in social skills (scores lower than 0.25 standard deviations below the mean of the sample in the CSES). Following the same criteria, the behavioural dysphoric group was made up of subjects who presented: (a) an intact attributional style, and (b) deficiencies in social skills. A previous study (Rodríguez-Naranjo and Godoy, 1996) lent support to the formation of these two groups on the basis of, among others, the following results: (a) Attributional style (ASQ) and social skills (CSES) demonstrated significant and independent predictive power for the scores in mild depression or dysphoria. (b) Other variables classically associated with depression, such as selfesteem, negati ve life events and anxiety were not found to associate significantly with the scores in dysphoria. (c) Dysphoric subjects that did not present deficiencies in either causal attributions or in social skills did not attain the criterion of mild depression in the BDI, and those who presented deficiencies in the two variables were the ones that sco red highly on this depression scale. (d) The most inte resting finding in this study was that subjects who presented deficiencies in causal attribution style, but not in social skills (cognitive dysphoric group), reported having experienced a significantly greater number of important negative life events than those subjects with deficiencies in social skills, but with an intact attributio nal style (behavioural dysphoric group). This latter group presented the same scarcity of negative life events as the nondysphoric subjects. These findings, obtained in two independent samples of dysphoric adolescents, allowed us to differentiate two types of dysphoria, in line with the suggestions of Alloy et al. (1988). In the current study, then, we follow the same criteria for the formation of the groups cognitive dysphoric and behavioural dysphoric. Of all the subjects of the sample assigned to one group or the other, we discarded those who, according to the criteria of the DSMIII R, suffered from any of the following disorders: anxiety disorder, severe depression, obsessivecompulsive reactions and addiction (current or previous) to drugs or alcohol. Also discarded were subjects that had received psychological or psychiatric treatment at some point in their lives, and those who had suffered serious physical disorders. Lastly, we excluded those subjects that the BDI indica ted were considering suicide and those that were unwilling to participate in the treatment.

Design used to constitute the treatment conditions
The final sample, used for comparing the effectiveness of matched and nonmatched treatments with the two types of mild depression or dysphoria described above, was made up of 30 subjects (9 males and 21 females) ranging in age from 14 to 23 years (mean age: 15.50; S.D. = l.74). Subjects were randomly assigned to one of the two following conditions: (a) matched treatment condition (cognitive dysphoric group treated with cogni tive therapy and behavioural dysphoric group treated with behavioural therapy) (n = 14); and (b) nonmatched treatment condition (cognitive dysphoric group treated with behavioural therapy and behavioural dysphoric group treated with cognitive therapy) (n = 16) (Table 1). The respective analyses were carried out on the subjects who responded to the scales completely: 30 subjects for the scale CESD; 29 subjects for the scale BDI (matched treatment: 14; nonmatched treatment: 15); and 19 sub jects on the scale SR (matched treatment: 9; nonmat ched treatment: 10).

Procedure carried out for the application of the cognitive and behavioural therapies
For the application of the treatments, two groups of cognitive therapy were formed, and another two of beha vioural therapy, each consisting of 7 or 8 subjects, mixed with regard to type of dysphoria, cognitive or behavioural. Both those subjects undergoing cognitive therapy and those undergoing behavioural therapy had weekly meetings with two therapists over a period of eight weeks. The therapists were blind with respect to the experimental conditions.
Cognitive therapy. The modification of the depressive style of causal explanation has received very little atten tion from the clinical perspective, such attention being limited, in terms of treatment for depression, to the reat tribution strategies suggested by Beck in his therapeutic procedure. This procedure has, however, been carried out in the experimental literature on achievement motivation (Försterling, 1980; Seligman, 1981).
In applying attributional retraining to those subjects with mild depression, our basic objective was that they learn to formulate adaptive causal explanations for negative events. To this end, we proceeded with the follo wing phases: (1) Subjects were provided with a cognitive explanation of emotional distress (Evans and Hollon, 1988). (2) The target situations of the treatment were determined (e.g., arguments with family and friends, disappointments in love, failing or having to repeat school subjects or years, etc.). We reviewed the causal explanations most frequently formulated by subjects about negative situations and analysed them on the basis of the attributional dimensions: internality, uncontrolla bility, stability and globality. (3) We analysed the evi dence for maintaining these causal interpretations about the target situations. Possible alternative interpretations were suggested and discussed with subjects. The process included the elements: (a) moving from internal causal explanations to external ones; (b) moving from internal and uncontrollable explanations with internal founda tions (e.g., "I'm to blame for having failed") to internal and controllable ones ("I failed because I didn't study enough"); (c) moving from stable explanations to unsta ble ones; and (d) moving from global explanations to specific ones. In cases where the attributed internal cause is considered pertinent, the objective is for sub jects to perceive this cause in terms of another, more adaptive dimension (for example, "they argued because I argued with them", moving from the global pole, "I always argue with everybody", to the specific pole, "on that occasion I argued with them"). (4) We analysed and discussed the negative conclusions derived from the nonadaptive causal explanations, mainly those related to selfblame and selfcriticism (Beck et al., 1979). With respect to point (d) above, we analysed how global explanations tend to lead to negative general conclu sions (for example, "my failures are due to my general lack of ability, so I'll probably fail in everything I do" to "I think I'm a failure"). (5) Lastly, we analysed the positive consequences of formulating alternative interpreta tions (external, controllable, unstable and/or specific) for negative events. Following the previous example, if failures are something specific that have a series of spe cific causes (what these causes may be are determined with the active cooperation of the subjects), the subject will not expect to fail in other activities. This is an inte ractive process in which, following the example, we move again to point (4), asking subjects to formulate examples of other negative conclusions that can be drawn from failing an exam or getting bad marks, analy sing the implicit attributional dimensions, generating alternatives and reassessing the previous conclusions.
Behavioural therapy. This consisted in carrying out training in coping skills for social interaction situations, giving particular importance to training in social skills. Specifically, the techniques used were those deriving from the work of Lewinsohn, Biglan and Zeiss (1976) for training in social skills, which have been shown to be effective for the treatment of depressive reactions (Antonuccio, Ward & Tearnan, 1989; Reed, 1994). The procedure consists of the following elements: (1) The social skills basis is explained as the cause of the emo tional distress. (2) The target behaviour is determined. (3) The target behaviour is modelled and practised. (4) Feedback and reinforcement from the group is imple mented after the realisation of the target behaviour. These steps are carried out for the training of the following behaviours: giving and receiving affection and praise, making and rejecting requests, dealing with criti cism, initiating and maintaining conversations, and dea ling with intimate relationships. Also included were skills of assertive acceptance and expression of positive feelings for increasing the social support received by subjects (Henderson, 1974). Both procedures being directed towards the treatment of subjects with mild depression, the treatment pro gramme has a marked didactic character; consequently, it is carried out without the inclusion of "homework" tasks.

RESULTS
To assess the differential effectiveness of matched and nonmatched treatments for the specific type of mild depression or dysphoria presented by subjects, we carried out an ANOVA for each depression measure (BDI and CESD) separately. The ANOVA was used in a mixed design of two factors with repeated measures in one of them. One factor was treatment condition (mat ched versus nonmatched) and the second was constituted by the treatment phase (preversus posttreatment). Thus, in the case of differences in effectiveness occurring bet ween the matched and nonmatched treatments, these would be produced by the interaction of the two factors. The results obtained show that the interaction is very close to statistical significance in the case of the BDI (F (1,27) = 3.81, p = .06), and statistically significant for the scores on the CESD (F(1,28) = 4.09, (p< .05). Figure 1 shows the differences between preand posttreatment scores for the BDI and CESD, in the two treatment conditions. Lastly, we carried out the same type of analysis to determine whether treatment condition (matched versus nonmatched) affected subjects' anxiety. The results show a decrease in anxiety after the matched treatment, whilst the nonmatched treatment was seen to produce a slight increase in these reactions. However, these diffe rences were not statistically significant (F(1,17) = 2.69, p = .12).

DISCUSSION
The results of this study allow us to draw a series of con clusions. Firstly, subjects with dysphoria or mild depres sion appear to improve differentially according to whet her they undergo treatment that is matched or nonmat ched with the type of mild depression from which they suffer. As expected, the dysphoria of those subjects to whom the treatment matched with the dysphoric state was applied decreased significantly more than that of those subjects to whom the nonmatched treatment was applied, at least as it is measured by the CESD. These results are concordant with the latest predictions of the reformulated model of learned helplessness in that the style of causal attribution is a predominant factor for a depression type (Abramson et al., 1989), in our study mild depression or dysphoria, referred to as cognitive. The results obtained also support previous findings about the role played by social skills for predicting depressive reactions (Lewinsohn, 1975). The differential effectiveness for decreasing dysphoria of attributional training, on the one hand, and of social skills training, on the other, according to the presence or absence of defi ciencies in these factors, supports our distinguishing of two types of mild depression or dysphoria (Rodríguez Naranjo and Godoy, 1996). Following García Hurtado, Fernández-Ballesteros, Montero and Heiby (1995), and in terms of the para digmatic behavioural theory (Heiby and Staats, 1990; Staats and Heiby, 1985), the deficiencies in social skills of dysphoric subjects may be constituted in the sensorymotor repertoire and the negative attributional style in the linguisticcognitive repertoire, which res pectively trigger two different types of dysphoria, which in turn respond differentially to therapy. We also feel that this approach might be complemented by the distinction between proximal and distal causes of depression, made by Alloy et al. (1988), since this approach allows a better understanding of the develop ment of depressive reactions. Thus, the interaction of negative life events with depressive attributional style (distal causes) permits us to predict the probable appe arance of a specific type of mild depression or dyspho ria, which we call cognitive dysphoria, and to differen tiate it from another type generated by deficiencies in social skills, which we call behavioural dysphoria (Rodríguez-Naranjo and Godoy, 1996). Other types of proximal causes, such as expectations of lack of con trol over one's own depressive reactions, may prolong or exacerbate depressive reactions. Teasdale (1985) refers to this as depression about depression, and emp hasises the important repercussions of this factor for therapy. In this case we would not be speaking, then, about factors that lead to dysphoria, but rather about factors that contribute to depressive behaviours being maintained, or even aggravated. The vicious circle that appears to constitute clinical depression means that it is very difficult to make valid generalisations about clinical depressives from research carried out with students that present subclinical depres sion or dysphoria (Coyne, 1994; FechnerBates, Coyne and Schwenk, 1994). Without contradicting the above, this study was carried out on the assumption that the dif ferential factors that predispose people to depression should be studied in subjects with low levels of depres sion, followed, ideally, by longitudinal studies. Consequently, the study of the effectiveness of treat ment for mild depression ceases to be a mere analogy of clinical research and becomes a study carried out under real treatment conditions. In conclusion, these results demonstrate the usefulness of distinguishing different types of depression, at least where mild depression or dysphoria is concerned. The fact that we have identified predictive, and possibly determinant, factors of two types of mild depression has helped us to clarify some of the factors on which the effectiveness of different treatment strategies depends. As early as the 1950s it was suggested that if the experi mental method could be applied to the clarification of the basic causal factors underlying functional abnorma lities, then, evidently, the same techniques (aimed at the experimental manipulation of these factors) could be applied to changing the abnormal behaviour (Yates, 1970). Today, research has developed from the study of spe cific cases to levels of more generalisability when making predictions about change. Thus, studies that identify factors of vulnerability or risk for suffering from future disorders are more and more frequent. The next step in our work must therefore be to test whether treatments matched with a type of dysphoria are also more effective than nonmatched treatments for pre venting subjects demonstrating highrisk characteris tics from developing depressive behaviour in the futu re, both in the school context and in that of institutions whose purpose is to promote health. This is an area in which great advances can, and should, be achieved in coming decades.

REFERENCES
Abramson, L. Y., Metalsky. G. and Alloy. L. B. (l989). Hopelessness depression: A theorybased subtype of depression. Psychological Review, 96, 358-372.

Alloy. L. B., Abramson L. Y., Metalsky, G. I. and Hartlage, S. (1988). The hopelessness theory of depression: Attributional aspects. British Journal of Clinical Psychology, 27, 5-21.

Alloy. L. B., Lipman. A. J. and Abramson, L. Y. (1992). Attributional style as a vulnerability factor for depression: Validation by past history of mood disorders. Cognitive Therapy and Research, 16, 391-407.

American Psychiatric Association (APA) (1987). Diagnostic and statistical manual of mental disorders: DSM IIIR. (3. ed. rev.). Washington, D. C.: APA. (Spanish version: Masson, Barcelona, 1988).

Antonuccio, D. O., Ward. C. H. and Tearnan, B. H. (1989). The behavioral treatment of unipolar depres sion in adult outpatients. In M. Hersen, R. M. Eisler and P. M. Miller (Eds.). Progress in Behavior Modification, Volume 24 (pp. 152-191). Newbury Park: Sage Publications.

Barnett, P. A. and Gotlib. I. H. (1988). Psychosocial functioning and depression: Distinguishing among antecedents, concomitants and consequences. Psychological Bulletin. 104, 84-96.

Beck, A. T. (1967). Depression: Clinical, experimental and theoretical aspects. New York: Harper & Row.

Beck, A. T., Rush, A. J., Shaw, B. F. and Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press. (Spanish version: Desclèe de Brouwer, Bilbao, 1983).

Buceta, J. M. and Polaino-Lorente, A. (1982). Reformulación del modelo de "Learned Helplesness" desde el punto de vista de la psicología atribucional (Reformulation of the "learned helplessness" model from the perspective of attributional psychology). Revista de Psicología General y Aplicada, 37, 13-29.

Brewin. C. R. (1985). Depression and causal attribu tions: What is their relation? Psychological Bulletin, 98, 297-309.

Caballo, V. E. and Carrobles, J. A. I. (1987). Adaptación de la escala "The College Self-Expression Scale (CSES)" (Adaptation of the CSES) of Galassi, Deleo, Galassi and Bastien (original version from 1974). In R. Fernández-Ballesteros and J. A. I. Carrobles (Eds.), Evaluación conductual. Madrid: Pirámide.

Compas, B. E., Ey, S. and Grant. K. E. (1993). Taxonomy, assessment and diagnosis of depression during adolescence. Psychological Bulletin, 114, 323-345.

Conde. V., Esteban, T. and Useros, E. (1976). Revisión crítica de la adaptación castellana del Cuestionario de Beck (Critical review of the Spanish adaptation of Beck's questionnaire). Revista de Psicología General y Aplicada, XXXI (140), 469-497.

Coyne, J. C. (1976a). Depression and the response of others. Journal of Abnormal Psychology, 85, 186-193.

Coyne, J. D. (1976b). Toward an interactional descrip tion of depression. Psychiatry, 39, 28-40.

Coyne, J. C. (1994). Selfreported distress: analog or ersatz depression? Psychological Bulletin, 116, 29-46.

Endler, N. S., Hunt, J. and Rosenstein, A. J. (1962). An SR Inventory of Anxiousness. Psychological Monographs: General and Applied, 536.

Endler, N. S. and Okada, M. A. (1975). A multidimen sional measure of trait anxiety: The SR Inventory of general trait anxiousness. Journal of Consulting and Clinical Psychology, 43, 319-329.

Ellis, A. (1962). Reason and emotion in psychotherapy. Secaucus, NJ: Lyle Stuart. (Spanish version: Desclèe de Brouwer, Bilbao, 1980).

Ellis, A. and Grieger, R. (1977). Handbook of rational emotive therapy. New York: Springer. (Spanish version: Desclèe de Brouwer, Bilbao, 1981).

Ellis, A. (1987). A sadly neglected cognitive element in depression. Cognitive Therapy and Research 11, 121-146.

Evans, M D. and Hollon, S. D. (1988). Patterns of per sonal and causal inference: Implications for the Cognitive Therapy of Depression. In L. B. Alloy (Ed.). Cognitive Processes in Depression (1 st . Ed), (pp. 344-377). New York: Guilford Press.

Fechner-Bates, S., Coyne, J. C. and Schwenk, T. L. (1994). The relationship of selfreported distress to depressive disorders and other psychopathology. Journal of Consulting and Clinical Psychology, 62, 550-559.

Feinberg, M. (1992). Comment: Subtypes of depression and response to treatment. Journal of Consulting and Clinical Psychology, 60, 670674.

Försterling, F (1980). Attributional aspects of cognitive behavior modification: A theoretical approach and suggestions for techniques. Cognitive Therapy And Research, 4, 27-37.

Galassi, J. P., Deleo, J., Galassi, M. D. and Bastien, S. (1974). The College SelfExpression Scale: A measu re of assertiveness. Behavior Therapy, 5, 165-171.

Garber, J., Miller, W. R. and Seaman, S. F. (1979). Learned helplessness, stress and the depressive disorders. In R. A. Depue (Ed.), The Psychobiology of the Depressive Disorders: Implications for the Effects of Stress, (pp. 335364). New York: Academic Press.

García Hurtado, J., FernándezBallesteros, R., Montero, I. and Heiby, E. M. (1995). Multiple correlates of unipolar depression: contributions from the paradigmatic behavioral theory. Psicothema, 7, 41-50.

Gotlib, I. H., Lewinsohn, P. M. and Seeley, J. R. (1995). Symptoms versus a diagnosis of depression: differences in psychosocial functioning. Journal of Consulting and Clinical Psychology, 63, 90-100.

Heiby, E. M. and Staats, A. W. (1990). Depression and its classification. In G. Eifert and I. Evans (Eds.), Paradigmatic Behavioral Therapy, New York: Springer. Henderson, S. (1974) Careeliciting behavior in man. Journal of Nervous and Mental Disease, 159, 172-181.

Joseph, S. and Lewis, C. A. (1995). Factor analysis of the Center for Epidemiological StudiesDepression scale. Psychological Reports, 76, 40-42.

Kendall, P. D. and Korgeski, G. P. (1979). Assessment and cognitive behavioral intervention. Cognitive Therapy and Research, 3, 121.

King, D. A. and Heller, K. (1984). Depression and the responses of others: A reevaluation. Journal of Abnormal Psychology, 93, 477-480.

Lewinsohn. P. M. (1975). The behavioral study and treatment of depression. In M. Hersen, R. M. Eisler and P. M. Miller (Eds.), Progress in Behavior Modification. Vol. I, New York: Academic Press.

Lewinsohn, P. M. and Amenson, C. S. (1978). Some relations between pleasant and unpleasant moodrela ted events and depression. Journal of Abnormal Psychology, 87, 655-664.

Lewinsohn. P. M., Lobitz, W. C. and Wilson, S. (1973). "Sensitivity" of depressed individuals to aversive sti muli. Journal of Abnormal Psychology, 81, 259-263.

Lewinsohn, P. M., Mischel, W., Chaplin. W. and Barton, R. (1980). Social competence and depression The role of illusory selfperceptions? Journal of Abnormal Psychology, 89, 213-219.

Lewinsohn, P. M., Muñoz, R., Youngren, M. A. and Zeiss, A. (1986). Control your depression (rev. ed.). Englewood Cliffs, NJ: Prentice Hall.

Lewinsohn, P. M., Steinmetz, J. L., Larson, D. W. and Franklin, J. (1981). Depressionrelated cognitions. Antecedent or consequence? Journal of Abnormal Psychology, 90, 213-219.

Lewinsohn, P M., Youngren, M. A. and Grosscup, S. J. (1979). Reinforcement and depression. In R. A. Depue (Ed.), The Psychobiology of Depressive Disorders: Implications for the Effects of Stress, New York: Academic Press.

MacNeil, D. E., Arkowitz, H. S. and Pritchard, B. E. (1987). The response of others to facetoface interac tion with depressed patients. Journal of Abnormal Psychology ,96, 341-344.

Maldonado, A. (1984). Terapia de conducta y depresión: "Un análisis experimental de las interacciones entre tratamientos cognitivos y conductuales con tra tamientos farmacológicos en sujetos depresivos" (Behavioural therapy and depression: "An experi mental analysis of the interactions between cognitive and behavioural treatments with pharmacological tre atment in depressed subjects"). Revista de Psicología General y Aplicada, 39, 517-535.

Myers. J. K. and Weissman, M. M. (1980). Use of a self report symptom scale to detect depression in a com munity sample. American Journal of Psychiatry, 137, 1081-1084.

Peterson, C., Semmell, A., Von Baeyer, C. Abramson, L. Y. Metalsky, G. I. and Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive Therapy and Research, 6, 287-299.

Polaino, A., Barcelo, M. and Maldonado, C. L. (1991). Versatilidad atribucional y tratamiento farmacológico de la depresión (Attributional versatility and phar macological treatment of depression). Revista de Psiquiatría de la Facultad de Medicina de Barcelona, 18, 173-184.

Radloff. L. S. (1977). The CESD scale: a selfreport depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.

Raich, R. M., Carregal, M. D., Hernández. L. and Sánchez. M. L. (1987). Estudio sobre creencias irracionales, asertividad y depresión en estudiantes universitarios (Study on irrational beliefs, assertiveness and depression in university students). Revista de Psiquiatría y Psicología Médica, 18, 97-106.

Raps, C. S. , Peterson, C., Reinhard, K. E., Abramson, L. Y. and Seligman, M. E. P. (1982). Attributional style among depressed patients. Journal of Abnormal Psychology, 91, 102-108.

Reed, M. K. (1994). Social skills training to reduce depression in adolescents. Adolescence, 29, 293-302.

Rodríguez-Naranjo and Godoy (1996). Attributional style and social skills deficits as predictors of dysp horic states and response to treatment. Manuscript submitted for publication.

Rosenblatt, A. and Greenberg, J. (1991). Examining the world of the depressed: Do depressed people prefer others who are depressed? Journal of Personality and Social Psychology, 60(4), 620-629.

Sánchez, V. and Lewinsohn, P. M. (1980). Assertive behavior and depression. Journal of Consulting and Clinical Psychology, 48(1), 119-120.

Sanz, J. and Graña, J. L. (1991). Factores psicosociales y síntomas depresivos: El caso de la autoobservación (Psychosocial factors and depressive symptoms: The case of selfobservation). Psicothema, 3, 381-399.

Seligman. M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco: W. H. Freeman.

Seligman, M. E. P (1981). A learned helplessness point of view. In L. P. Rehm (Ed.), Behavior Therapy for Depression. Present Status and Future Directions. New York: Academic Press.

Seligman, M. E. P. Abramson, L. Y. . Semmell, A. and Von Baeyer, C. (1979). Depressive attributional style. Journal of Abnormal Psychology, 88, 242-247.

Smith, T. W. (1989). Assessment in rational emotive the rapy: Empirical access to the ABCD model. In M. E. Bernard and R. DiGiuseppe (Eds.), Inside Rational Emotive Therapy. A Critical Analysis. New York: Academic Press.

Staats. A. W. and Heiby, E. M. (1985). Paradigmatic behaviorism's theory of depression: Unified, explana tory and heuristic. In S. Reiss and R. R. Bootzin (Eds.), Theory Issues in Behavior Therapy. Orlando, FL. Academic Press.

Sweeney, P. D., Anderson, K. and Bailey. S. (1986). Attributional style in depression: a metaanalytic review. Journal of Personality and Social Psychology, 50, 974-991.

Teasdale, S.D. (1985). Psychological treatments for depression. How do they work? Behaviour Research and Therapy, 23(2), 157-165.

Vázquez, C., Avia, M. D., Alonso, B. and Fernández, E. (1989). A test of the diathesisstress model of depres sion Attributional style as a predictor of dysphoric reactions after an academic failure. Communication presented to the 1st European Congress of Psychology, Amsterdam.

Yates, A. (1970). Behavior Therapy. New York: Wiley. (Spanish version: Trillas, México, 1980).

Wierzbicki, M. (1984). Social Skills Deficits and Subsequent Depressed Mood in Students. Personality and Social Psychology Bulletin, 10(4), 605-610.

View this Table:
[in this window]
[in a new window]

Table 1. Assignment of subjects to the treatment conditions "marched and non-matched with type of dysphoria".

View this Figure:
[in this window]
[in a new window]

Figure 1 . Differences between pre- and post-treatment scores on the BDI and CES-D for the treatment conditions (matched versus non-matched) with type of dysphoria. BDI: Beck's Depression Inventory. CES-D: Centre for Epidemiological Study of Depression Scale.

Copyright 1998 by the	Psychology in Spain, Vol 2 No 1, 17-26
Colegio Oficial de Psicólogos