Printer Friendly

The Impact of an Enriched Environment on the Relationship Between Activation and Depression in Latinx and Non-Latinx Students.

Depression is one of the most prevalent psychological disorders and is the leading cause of disability globally (World Health Organization, 2017). Given the frequently debilitating nature of depression and the high prevalence rate of depressive episodes (Substance Abuse & Mental Health Services Administration, 2017), there has been much theoretical speculation about the variables affecting the course and magnitude of depressive symptoms. One approach that has received a large amount of both theoretical and empirical investigation is anchored in behavioral theory (Skinner, 1974). From this viewpoint, the etiology and maintenance of depression is heavily influenced by both an organism's own behavior, specifically activation (Kanter, Mulick, Busch, Berlin, & Martell, 2007), as well as by the environment of the organism (Armento & Hopko, 2007). Although this conceptualization of depression has received extensive support in the form of component analysis of CBT (Jacobson et al., 1996) and treatment studies (Ekers et ah, 2014), few studies have examined the association between activation and depression simultaneously. Thus, the goal of the current study was to examine the moderating role of environmental enrichment in the association between activation and depressive symptoms.

Early functional analysis of depression (Ferster, 1973) implicated lower frequencies of positive reinforcement of nondepressive behaviors and increased frequency of negative reinforcement of behaviors associated with depression, such as complaints and requests for help, as a root cause of depression. Contemporaries of the time such as Lewinsohn (1974) endorsed similar viewpoints, characterizing depression as the loss of, or decrease in, response contingent positive reinforcement (i.e., fewer instances of the frequency or duration of a behavior increasing as a result of positive outcomes). A more recent functional analysis of depression (Kanter, Cautilli, Busch, & Baruch, 2005) built upon earlier conceptualizations to depict a more nuanced understanding of the behaviors of interest. In their analysis, Kanter et al. identify five factors generally agreed upon in behavioral conceptualizations of depression. These factors include lack of response contingent reinforcement of behavior, excessive punishment of behavioral responses, loss of effective operant behavior, positive reinforcement for depressive behavior, and negative reinforcement of depressive behavior. Kanter et al. identified two additional factors that may contribute to depression but lack scientific consensus. The first of these factors, overreliance on rule-governed behavior, contributes to depression by preventing naturally occurring contingencies from governing behavior. Instead, individuals relying on rule-governed behavior to guide action engage in rigid behaviors that are verbally mediated and not naturally consequated. The second additional factor is environmental factors that act as establishing operation for depression. Factors such as obtaining less sleep and ending relationships may increase the likelihood that a person engages in escape behavior, which is a key factor in a behavioral interpretation of depression. When taken in conjunction, these analyses point to a model of depression that is self-perpetuating; a lack of positive reinforcement leads to less activation and more depressive symptoms, which further decreases positive reinforcement and leads to a strengthening of the cycle and maintenance of depressive symptoms through negative reinforcement.

These functional analyses also indicate the mechanism by which Behavioral Activation (BA; Jacobson et al., 1996) treatments for depression are thought to work. BA is a frontline treatment for depression. In general, BA involves individuals acting in mood-independent behavior, resulting in effective engagement with the environment, increasing rates of response contingent reinforcement, and decreasing reinforcement for behaviors that are characterized as "depressed" in nature. BA can be viewed as explicitly addressing all factors identified in Kanter et al. (2005) functional analysis (Kanter et al., 2010). In addition to being included as a component of empirically supported treatments such as Cognitive Therapy for depression (Beck, 1979) and Acceptance and Commitment Therapy (Hayes, Strosahl, & Wilson, 1999), there is a large, accumulating base of evidence supporting BA as a standalone treatment for depression (e.g., Dimidjian et al., 2006; Mazzucchelli, Kane, & Rees, 2009). Although the evidence supporting BA as an intervention for depression appears to be consistent with a behavioral model of depression, it does not provide conclusive evidence for the behavioral conceptualization of depression itself.

Any conclusive empirical evaluation of a behavioral model of depression must capture three critical variables. The first critical variable is the level of activation that an individual is engaging in. To frilly capture the activation construct, it's important to not only capture activation, but also avoidance of aversive stimuli (Martell, Addis, & Jacobson, 2001). In essence, in order to have nondepressed behavior reinforced, a person must be engaging in nondepressed behavior, i.e., activation. In addition to activation and avoidance (henceforth referred to as activation), the ability of the environment to promote and maintain activation must also be understood. Based on the behavioral model of depression, depression initially develops due to the inability of the environment to effectively reinforce nondepressed behavior. As such, the availability of reinforcement (sometimes referred to as reward; Armento & Hopko, 2007) or probability of response-contingent positive reinforcement (Carvalho et al., 2011) in the environment is an important construct to evaluate in conjunction with activation. The conceptual incorporation of environmental reward into a behavioral model of depression can be conceptualized as follows: if an individual engages in nondepressed behaviors but that behavior is not reinforced, the behavior will stop occurring. In looking at activation and environmental reward, simultaneous evaluation of the constructs stands out as uniquely important. Individuals need to both engage in behaviors (activation) and have those behaviors consequated in order to prevent depression. If a person fails to either engage in behavior or the environment doesn't effectively consequate the behavior, the likely result is depression. Finally, in order to evaluate a model of depression, researchers must collect a measure of depression. Although there are many validated and commonly used measures of depression such as the Beck Depression Inventory-II (Beck, Steer, & Brown, 1996) and the Hamilton Rating Scale for Depression (Hamilton, 1960), tools for evaluating activation and environmental reward have less empirical support and have been infrequently used (Manos, Kanter, & Busch, 2010).

Research groups seeking to elucidate the behavioral model of depression have adopted different strategies to measure the availability of rewards in the environment (Manos et al., 2010). Early techniques included diary-card monitoring (Lewinsohn & Atwood, 1969), assessment of pleasant events that are being engaged in (MacPhillamy & Lewinsohn, 1982), measurement of aversive events (Lewinsohn & Talkington, 1979), and measurement of interpersonal events (Youngren & Lewinsohn, 1980). More recent techniques have focused on self-reported levels of activation (Kanter et al, 2007), functional, rather than topographical, approach to assessing availability of reinforcement in an individual's environment (Armento & Hopko, 2007) or the probability of environmental reward (Carvalho et al., 2011). Evidence supporting a behavioral model of depression using these measurement strategies is promising, but limited (Manos et al., 2010). Studies assessing the importance of the environment and activation have been conducted in parallel studies, rather than in conjunction with limited exceptions (e.g., Hill, Buitron, & Pettit, 2017). Understanding the relationship between activation, the availability of environmental rewards, and depression may have important implications for the implementation of BA and conceptualization of depression using a behavioral model.

The current study was designed to explore the relationship among activation, availability of environmental rewards, and depression. These relationships are explored in the context of a behavioral model of depression with a cross-sectional analysis of an undergraduate sample of convenience. Two hypotheses were tested. The first hypothesis is that the availability of environmental reward and activation would both contribute to the variance in depression. Further, it was hypothesized that each of these constructs would provide unique contributions. Second, it was hypothesized that the relationship between activation and depression would be moderated by the availability of environmental rewards.


The study, approved by the human subjects institutional review board, included administration of an online cross-sectional survey created to answer the proposed research questions. The current study included analysis of three questionnaires from a larger questionnaire battery. Subjects were recruited from undergraduate psychology courses at a 4-year, medium-sized, primarily nonresidential, Hispanic-Serving institute. To be eligible to participate in the study, potential participants had to be undergraduate students enrolled at the institute conducting the research and over the age of 18. Course instructors sent out email announcements informing students of the research opportunity, informed students that the study consisted of completing a series of questionnaires, the anticipated time commitment of the survey, and a link to an online questionnaire platform to complete the surveys. After reading an information statement, 409 participants agreed to complete the questionnaires provided. In order to prevent participants from accidentally skipping questions, answers had to be provided to all questions on the assessment battery. Participants were given the option to indicate that they chose not to respond to questions. As such, all missing data was intentionally omitted by participants. Approximately 75% of participants completed all items on the measures of interest, e.g., provided answers other than "I choose not to respond." Participants who had more than 20% of their data missing on any measure were not included in the final analyses (N = 374). Creating a mean score for participants with missing data offers a closer approximation of a participant's responses in comparison to a sum score. Thus, mean scores were used for all analyses rather than sum scores. The sample was overwhelmingly female (85.3%) and Latinx (60.4%), and 90% of the sample was between ages 18 to 24 years old (range: 18-84). Remaining demographic information can be found in Table 1.


Behavioral Activation for Depression Scale (BADS) The BADS is a 25-item scale developed to measure avoidance and activation. Factor analysis of the measure identified four factors (activation, avoidance/rumination, work/school impairment, and social impairment; Kanter et al., 2007). Items are scored on a 7-point (0-6) scale where higher scores indicate more activation and items were averaged to create a mean behavioral activation score. The measure has good internal consistency (a = .87) and acceptable 1-week test-retest reliability (r = .74). Internal consistency within the current sample was excellent ([alpha] = .92). Normative data between the original sample (M- 110.51, SD = 21.04) and the current study (M = 90.23, SD = 24.25) indicated substantial differences. Previous research has found strong correlations between the BADS and depression.

Environmental Reward Observation Scale (EROS) The EROS is a 10-item measure that assesses environmental reward. The measure is designed to assess the availability of reward in the environment and the ability of individuals to come in contact with the available reinforcement. Although typically administered as a 4-point Likert scale, a modified 7-point Likert scale version of the measure was used in the current study, and items were averaged to compute a mean score. The nonmodified EROS possesses excellent 1-week test-retest reliability and internal consistency (r = .85, [alpha] = .86; Armento & Hopko, 2007), and this modified EROS possesses similar strong internal consistency ([alpha] = .87). As would be expected, normative data on the modified version of the EROS (M = 46.59, SD = 10.91) differed from the results in the seminal article (Administration 1 [M = 29.46, SD = 4.86], Administration 2 [M = 30.33, SD = 4.86]; Armento & Hopko, 2007). When mean values from the present study are weighted to equate values to a 4-point scale, the obtain means are similar but slightly lower than those obtained in the seminal article whereas the standard deviations are higher. The skewness and kurtosis of the nonmodified EROS (-.18 and -.69, respectively) were similar to those found in the modified version (-.14 and -.37, respectively). Previous research has found strong correlations between the EROS and depression.

Beck Depression Inventory II (BDI) The BDI (Beck et al., 1996) is a measure designed to assess the presence and severity of depression. It includes 21 items rated on a 4-point Likert Scale that were averaged to compute a mean score (a = .93). The mean score in the present study (M = 14.30, SD = 10.92) differed slightly from that given by Beck et al. (1996; a = .92, M = 12.55, SD = 9.93). Higher scores on the BDI indicate higher levels of depression. The instrument is widely used and has excellent reliability and validity (Huang & Chen, 2015).

Data analytic approach There were no differences on the predictors and outcome of interest between those whose data were excluded and included. Independent sample (-tests were conducted to ascertain any differences in variables of interest across gender (male/female) and ethnicity (Latinx/non-Latinx; Salinas & Lozano, 2017). One-way ANOVAs were conducted to ascertain any differences in the variables of interest across race and employment status. To ascertain the presence of any outliers, we ran and examined descriptive statistics (see Table 2) and histograms of all variables. No data was removed as a result of this process.

Hypotheses testing Hierarchical multiple regression analyses using R (R Core Team, 2017) and RStudio software (RStudio Team, 2016) was used to examine the three hypotheses. Because BADS and EROS were highly correlated (see Table 2), partial regression coefficients were calculated for all models to ascertain the unique effects of each variable on BDI. Hypothesis 1 examined the unique contribution of BADS and EROS on BDI. To test Hypothesis 1, variables were centered, and BADS and EROS were entered as main effects of BDI. Hypothesis 2 examined EROS as a moderator of the association between BADS and BDI. To examine the moderating effect of EROS on the association between BADS and BDI, BADS and EROS were entered as main effects in Step 1. BADS, EROS, and an interaction between BADS and EROS were entered in Step 2. Given the high correlation between EROS and BADS, the variance inflation factor (VIF) was included to detect the degree of multicollinearity. VTF values have a lower bound of 1 and increase as the single variable is highly correlated with the other predictors. Although there is debate regarding at what point VIF values suggest unstable estimation due to multicollinearity (O'Brien, 2007), general guidelines suggest VIF values greater than 10 suggest high levels of multicollinearity (James, Witten, Hastie, & Tibshirani, 2013), whereas more conservative recommendations suggest values greater than 2.5 warrant caution in interpretation (Allison, 2012).

Interpretation of significant interactions necessitates ascertaining whether the simple slopes are significantly different than 0 (Hayes et al., 2013). The interactions were graphed and probed by examining the conditional effects (simple slopes) of BADS on BDI by setting the values of the moderator (EROS) at five different values (-2SD, -1 SD, Mean, 1 SD, 2SD) using interActive, a web-based, data visualization application (McCabe, Kim, & King, 2018). Multiple values of the moderator were chosen to enable a precise examination of the full spectrum of effects of EROS on the BADS and BDI association. InterActive creates an individual plot for the simple slope at each level of the moderator, thus enabling the viewer to understand changes in the region of significance across the moderator (McCabe, Kim, & King, 2018). Marginal-effects plots were also plotted to visualize the regions of significance (Preacher, Curran, & Bauer, 2006).


There were no significant differences across any of the variables of interest among gender, race, ethnicity, and employment status (all p > .07). Means, standard deviations, and zero-order correlations of the model variables appear in Table 2. Because the majority of the extant literature utilizes sum scoring procedures to score these variables, total scores were computed and presented in Table 1 and Table 2 whereas all analyses were conducted using mean scores. Unstandardized and standardized coefficients for hypotheses testing appear in Table 3. The [R.sup.2] of BADS on BDI was .55, indicating that BADS contributes 55% of the variance of BDI. The [R.sup.2] of EROS on BDI was .49, indicating that EROS contributes 49% of the variance of BDI. It should be noted that the correlation between BADS and EROS was .736, thus there is considerable shared variance in the contribution of EROS and BADS on BDI. All VIF values fell below 2.5, suggesting that by conservative recommendations for VIF, the regression results may be interpreted (see Table 3).

Hypothesis 1 was supported. Both BADS and EROS were significantly, negatively associated with BDI [[beta] = -.49 (95% CI:, -.33, -.22), and [beta] = -.34 (95% CI:, -.21, -.12)], respectively. This model contributed 60.3% of the variance of BDI. The partial coefficients for BADS and EROS were .468 and -.342, respectively. This model was rerun as Step 1 in the model testing Hypothesis 2 that evaluated a moderation model of EROS and BADS on BDI. Hypothesis 2 was supported. There was a significant interaction between BADS and EROS on BDI (see Figure 1). Visualization through the marginal effects plot suggests that BADS was significantly, negatively associated with BDI between 2 standard deviations above and below the mean of EROS, and 99% of the values of EROS lay within this range (see Figure 2). However the magnitude of the effect of BADS on BDI differed across EROS (see Figure 3). At low levels of EROS (e.g., 1 and 2 standard deviations below the mean of EROS), the magnitude of the effect of BADS on BDI was stronger than at higher levels of EROS (e.g., 1 and 2 SD above the mean of EROS), suggesting the strength of the association between BADS and BDI was strongest when EROS was low relative to when it was high. The full model contributed 64% of the variance of BDI. The change in [R.sup.2] with the addition of the interaction term in Step 2 was statistically significant [F(1) = 35.58, p < .001)].


Findings from the present study replicate and expand upon previous research on mechanisms of depression from a behavioral perspective. The results presented support previous findings linking activation and avoidance to depression (Kanter, Rusch, Busch, & Sedivy, 2009) and the availability of environmental reward to depression (Armento & Hopko, 2007). Significant additions to this simple model can be found with a path between activation and depression moderated by the availability of environmental reward. These findings provide incremental evidence to the understanding of a behavioral framework of depression. The magnitude of the incremental increase in understanding is limited by the cross-sectional nature of the data collected. As such, the interpretations of the data are guided by theory rather than data-based interpretations. Nonetheless, these data provide an important foundation for more resource intensive and methodologically rigorous studies to build from.

To the best of our knowledge, these data represent the first time that the BADS and the EROS have been investigated concurrently. This provides unique insight into the measurement of behavior and contact with environmental reinforcement. Consistent with hypothesized models, our findings suggest that the less enriched the environment, the more activation plays a role in depression. These results compliment recent findings on reward probability (Hill et al., 2017). The findings suggest that when the environment doesn't offer sufficient reward, individuals will need to engage in more behavior to come in contact with a reward. As the environment becomes more enriched, individuals will come in contact with reinforcement on a more frequent schedule of reinforcement, reducing the importance of activation in isolation as it relates to depression.

Theorists have posited a bidirectional relationship between environmental reward and activation (Manos et al., 2010). Findings from the current study are consistent with this model, indicating that the initial primary treatment target of many approaches to BA, activation (Jacobson et al., 1996; Lejuez, Hopko, Aciemo, Daughters, & Pagoto, 2011), is an important target; that behavior change is suggested when the environment is not rewarding, but as environmental reward increases, intentional activation become less important. These findings may have important treatment implications. Although the goal of BA is ultimately to decrease depressive symptoms, these findings suggest that at different parts of the intervention, treatment providers should be emphasizing different components of the intervention. In developing a case conceptualization for a BA intervention, including both the BADS and EROS could suggest what to target in the intervention. For instance, a low score on the EROS but a high score on the BADS would suggest the therapist work to increase the availability of reward in the environment. This could potentially include building in different activities, working to change how the client engages with their activities, or enhancing opportunities for reinforcement of behaviors by increasing the client's social network or contact with their social network. Conversely, a client with a low score on the BADS but a high score on the EROS would suggest developing a robust repertoire of mood independent behaviors that can facilitate the client starting behaviors that will theoretically be maintained by the environment. The client who is both active and reports a high level of available environmental reward, would suggest an intervention relating to more complex behaviors that would facilitate their self-report more accurately representing their environment and behaviors. It is possible that these are the situations in which values work is best suited or repertoires related to verbal behavior and relational framing are modified. Finally, low scores on both measures would indicate a need to work at increasing mood independent behavior, while simultaneously creating a more enriched environment. Through this data-driven case-conceptualization process, treatment providers may seek to utilize principle driven modalities of treatment as opposed to a manualized version that may be less sensitive to contextual factors.

The current study substantially increases the support for a behavioral model of depression across populations. The fact that no differences were found between groups (e.g., males and females, Latinx and non-Latinx) in a large diverse sample corroborates evidence on the effectiveness of the intervention with traditionally underserved populations (Kanter et al., 2015; Benson-Florez, Santiago-Rivera, & Nagy, 2017). Although further research is needed on the applicability of the therapy to diverse samples, initial findings are promising for the validity of a behavioral model with individuals from diverse backgrounds. One variable that has been incorporated into treatment manuals (Kanter et al., 2010) that needs further empirical investigation is values. Conceptually, values are thought to augment the reinforcing properties of the consequences with which various behaviors come in contact. Given the significant differences in values of various cultures, treatment manuals for the application of culturally sensitive BA have emphasized values. However, little empirical investigation on values has been conducted in studies of behavioral models of depression or BA with or without individuals from traditionally underserved populations. In this regard, treatment implementation research (e.g., Mir et al., 2015) in BA may be outpacing research on the theoretical model and empirical investigation of theoretically important components. If this is the case, it is possible that therapists are currently engaging in ineffective clinical work or adding components to the treatment that although appetitive or effective, are unneeded.

Findings from the current study are notably discrepant from previous studies. In the current study, activation and the availability of environmental rewards were found to contribute 60.3% of the variance in depression. In a study designed to assess the mediating effect of behavioral activation on the relationship between reward probability and environmental suppression on depression, Hill et al. (2017), found their model only accounted for 26.3% of the variance in depression scores. Although the phenomena being measured and the methodology used are significantly different, the magnitude of the differences in findings are notable. If future investigations replicate findings from the current study, it may be indicative of specific constructs that should be both targeted and measured in research on behavioral activation.

Limitations and Future Directions

The current study has several significant limitations. The data collected in the study are cross-sectional in nature. As such, interpretations of the data are guided by theory rather than methodology that would support the interpretations. Although it is important for the relationship between the depression, activation, and environmental reward to be established, there are a number of possible explanations for the findings that should be explored. For instance, it is possible that activation and environmental reward are a consequence of depression rather than a cause of depression. Further, it is entirely possible that BADS scores should be interpreted as determining the strength of the relationship between the EROS and BDI. In addition, it is possible that other variables, such as social interactions, may play a role in a comprehensive model of depression. To this end, future research should explore the possibility of manipulating activation or environmental enrichment in order to provide support for causality. It is also possible to explore these phenomena in a nonexperimental context via longitudinal assessment of the constructs.

A second significant limitation to the current study can be found in the measurement strategies utilized. For instance, the current study utilized a modified version of the Environmental Reward Observation Scale. Although there is no reason to suspect that a 7-point Likert scale would create meaningful differences from the 4-point version originally developed, the psychometric properties of the version utilized in the current study warrant further investigation. Investigation of the psychometric properties of the 7-item EROS may include studies investigating test-retest reliability, measures of internal consistency other than Cronbach's Alpha, and the predictive validity of the instrument. In addition, the current study was conducted in a nonclinical sample of convenience. Although it is important to establish the relationship of activation and environmental reward availability in a nondepressed sample in addition to a depressed sample, the clinical relevance of the current study may be limited.

Given the promising findings of the current study, the potential of behavioral activation as an intervention for depression (Mazzucchelli et al, 2009), and the theoretical ease of training BA, investigation of the current topic with increased methodological rigor is justified. Future research into a behavioral model of depression would benefit from measurement of environmental reward and behavioral activation in a treatment setting. Regular assessment of the constructs of interest as symptoms change will provide evidence clarifying the appropriate time to target activation and environmental rewards. Further, assessment in a clinical setting would provide evidence for the appropriate approach to assessing activation and environmental rewards impact on depression. Given the current evidence, a strong case can be made for both a moderating and mediation model.

Finally, further research is needed on what factors contribute to the availability of environmental rewards. One prospective area of investigation is the role of the social environment on depression. Early research on activation assessed the importance of interpersonal events at a topographical level (Youngren & Lewinsohn, 1980), but more recent investigations of BA have not incorporated the social environment. Relationships may play an important role in promoting and maintaining activation as they provide immediate consequences to actions. To that end, it may not be sufficient to assess social support (e.g., social isolation). Rather, it is important to understand the types of interactions that depressed individuals have with the individuals who make up their social network. The framework of intimacy provided by Cordova and Scott (2001) may be useful in guiding empirical investigation. This framework places an emphasis on how suppression of behavior vulnerable to interpersonal punishment can lead to a lack of interpersonal connection. Investigation into who suppresses and who expresses behavior vulnerable to interpersonal punishment might provide clarification on the relationship between loneliness and depression (Erzen & Gikrikci, 2018). In addition, because the EROS is a self-report measure rather than an objective measure of environmental reward, more complex behaviors such as verbal behavior (Skinner, 1957), relational framing (Hayes, Barnes-Holmes, & Roche, 2001), and other private behaviors such as remembering and attending (Donahoe & Palmer, 1994) may contribute to understanding why perceived environmental reward contributes to depression. Conceptualization of the role of these complex behaviors in depression is beyond the scope of the current study but has been presented by other researchers (e.g., Dougher, 2000).

To explore these components, researchers are encouraged to look at environmental reward and activation in a treatment setting as well as continue refining their understanding of environmental enrichment, including exploration of values, types of social interaction, and contact with the present moment while engaging in activities. Once empirically derived hypotheses are developed, a cross-sequential study can be conducted tracking meaningful variables in multiple cohorts over time. This study technique allows for assessment of changes in variables of interest while pragmatically allowing for researchers to present novel information at conferences and symposium. As variables separate themselves as meaningful, more resource intensive studies, such as randomized clinical trials, can be conducted to provide stronger evidence of the unique contribution of each component. Measuring activation and environmental reward availability can elucidate mechanisms of change in BA interventions, thus satisfying an important criterion in contemporary evaluations of evidence-based treatments (Tolin, Mckay, Forman, Klonsky, & Thombs, 2015).


Findings from the present study provide evidence for a model of depression grounded in behavioral theory. The findings provide support for the importance of both behavioral activation in addition to the availability of environmental reward in depression. Further, these findings provide support for the idea that this relationship is a moderated relationship. These findings may help refine conceptualization and implementation of treatments that include a behavioral activation component by identifying the appropriate time to help create environmental enrichment and when it is appropriate to focus on activation.

Availability of Data and Materials All data supporting these findings is available via the following link: 8f26c5e5d57a4078b8161059f32eae93

Funding The current study did not receive any funding

Compliance with Ethical Standards

Conflict of Interest All authors declare they have no conflict of interest All authors confirm that all the research meets ethical guidelines, including adherence to the legal requirements of the study country.

Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All research conducted received approval from the university institutional review board before being conducted.

Informed Consent Informed consent was obtained from all individual participants included in the study.


Allison, P. (2012). When can you safely ignore multicollinearity? Statistical Horizons. Retrieved from multicollinearity.

Armento, M. E., & Hopko, D. R. (2007). The environmental reward observation scale (EROS): Development, validity, and reliability. Behavior Therapy, 38(2), 107-119. 2006.05.003.

Beck, A. T. (1979). Cognitive therapy of depression. New York, NY: Guilford Press.

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-11. San Antonio, TX: Psychological Corporation.

Benson-Florez, G., Santiago-Rivera, A., & Nagy, G. (2017). Culturally adapted behavioral activation: A treatment approach for a Latino family. Clinical Case Studies, 16(1), 9-24. 1534650116668630.

Carvalho, J. P., Gawrysiak, M. J., Hellmuth, J. C., McNulty, J. K., Magidson, J. F., Lejuez, C. W., & Hopko, D. R. (2011). The Reward Probability Index: Design and validation of a scale measuring access to environmental reward. Behavior Therapy, 42(2), 249262.

Cordova, J. V., & Scott, R. L. (2001). Intimacy: A behavioral interpretation. The Behavior Analyst, 24(1), 75-86. BF03392020.

Dimidjian, S., Hollon, S. D., Dobson, K. S., Schmaling, K. B., Kohlenberg, R. J., Addis, M. E., ... Jacobson, N. S. (2006). Randomized trial of behavioral activation, cognitive therapy, and antidepressant medication in the acute treatment of adults with major depression. Journal of Consulting & Clinical Psychology, 74(4), 658-670.

Donahoe, J, W., & Palmer, D. C. (1994/2004). Learning and complex behavior. Richmond, VA: Ledgetop Publishing.

Dougher, M. J. (Ed.). (2000). Clinical behavior analysis. Reno, NV: Context Press.

Ekers, D., Webster, L., Van Straten, A., Cuijpers, P., Richards, D., & Gilbody, S. (2014). Behavioural activation for depression: An update of meta-analysis of effectiveness and sub-group analysis. PloS one, 9(6), 1-11.

Erzen, E., & Gikrikci, O. (2018). The effect of loneliness on depression: A meta-analysis. International Journal of Social Psychiatry, 64(5), 427-435.

Ferster, C. B. (1973). A functional analysis of depression. American Psychologist, 25(10), 857-870.

Hamilton, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery, & Psychiatry, 23(1), 56-62. jnnp.23.1.56.

Hayes, S. C., Barnes-Holmes, D., & Roche, B. E. (2001). Relational Frame Theory: A post-Skinnerian account of human language and cognition. New York, NY: Kluwer Academic/Plenum Publishers.

Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (1999). Acceptance and commitment therapy. New York, NY: Guilford Press.

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

Hill, R. M., Buitron, V., & Pettit, J. W. (2017). Unpacking response contingent positive reinforcement: Reward probability, but not environmental suppressors, prospectively predicts depressive symptoms via behavioral activation. Journal of Psychopathology & Behavioral Assessment, 39(3), 498-505. sltb. 12314.

Huang, C., & Chen, J. H. (2015). Meta-analysis of the factor structures of the Beck Depression Inventory-II. Assessment, 22(4), 459-472.

Jacobson, N. S., Dobson, K. S., Truax, P. A., Addis, M. E., Koemer, K., Gollan, J. K., ... Prince, S. E. (1996). A component analysis of cognitive-behavioral treatment for depression. Journal of Consulting & Clinical Psychology, 64(2), 295-304. 10.1037/0022-006X.64.2.295

James, G., Wittne, D., Hastie, T., & Tibshirani, R. (2013). Introduction to statistical learning with applications in R (7th ed.). New York, NY: Springer.

Kanter, J. W., Cautilli, J. D., Busch, A. M., & Baruch, D. E. (2005). Toward a comprehensive functional analysis of depressive behavior: Five environmental factors and a possible sixth and seventh. The Behavior Analyst Today, 6(1), 65. h0100055.

Kanter, J. W., Manos, R. C., Bowe, W. M., Baruch, D. E., Busch, A. M., & Rusch, L. C. (2010). What is behavioral activation? A review of the empirical literature. Clinical Psychology Review, 30(6), 608-620.

Kanter, J. W., Mulick, P. S., Busch, A. M., Berlin, K. S., & Martell, C. R. (2007). The Behavioral Activation for Depression Scale (BADS): Psychometric properties and factor structure. Journal of Psychopathology & Behavioral Assessment, 29(3), 191-202. 0862-006-9038-5.

Kanter, J. W., Rusch, L. C., Busch, A. M., & Sedivy, S. K. (2009). Validation of the Behavioral Activation for Depression Scale (BADS) in a community sample with elevated depressive symptoms. Journal of Psychopathology & Behavioral Assessment, 37(1), 36-42.

Kanter, J. W., Santiago-Rivera, A. L., Santos, M. M., Nagy, G., Lopez, M., Hurtado, G. D., & West, P. (2015). A randomized hybrid efficacy and effectiveness trial of behavioral activation for Latinos with depression. Behavior Therapy, 46(2), 177-192. https://doi.oig/10. 1016/j.beth.2014.09.011.

Lejuez, C. W., Hopko, D. R., Aciemo, R., Daughters, S. B., & Pagoto, S. (2011). Ten year revision of the Behavioral Activation Treatment for Depression (BATD-R): Revised treatment manual. Behavior Modification, 35(2), 111-161. 0145445510390929.

Lewinsohn, P. M. (1974). A behavioral approach to depression. In R. M. Friedman & M. M. Katz (Eds.), The psychology of depression: Contemporary theory and research. 157-174 New York, NY: Wiley

Lewinsohn, P. M., & Atwood, G. E. (1969). Depression: A clinical research approach. Psychotherapy, 6(3), 166-171. https://doi.oig/10. 1037/h0088744.

Lewinsohn, P. M., & Talkington, J. (1979). Studies on the measurement of unpleasant events and relations with depression. Applied Psychological Measurement, 3(1), 83-101. 014662167900300110.

MacPhillamy, D. J., & Lewinsohn, P. M. (1982). The pleasant events schedule: Studies on reliability, validity, and scale intercorrelation. Journal of Consulting & Clinical Psychology, 50(2), 363-380.

Manos, R. C., Kanter, J. W., & Busch, A. M. (2010). A critical review of assessment strategies to measure the behavioral activation model of depression. Clinical Psychology Review, 30(5), 547-561. https://doi. org/10.1016/j.cpr.2010.03.008.

Martell, C. R., Addis, M. E., & Jacobson, N. S. (2001). Depression in context: Strategies for guided action. New York, NY: Norton.

Mazzucchelli, T., Kane, R., & Rees, C. (2009). Behavioral activation treatments for depression in adults: A meta-analysis and review. Clinical Psychology: Science & Practice, 16(4), 383-411. https://

McCabe, C. J., Kim, D. S., & King, K. M. (2018). Improving present practices in the visual display of interactions. Advances in Methods & Practices in Psychological Science, 1(2), 147-165. https://doi. org/10.1177/2515245917746792.

Mir, G., Meer, S., Cottrell, D., McMillan, D., House, A., & Kanter, J. W. (2015). Adapted behavioural activation for the treatment of depression in Muslims. Journal of Affective Disorders, 180, 190-199. https://doi.Org/10.1016/j.jad.2015.03.060.

O'Brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 41, 673-690. https://doi.oig/ 10.1007/sl 1135-006-9018-6.

Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational & Behavioral Statistics, 31(4), 437-448. 10769986031004437.

R Core Team. (2017). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing Retrieved from

RStudio Team (2015). RStudio: Integrated development for R. RStudio. Retrieved from

Salinas, C., & Lozano, A. (2017). Mapping and recontextualizing the evolution of the term Latinx: An environmental scanning in higher education. Journal of Latinos & Education, 1-14. 1080/15348431.2017.1390464.

Skinner, B. F. (1957). Verbal behavior. New York, NY: Appleton-Century-Crofts.

Skinner, B. F. (1974). About behaviorism. New York, NY: Knopf.

Substance Abuse & Mental Health Services Administration. (2017). Key substance use and mental health indicators in the United States: Results from the 2016 National Survey on Drug Use and Health (HHS Publication No. SMA 17-5044, NSDUH Series H-52). Rockville, MD: Author Retrieved from data/.

Tolin, D. F., McKay, D., Forman, E. M., Klonsky, E. D., & Thombs, B. D. (2015). Empirically supported treatment: Recommendations for a new model. Clinical Psychology: Science & Practice, 22(4), 317-338.

World Health Organization. (2017). Depression and other common mental disorders: Global health estimates. Retrieved from https://www. health_estimates/en/.

Youngren, M. A., & Lewinsohn, P. M. (1980). The fimctional relation between depression and problematic interpersonal behavior. Journal of Abnormal Psychology, 89(3), 333-341. 0021-843X.89.3.333.

Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

[mail] Daniel W. M. Maitland

Daniel W. M. Maitland (1)(iD), Elizabeth C. Neilson (2), Emily A. Munoz (1), Alexis Ybanez (1), Amanda L. Murray (1)

(1) Department of Psychology, Texas A&M University-Corpus Christi, 6300 Ocean Drive, Bay Hall 3.51, Mail Stop 5827, Corpus Christi, TX 78412, USA

(2) University of Washington, Washington, DC, USA

Published online: 30 August 2019

Caption: Fig. 1 A Interaction of behavioral activation and environmental enrichment on depressive symptoms (Mean Model)

Caption: Fig. 2 Marginal effects plot of behavioral activation and environmental enrichment on depressive symptoms. Note: Shaded area indicates 95% confidence region for the marginal effect. The moderating effect of environmental enrichment on the behavioral activation-depression relation is not significant when environmental enrichment is greater than 2 standard deviations above the mean

Caption: Fig. 3 The effect of behavioral activation on depressive symptoms at 2 standard deviations below and above the mean of environmental enrichment. Note: PTCL = Percentile of data
Table 1 Sample (N = 374) descriptive statistics

Variable                      N     Percentage

White                         245      65.6%
African American/Black         23       6.1%
Native Hawaiian                 6       1.6%
Pacific Islander                1        .3%
Asian/Asian American           12       3.2%
American Indian                 6       1.6%
Multiracial                    33       8.8%
Prefer not to answer           47      12.6%
Latinx                        226      60.4%
Not Latinx                    144      38.5%
Prefer not to answer            4       1.1%
Freshman                       74      19.8%
Sophomore                      90      24.1%
Junior                        114      30.6%
Senior                         92      24.7%
Graduate student                3        .8%
Employment Status
Full time                      40     10.08%
Part time                     158      42.5%
Unemployment                  174      46.5%
Level of Depression
Minimal Depression (6-13)     207      54.6%
Mild Depression (14-19)        59      16.3%
Moderate Depression (20-28)    56      15.2%
Severe Depression (29-63)      45      12.1%

Table 2 Means, standard deviations, and zero-order correlations
for total score and mean score

                   BDI (sum)      BADS (sum)      EROS (sum)

BDI (sum)          --
BADS (sum)         -.742 **       --
EROS (sum)         -.701 **       .736 **         --
Mean (SD) (sum)    14.30(10.92)   90.23 (24.25)   46.59 (10.91)
Range (sum)        0-49           31-149          16-70
BDI (mean)         BDI (mean)     BADS (mean)     EROS (mean)
BADS (mean)        -.743 **       --
EROS (mean)        -.703 **       .739 **         --
Mean (SD) (mean)   .69 (.53)      3.62 (.949)     4.64(1.10)
Range (mean)       0-2.33         1.96-6.68       1.60-7.00

* p < .05, ** p < .01, *** p < .001

Table 3 Unstandardized and standardized coefficients for
regression models for sum and mean models

                       B (sum)    SE (sum)    [beta] (sum)

Step 1   BADS          -.22       .02         -.48 ***
         EROS          -.37       .05         -.36 ***
Step 2   BADS          -.22       .02         -.49 ***
         EROS          -.34       .05         -.34 ***
         BADS * EROS   .007       .001        .19 ***

                       B (mean)   SE (mean)   [beta] (mean)

Step 1   BADS          -.27       .027        -. 49 ***
         EROS          -.16       .023        -.34 ***
Step 2   BADS          -.28       .026        -.50 ***
         EROS          -.15       .022        -.32 ***
         BADS * EROS   .09        .014        .19 ***

         Partial              95% CI (sum)    VIF (sum)
         Correlation (sum)

Step 1   -.45                 -.26, -.17      2.20
         -.36                 -.46, -.26      2.20
Step 2   -.48                 -.27,-.18       2.20
         -.36                 -.44, -.25      2.22
         .30                  .005, .10       1.01

         Partial              95% CI (mean)   VIF (mean)
         Correlation (mean)

Step 1   -.47                 -.24, -.22      2.20
         -.34                 -.21,-.12       2.20
Step 2   -.47                 -.33, -.23      2.21
         -.34                 -.20, -.11      2.22
         .30                  .006,.ll        1.01

* p < .05, ** p < .01, *** p < .001
COPYRIGHT 2019 Springer
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Maitland, Daniel W.M.; Neilson, Elizabeth C.; Munoz, Emily A.; Ybanez, Alexis; Murray, Amanda L.
Publication:The Psychological Record
Article Type:Survey
Geographic Code:1U2NY
Date:Dec 1, 2019
Previous Article:Symbolic Generalization of Discriminative Functions in Accordance with a Five-Member Comparative Relational Network.
Next Article:Using the Implicit Relational Assessment Procedure (IRAP) as a Measure of Reaction to Perceived Failure and the Effects of a Defusion Intervention in...

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters