2a. Students should be trained to understand and appreciate how heuristics and biases, particularly confirmation bias, will limit the accuracy of their judgments (Arkes, 1991; Garb, 1998).
This information can assist the student in learning how to recognize when their decisions are guided by biases, how to correct this, and how to use valid psychological measures to continually evaluate/double-check their clinical impressions (Croskerry, 2002, 2003; Meehl, 1954).
2b. Students should be trained and supervised in the application of scientific thinking to practice, in particular hypothesis testing, data collection, Bayesian decision-making, etc. within a clinical context (Dixon, et al., 2009; Lueger, 2002; Straus, et al., 2011).
As mentioned above, they should learn how to monitor progress to make sure that treatment is helping and not having unintended consequences (Jacobson & Truax, 1991; Lambert & Brown, 1996; Lambert, et al., 2001; Powsner & Tufte, 1994). They should also be trained to be keenly aware of any ethical concerns that might impact their clinical or clinical research activities (Hoagwood & Cavaleri, 2010).
2c. Students should be trained in how to proceed clinically in the absence of highly relevant scientific knowledge (the “rigor versus relevance” dilemma, where clinical practice involves some individual cases that will not be well represented in rigorous research) (Schon, 1983).
Students should learn how to use guiding principles and generalizations from evidence to shape practice, rather than switching to unstructured, unreflective improvisation in the absence of strong evidence (Castonguay & Beutler, 2006; Norcross, Hogan, & Koocher, 2008; Stricker & Gold, 1996). For example, exposure is a technique with empirical support for a wide range of anxiety and fear conditions. Even if a client’s anxiety problems do not meet diagnostic criteria for a specific anxiety disorder with a corresponding empirically-supported treatment, critical thinking about the scientific literature may lead a therapist to identify exposure as a research-informed treatment option (Woody & Ollendick, 2006). As another example, basic and psychotherapy research converge in suggesting that when treating clients with depressive symptomatology, students should be trained to use interventions that foster the awareness, acceptance and regulation of emotion. Specifically, while psychopathology research has demonstrated the positive impact of emotional disclosure (and the negative impact of suppression of emotion), psychotherapy studies have found positive relationships between emotional deepening and outcome not only in experiential therapy (which emphasize this principle of change) but, as mentioned above, in cognitive therapy for depression (see LeMoulth, Castonguay, Joorman, & McAleavey, in press). Interventions that focus on emotional regulation would seem appropriate even if the client does not meet diagnostic criteria for major depression. Although existing evidence-based interventions have been identified according to diagnostic categories, treatment need not be organized around a diagnosis, but should have a formulation that guides choice of strategies and which can be measured to show progress (Crits-Christoph, 1998; Luborsky, 1984; Sanderson & McGinn, 1997; Stricker & Gold, 1996). These suggestions not only reflect the importance of using critical thinking in the conduct of psychotherapy but are also consistent with the recommendation that clinicians be guided by empirically-based principles of change (e.g., Follette & Greenberg, 2006; Woody & Ollendick, 2006, Norcross, et al., 2008; Spring, 2007) to help them adapt (and/or enhance the impact of) empirically-supported treatments for individual clients (Stricker & Gold, 1996).
2d. Students should learn to be critical consumers and producers of the research literature, recognizing common sources of bias in assessment (Bossuyt, et al., 2003; Campbell & Fiske, 1959; Jaeschke, Guyatt, & Sackett, 1994; Meehl, 1954) and treatment studies (Chambless & Ollendick, 2000; Moher, et al., 2001; Silverman, 1998), and both the strengths and limitations of different study designs for yielding knowledge that generalizes to clinically relevant and diverse populations.
Students should be trained to be attuned to how ethical issues may operate in clinical research studies (e.g., selection of control groups, inclusion/exclusion criteria, informed consent; Hoagwood & Cavaleri, 2010). In addition, students should learn how biases and heuristics, including confirmation bias can affect their perspective in research contexts, constrain the nature of hypotheses they consider, and affect their interpretation of both previous research and their own study data.