Social Thinking®, Pseudoscientific, Not Empirically Supported, and Non-Evidence Based: a Reply to Crooke and Winner
Social Thinking® lacks the data we require; choose interventions with proven social skill outcomes.
01Research in Context
What this study did
Leaf and colleagues wrote a position paper. They asked, 'Does Social Thinking® meet behavior-analytic standards?'
They checked the published studies on Social Thinking®. They looked for single-case design, measurement of real social behavior, and peer review.
What they found
The authors found almost no data. Most Social Thinking® books lack single-case graphs or group trials.
They conclude the package is 'pseudoscientific' and 'non-evidence-based' for clients with autism.
How this fits with other research
Peters et al. (2018) backs them up. That review also says stand-alone perspective-taking drills do not help kids with autism make friends.
Begeer et al. (2015) seems to disagree. Their RCT showed short ToM lessons improved false-belief scores. But the gains stayed on the test and did not spread to recess or home life. The papers clash only on the surface; both agree narrow drill success is not social success.
Bahry et al. (2023) extends the warning. They say we must pick goals that matter in adulthood, not just goals that sound social.
Why it matters
If you write treatment plans, this paper is your shield. You can show teachers, parents, or funders why you will not use Social Thinking® worksheets. Replace them with scripts, peer modeling, or BST that already have single-case evidence. You stay ethical, and the client gets skills that actually work outside the clinic.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Audit your current social goals—swap any Social Thinking® worksheet for a peer-mediated BST session with data sheets.
02At a glance
03Original abstract
J. B. Leaf et al. (Behavior Analysis in Practice, 9, 152–157, 2016) wrote a commentary on social thinking (ST), an intervention commonly implemented for individuals diagnosed with autism spectrum disorder (ASD). The authors described what constitutes scientific, pseudoscientific, and antiscientific evidence and contended that ST aligns with the definition of pseudoscience and, to date, is not empirically supported or evidence based. Crooke and Winner (Behavior Analysis in Practice, 9, 403–408, 2016) responded, arguing that ST meets their definition of an evidence-based practice and identifying purported misconceptions and inaccuracies described by J. B. Leaf et al. In the current article, the authors clarify the original arguments, critically evaluate Crooke and Winner’s definition of what constitutes evidence-based practice, further evaluate the research on ST, discuss issues regarding how ST is conceptualized, and express concerns about the endorsement and use of an eclectic approach to treating ASD. As this response was written by behavior analysts, it specifically addresses the conceptual consistency of this approach from a behavior–analytic worldview.
Behavior Analysis in Practice, 2018 · doi:10.1007/s40617-018-0241-0