Abstract
<p>Relational Frame Theory (RFT) conceptualises perspective taking and related constructs of empathy, theory of mind, and false-belief understanding in terms of deictic relational responding. RFT's Perspective Taking Protocol was originally developed to assess and train these relational repertoires, and increasingly became used to measure individual differences in perspective taking in adult and clinical samples. This slippage was recently raised as a concern regarding its appropriate use: the latter presupposes that PTP scores are reliable and valid measure of a perspective-taking trait, yet these properties have rarely been tested. This puts it at risk of jingle fallacy: the assumption similarly named measures measure similar things. We administered a 9-item PTP to 134 adults online, retesting 98 participants after a 10-day interval, alongside established self-report and behavioural measures of perspective taking, empathy, and theory of mind. The PTP showed poor internal consistency (α = .40, 95% CI [.24, .54]) and test–retest reliability indistinguishable from zero (ICC2,1 = .08, 95% CI [−.12, .27]); its trial-type subscales did not cohere (subscale correlations from r = -.08 to .22); and it was essentially uncorrelated with existing measures even after disattenuating for unreliability. A meta-analysis demonstrated that the PTP is on average unrelated to other measures (r = .01, disattenuated r = .03) whereas those measures are generally associated with one another (r = .28, disattenuated r = .35). Results question whether the PTP should be used as a measure of individual differences in perspective taking; naming it a perspective-taking measure may represent a jingle fallacy.</p>