Transference in social cognition: Persistence and exacerbation of significant-other-based inferences over time

A social-cognitive model of transference - defined as the activation and application of a mental representation of a significant other to a new person - has been verified experimentally in terms of relevant inferences and memory about the new person (e.g., Andersen and Cole, 1990; Andersen, Glassman, Chen, and Cole, 1995). The model suggests that transference should persist and increase over time, indicating that the phenomenon is not fleeting or self-correcting, and is therefore of clinical importance. In two within- subject experiments, participants learned about four fictional people, one of whom resembled their own significant other. They then completed a recognition-memory test. In Study 1, the test was administered both immediately after learning about the new people, and again 2 to 3 weeks afterward As predicted, greater confidence in having learned representation- consistent attributes that had not been presented in the learning task occurred in the significant-other condition relative to the control conditions - both immediately and after the delay, with the effect increasing over time. The potential artifact of the first memory test vis-a-vis the second was ruled out in Study 2, which showed the persistence effect using a test administered only once, 2 to 3 weeks after the learning task. Persistence and exacerbation in the effect have theoretical and clinical implications, as does the general notion that transference occurs in everyday social perception on the basis of significant-other representations.