I am having trouble finding the social psychology theory that relates to this example.

Researchers are curious whether detectives might influence forensic scientists by providing information about unrelated evidence that suggests a defendant's guilt. In a between subjects design, experimenters informed fingerprint analysts either that a positive witness ID had been made of the suspect or no positive ID had been made. Results indicated that fingerprint analysts reported more suspect matches when they had been told that a witness had positively identified the suspect.