This study describes and tests an empirical-based theoretical model of rapport in an investigative interview context. Essential in this study is whether rapport, operationalized as the humanitarian interview, in two interviews with a six-month retention interval, had any causal effects on the respective memory performance of 146 and 127 interviewees. Independent-samples t‐tests revealed, on both occasions, that a humanitarian rapport interview led to a larger amount of reported information altogether, with more central and peripheral information, than a dominant non-rapport interview did. Regardless of the interview approach, mixed between-within analysis of variance showed a substantially larger amount of reported information in the first interview than the second. The amount of false information reported in both interviews was statistically invariable, regardless of interviewing style.