“…Results found here agree with those of Rose et al. () who found no stability for processing speed or overall percent novel using the serial task when infants were tested at more than one age, and they agree with results from a variety of preference tasks reporting poor reliability (e.g., Andersson, ; Bornstein, Brown, & Slater, ; Colombo et al., ; Lavoie & Desrochers, ; Rose et al., ; Tashihsazan, Nettelbeck, & Kirby, ; Thompson, Fagan, & Fulker, ). Moderate test–retest correlations were nonetheless reported when novelty preferences were averaged across trials during traditional habituation and paired‐comparison studies (e.g., Colombo et al., ; Rose & Feldman, ; for reviews, see Bornstein & Sigman, ; McCall & Carriger, ; Rose et al., ; Slater, ).…”