•Evidence that UAE is effective for extraction of anthocyanins from haskap berries.•Significant variables can be screened by Plackett–Burman design.•The extraction of anthocyanins was optimized by RSM using Box–Behnken design.•Proximity of predicted and practical responses indicate the adequacy of the model. Haskap berries (Lonicera caerulea L.) are a rich source of bioactive molecules. As such, the extraction of anthocyanins is important for the development of many value-added products and functional food ingredients. In this paper, the ultrasound-assisted extraction (UAE) of anthocyanins from haskap berries was investigated. Significant independent variables were screened and optimized using Plackett–Burman (PB) and Box–Behnken (BB) designs, respectively. The mathematical model showed a high coefficient of determination (R2=0.9396) and the optimum conditions for the extraction were as follows: liquid/solid ratio 25:1 (mL/g), solvent composition of 80% ethanol, addition of 0.5% formic acid, ultrasound bath temperature of 35°C for 20min. Under these conditions, the total anthocyanin content of 22.73mg cyaniding 3-glucoside equivalents (C3G)/g dry weight (DW) was consistent with the predicted response of 22.45mg C3G/g DW from the model (mean error of 1.28%). Five anthocyanins were identified in the optimized extract, namely cyanidin 3,5-diglucoside, cyanidin 3-glucoside, cyanidin 3-rutinoside, pelargonidin 3-glucoside, and peonidin 3-glucoside. Thus, UAE is a suitable technique for the extraction of anthocyanins from haskap berries.