Does social collaboration benefit face-matching accuracy over simply fusing individuals' responses?