*2.2. Subjects*

Twenty children were included in the study at a median age of 0.87 years (0.58–2.53 years, 8 females) (Table 1). Parental informed consent was obtained for all children. Children were implanted bilaterally at a median age of 0.58 years (0.42–2.3 years) with devices from Cochlear (Cochlear Corporation, Sydney, Australia) or Med-El (Med-El GmbH, Innsbruck, Austria).

Thirteen children who met the inclusion criteria were not asked to participate due to limited time during regular clinical follow-up. Another two children declined participation. These 15 children were implanted at the same median age as the included subjects.


**Table 1.** Background data on the children who participated in the study. Children are sorted in ascending age order. Two of twenty included children did not cooperate to sound localization testing and are not shown.

CI522, CI512, CI532, and CI612; cochlear implants manufactured by Cochlear Corporation, Sydney, Australia. Synchrony 2 Flex28; cochlear implant manufactured by Med-El GmbH, Innsbruck, Austria. CP1000; sound processor manufactured by Cochlear Corporation, Sydney, Australia. Sonnet 2; sound processor manufactured by Med-El GmbH, Innsbruck, Austria. M; Male. F; Female. cCMV; congenital cytomegalovirus.

#### *2.3. Setup, Stimulus and Test Procedure*

The setup, stimulus, test procedure and acquisition of behavioral responses is described in detail previously [28]. Children were seated in the lap of a parent in front of 12 active loudspeakers (ARGON 7340A; Argon Audio, Sweden) spanning a 110-degree arc in the frontal horizontal plane (Figure 1) in an audiometric test room. Loudspeakers were at ear level and spaced 10 degrees, resulting in loudspeaker positions at ±55, ±45, ±35, ±25, ±15, and ±5 degrees azimuth with respect to the subject. A 7-inch thin film transistor (TFT) display was mounted below each loudspeaker, resulting in 12 loudspeaker/display (LD)-pairs. An eye-tracking system (Smart Eye Pro; Smart Eye AB, Gothenburg, Sweden) was used for objective positioning of children's pupil positions relative to the LD-pairs.

**Figure 1.** Experimental setup for determination of horizontal sound localization accuracy in infants and young children listening through bilateral cochlear implants. The left panel illustrates the position of the child relative to an array of loudspeaker/display-pairs. Loudspeakers were covered in black cloth to attract the child's gaze to the visual displays. A continuous auditory-visual stimulus was presented from a loudspeaker/display-pair and randomly shifted in azimuth. Simultaneously with an azimuthal shift, the visual part of the stimulus was stopped for 1.6 s and eye-gaze patterns in response to the auditory stimulus were recorded before the visual part of the stimulus returned. The right panel illustrates the implanted and external parts of a cochlear implant system. An array of electrode contacts resides in the cochlea, stimulating the auditory nerve. The electronics of the cochlear implant are driven by an external sound processor behind the ear. Illustration by Mats Ceder.

A sound localization test consisted of 24 azimuthal shifts of an ongoing auditoryvisual stimulus (a colorful cartoon movie playing a continuous melody with a long-term frequency spectrum similar to female speech) presented at 63 dBA. In each azimuthal shift, the sound was changed to another randomly assigned loudspeaker on average every 7th second (5–9 s) with a simultaneous stop of the visual stimulus. The visual stimulus was reintroduced on the visual display corresponding to the sounding loudspeaker 1.6 s after the azimuthal sound shift. The procedure allowed acquisition of gaze behavior during 1.6 s in response to a spatial change of the sound. A test lasted ≈3 min.

Localization accuracy was quantified by an Error Index [29,30]. An EI = 0 corresponds to perfect performance. An EI = 1 corresponds to average random performance. A Monte Carlo simulation showed that the 95% confidence interval (C.I.) for random performance using the current procedure was [0.72, 1.28].

Children were not given any instructions before or during testing. The parent having the child on their knee was instructed to remain seated and unmoving and to not talk to the child.
