*3.3. Comparison between Children with Early Bilateral Cochlear Implantation and Relatively Late Sequential Bilateral Cochlear Implantation*

To study the effect of age at implantation on development of sound localization accuracy, data were further compared to results from children with relatively late sequential bilateral cochlear implantation (median age first CI = 1.9 years, median age second CI = 4.1 years, *n* = 66) [23]. These children, implanted and tested at the same tertiary referral center as the subjects in the present study, were assessed at a median age of 5.6 years (range = 2.8–17.3 years), i.e., they were substantially older than the children in the present study. Despite methodological and procedural differences (i.e., number of sound-sources, spatial range and resolution of the test, spectral and temporal characteristics of the auditory stimulus, and quantification of behavioral responses), a striking resemblance in development of localization between early (this study) and late bilateral implantation existed (Figure 4, panel B). The rate of development was identical, whereas intercepts differed slightly (late cochlear implants: slope = 0.21/year, intercept = 0.79; early cochlear implants: slope = 0.21/year, intercept = 0.82).

#### *3.4. Reliability of Sound Localization Measurements*

The 95% C.I. of the test–retest differences (−0.098 to 0.037) included 0, reflecting that no significant learning effect occurred. The 95% C.I. for the EI for a single measurement was ±0.11 (*n* = 42). The test–retest differences did not depend on the number of obtained responses during a test (r = 0.13, *p* = 0.40).

## **4. Discussion**

We found that infants with bilateral severe-to-profound congenital hearing loss develop horizontal sound localization abilities after bilateral cochlear implantation. When contrasting data from the current study with previous data from infants with normal hearing, we found that the developmental rates in these groups were similar. While it seems unlikely that localization will reach the same accuracy as in normal hearing based on CI studies in adults [31], the rate of improvement emphasizes the need of early provision of hearing in both ears for children with severe-to-profound hearing loss to allow development of spatial hearing near ages for which development normally occurs. In addition to being a safety matter in for example traffic, adults with hearing loss report that difficulties in localization of sounds are associated with loss of concentration, confusion of sounds, and a wish to escape settings in which this occurs [32]. Additionally, accurate localization is likely to improve communication since audiovisual cues are important for speech understanding when hearing loss is present [33]. Less is known about how impaired sound localization during infancy affects learning and interaction in daily life and should be targeted in future research. For children with unilateral hearing loss, a condition that typically is associated with impaired localization [34], it is 10 times more common having to repeat at least one year in school [35].

When contrasting the current data with previous data from children with relatively late cochlear implantation, we found a striking resemblance between infants and children in early school-age in the rate of development following implantation. This demonstrates that a sensitive period for human spatial hearing is not restricted to early development, corroborating findings in adult humans [6] and ferrets [16] who adapt their behavior to altered spatial cues. Results differ from barn owls [36] and mice [37], for which age limits sensitive periods for development of sound localization or the binaural cues it is based on. In children implanted bilaterally after 5 years of age, localization performance is poor one year after implantation [38], but data on long-term localization performance for late bilateral implantation is unknown. It is noteworthy that the high similarity in the rate of development and between-subject variability of sound localization abilities found in the present study between younger and older children with cochlear implants occurred despite methodological differences in how localization accuracy was measured. Infants' responses in the present study were obtained through eye-gaze, whereas older children with implants [23] pointed at or verbally indicated the perceived sound-source azimuth. In addition, children in the present study listened to a continuous sound, whereas older children listened to sounds of relatively short duration. For both groups, between-subject variability in localization accuracy was high and time since activation of BiCI explained

the same amount of variance in localization accuracy (25%). The underlying causes for variability in binaural hearing in individuals with implants have been targeted in recent years [39–43], revealing etiology of the hearing loss, duration of hearing loss, and surgical procedures and subsequent bilateral fitting of sound processors as variables that may affect results. Importantly, while current and previous data presented together here show that the developmental rate of localization accuracy is comparable for children with normal hearing and cochlear implants, localization performance after prolonged cochlear implant use does not reach that of individuals with normal hearing [17,44,45]. One reason for less accurate localization despite many years of cochlear implant stimulation is that thresholds for important cues underlying accurate localization (interaural level and time differences) typically are substantially worse for listeners with cochlear implants compared to normal hearing [46], owing to technical and surgical limitations (see, e.g., ref. [47] for a discussion). Future studies including long-term follow up of children who received bilateral cochlear implants as infants may reveal if localization performance plateaus at higher accuracy than later implanted children, and if early localization abilities have implications for, e.g., learning, language and social interaction. Factors of interest in such future studies should be to determine underlying causes for between-subject variability through genetic testing (≈50% of congenital sensorineural hearing loss are genetic in origin [48,49]) and radiological investigation of bilateral cochlear implant electrode placement to assess the impact of interaural frequency mismatch which negatively affects binaural hearing [39].

A limitation of the comparison between data collected in the current study and data from children with NH from previous work is that previous data are cross-sectional and based on a relatively small sample. However, the data from children with NH should be representative given previously performed analyses [28] of localization accuracy in larger samples of children with NH aged 0.5 to 1.5 years (*n* = 80) showing a rate of improvement similar to what was found in our smaller sample [28,50].

Data presented here suggest an intrinsic mechanism for the development of horizontal sound localization abilities. Our study improves on previous work on spatial hearing in children with cochlear implants [17,23,51,52] due to its inclusion of children at a very young age and its longitudinal follow-up before school-age. As long as cochlear implantation may be performed safely in infants, our findings suggest that implantation should occur as early as possible to allow development of spatial hearing near ages for which development normally occurs.

**Author Contributions:** Conceptualization, F.A. and E.B.; methodology, F.A. and E.B.; software, F.A.; validation, F.A., E.K. and E.B.; formal analysis, F.A. and E.B.; investigation, F.A.; resources, F.A.; data curation, F.A.; writing—original draft preparation, F.A.; writing—review and editing, F.A., E.K. and E.B.; visualization, F.A. and E.B.; supervision, F.A. and E.B.; project administration, F.A. and E.K.; funding acquisition, F.A., E.K. and E.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Tysta Skolan Foundation (FB16-0023) and the Foundation for promotion and development of clinical research at Karolinska Institutet. The project was also supported by the regional agreement on medical training and clinical research (ALF) between Stockholm County Council and Karolinska Institutet, and the foundation of the Swedish Order of Freemasons (Konung Gustaf VI Adolfs frimurarefond, 2022).

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki, and approved by the regional ethical review board in Stockholm, Sweden (permit number 2012/189-31/3 and 2013/2248-32) for studies involving humans.

**Informed Consent Statement:** Parental informed consent was obtained for all subjects involved in the study. Written informed consent has been obtained from the parents to the children to publish this paper.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ethical, legal and privacy issues.

**Acknowledgments:** We wish to extend our appreciation to the children and parents that participated in this study. The authors are grateful for the continued support of Scientific Center for Advanced Pediatric Audiology at Karolinska Institute, to Åke Olofsson for technical assistance, and to Mats Ceder for illustrating the setup used for sound localization measurements in Figure 1.

**Conflicts of Interest:** The authors declare no conflict of interest.
