**4. Discussion**

The current paper focuses on early abilities that might be supporting language development in DS and TD 5–7-months old infants. In particular, we assessed infants' early visual attention, audiovisual speech processing, and communication skills. We will discuss each of the assessed measures and their implications for language development, particularly for DS infants. First, we observed that DS infants are slower in orienting their visual attention to stimuli in comparison to TD peers. This means that DS infants need more time to start attending to salient visual cues in their environment, here a flashing red light. This result is in line with previous studies on impaired visual attention in DS toddlers and children, especially in disengaging their visual attention [34–36]. However, our results differ from Landry and Bryson's study [35] where DS preschool children were similar in the visual orientation task to TD children matched in mental age with the DS group. There are at least two explanations for these between-studies differences. First, we tested a much younger population than in Landry and Bryson [35] and it is possible that by the preschool age DS children do improve their visual orientation attention. Second, there are important methodological differences between studies. Note that we compared DS and TD infants that were matched in their chronological, rather than in their mental age. Considering that we were interests in assessing DS infants between 5–7 months of age it would be difficult, if not impossible, to match groups in their mental age. Moreover, our task required infants to turn and orient their head to lateral/central position to flashing lights, whereas Landry and Bryson [35] used a set up with a central monitor and two lateral monitors placed in front of the child. It is therefore possible that our task was particularly challenging for the Down syndrome group. Nevertheless, our study is one of the first studies demonstrating that in first half of the first year of life, DS infants are impaired in orienting their visual attention to salient stimuli.

The second component we assessed was infants' attentional pattern during audiovisual speech processing. We presented an animated character that waived and talked at the infant. We were particularly interested in examining what visual speech/communicative cues infants attend to at this age early age. We observed that the two groups demonstrated certain similarities in their looking pattern: both groups do attend less to the waving arm at this age than the face and the background, suggesting that at this age the waiving gesture is not a particular salient communicative cue. Further, when attending to the face, both groups look more at the eyes than the mouth, in line with many recent studies done at similar ages (e.g., [41–43]). However, we observed a striking difference between the groups: TD infants attend more to the face than to the background, whereas DS infants attend similarly (~40 percent of their looking time to the screen) to the background and the face. This suggests that for DS infants the face is not a salient cue, at least not more salient than the background. Further implying that when observing a scene, DS infants attend equally to social/speech cues and to other cues that are not relevant for communication. This finding is in line with other studies that observed impairment in face and audiovisual speech processing in DS infants and toddlers [31–33]. Considering that we observed that DS infants are also slower to orient their visual attention, we can propose that such impairment affects their ability to orient to and fixate salient audiovisual speech cues (e.g., the face). Therefore, early interventions that are based on improving communication abilities have to take into account that DS attention to communicative cues is impaired. Moreover, we detected such impairment already in the first half of the first year of life, suggesting that interventions could target the ability to detect visual speech cues, even before interventions focusing on improving joint attention take place. We could also speculate that improving visual orientation might improve their ability to detect visual speech/communication cues and further research should address this possibility. It is important to note, however, that we have assessed audiovisual speech processing using an animated character, that certainly differs from a human face. In particular, the richness of the interplay between the acoustic signal, articulatory movements and facial expressions is reduced in animated characters, in comparison to a human face. Interestingly, even with the animated character,

DS and TD infants attend more to the eyes than the mouth, similar to previous studies assessing attention to a human face (e.g., [41–43]). Moreover, we also found that DS infants differ from TD infants in how they attend to communicative cues in an animated character talking face. It might be possible that for DS infants a human talking face would elicit greater attention, and future research should assess DS attention to a talking human face in relation to attractive background.

Finally, we observed important across-group differences regarding the relation between attending to audiovisual communicative cues and communicative development. First, we observed that attention to the eyes in the TD group relates to concurrent gesture and communication skills (and a trend for a positive correlation between looking to the arm and gesture skills was also found). Note that the items in the gesture and communication scales are mostly tackling non-verbal communication skills. For instance, assessing whether the child is pointing, waiving, asking for attention when a caregiver is not providing it, etc. So, we observed that in ~6-month-old typically developing infants, attending to relevant audiovisual cues, particularly to the eyes and the arm, supports early communication skills. This result is in line with previous studies on the importance of audiovisual cues for communication skills in typical development (e.g., [44]). However, in DS infants, we have not observed patterns supporting a relationship between performance in the audiovisual task and communication skills. We further inspected these results and observed that all DS infants have value 0 on the gesture score, meaning that gesturing in 5–7-month-old DS infants did not emerge yet. Considering that we have observed that DS infants attend much less to relevant audiovisual cues than TD infants, it is possible that less attention to audiovisual cues hinders their communicative skills. Alternatively, their poor communicative skills could drive away their attention from relevant audiovisual speech/communication cues. Either way, we observed that in the first 6 months of life, DS infants' link between audiovisual communicative attention and communication skills is not yet established. Further research is needed to address what is the relation between early attention to communicative cues and later communicative development.

A limitation of our study is certainly the small sample size, and future work should include larger samples. However, it should be noted that the population of DS infants is far less than that TD infants. According to the report from the National health institute [45] in the period from 2008–2017, in average, ~20 Down syndrome infants were born in Portugal per year. Thus, we assessed 35% of the population in a given year. An additional limitation of the current study is that our DS group was chronologically age matched with the TD group, and not mental age matched. Therefore, it is possible that the findings might change if groups were matched by mental age, in particular if an older group of DS infants was considered instead. However, that would leave our goal of investigating very early attention and audiovisual abilities in DS infants unaddressed, as well as of contributing to understand how DS communicative abilities develop from an early age. In future work, we plan to look at older DS infants to examine how these abilities develop. Nevertheless, based on the current findings we could speculate that undeveloped attentional orientation skills hinder DS infants' orientation to speech/communication cues. Therefore, future research should also explore intervention strategies that would focus on improving orientating attention, but specifically to audiovisual speech/communication cues, i.e., the face.
