Social context increases infants’ attentiveness – but why?

I’m a huge fan of the work of Linda Smith and Chen Yu – but this recent article from them wound me up a little. It reports that, when a social partner pays attention to an object, this increases the attention that the other social pays to that object, too. This, Chen and Linda interpreted as suggesting that social context leads to immediate increases in sustained attention.

This wound me up because sustained attention is, in the adult and child literature at least, widely considered a top-down, or executive, attention mechanism. In the infant literature the picture is much murkier (see here), with shorter looking times in young (<8 month) infants thought to associate with better long-term outcomes, and the relationship putatively flipping some time around the 12-month age boundary (see here, and here). But Chen and Linda didn’t mention these infant findings - they said that sustained attention is ‘generally believed to be the developmental product of increasing self-regulatory and endogenous (i.e., internal, top-down, voluntary) control over one’s attention and cognitive systems”.

Chen and Linda’s findings suggested that sustained attention was increased by social factors. So were they suggesting that social context leads to immediate increases in infants’ endogenous attention control?

In a Developmental Science paper that has just come out, we looked at this. First, we set out to replicate their finding, in a slightly different way. We asked parents and children to play in two contexts – either alone, with an identical but parallel set of toys (which we called Solo Play), or together (which we called Joint Play). We videoed the play, and micro-coded both partners’ looking patterns afterwards. Sure enough, we found that infants’ attention durations towards the objects were higher in the Joint Play condition – consistent with what Chen and Linda had found.

Next, we asked – why was this? Was it because social context led to immediate increases in infants’ endogenous attention control? Or was it because, in social settings, parents manipulate the object by making it more exogenously salient and attention-eliciting (e.g. by moving it around)? If this latter hypothesis were true then social context would increase infants attentiveness towards the play objects primarily via ‘bottom-up’ attention mechanisms – a completely different route to what Chen and Linda had (seemingly) claimed.

We couldn’t answer that question directly in the paper – for more, see here – but we provided some data that we argued was more consistent with the second hypothesis. First, we found that infants’ rate of change of attentiveness was faster during Joint Play than Solo Play, sugesting that endogenous attention factors, such as attentional inertia, influence looking behaviour less during Joint Play. Second, we found that adults’ attention forwards-predicted infants’ subsequent attention more than vice versa. Finally, we found that mutual gaze did not directly facilitate infant attentiveness.

These findings are relatively by-the-by – good research into parent-child interactions has already shown how parents use multi-modal cues to maintain infants’ attentiveness. But it’s important, we think, because of other research that suggests that infants who spend more time in joint engagement with parents during play show better subsequent visual attention control over longer time-frames. But how does scaffolding, if it operates primarily via exogenous attention cues, lead to better endogenous long-term attention control? Now that's a tough question to answer...!