There is a growing need to understand the global impact of poverty on early brain and behavioural development, particularly with regard to key cognitive processes that emerge in early development. Although the impact of adversity on brain development can trap children in an intergenerational cycle of poverty, the massive potential for brain plasticity is also a source of hope: reliable, accessible, culturally agnostic methods to assess early brain development in low resource settings might be used to measure the impact of early adversity, identify infants for timely intervention and guide the development and monitor the effectiveness of early interventions. Visual working memory (VWM) is an early marker of cognitive capacity that has been assessed reliably in early infancy and is predictive of later academic achievement in Western countries. Here, we localized the functional brain networks that underlie VWM in early development in rural India using a portable neuroimaging system, and we assessed the impact of adversity on these brain networks. We recorded functional brain activity as young children aged 4–48 months performed a VWM task. Brain imaging results revealed localized activation in the frontal cortex, replicating findings from a Midwestern US sample. Critically, children from families with low maternal education and income showed weaker brain activity and poorer distractor suppression in canonical working memory areas in the left frontal cortex. Implications of this work are far‐reaching: it is now cost‐effective to localize functional brain networks in early development in low‐resource settings, paving the way for novel intervention and assessment methods.

In the current study, we extend a previous methodological pipeline by adding a novel image reconstruction approach to move functional near-infrared (fNIRS) signals from channel-space on the surface of the head to voxel-space within the brain volume. We validate this methodology by comparing voxel-wise fNIRS results to functional magnetic resonance imaging (fMRI) results from a visual working memory (VWM) task using two approaches. In the first approach, significant voxel-wise correlations were observed between fNIRS and fMRI measures for all experimental conditions across brain regions in the fronto-parieto-temporal cortices. In the second approach, we conducted separate multi-factorial ANOVAs on fNIRS and fMRI measures and then examined the correspondence between main and interaction effects within common regions of interest. Both fMRI and fNIRS showed similar trends in activation within the VWM network when the number of items held in working memory increases. These results validate the image-based fNIRS approach.

Working memory is a central cognitive system that plays a key role in development, with working memory capacity and speed of processing increasing as children move from infancy through adolescence. Here, I focus on two questions: What neural processes underlie working memory, and how do these processes change over development? Answers to these questions lie in computer simulations of neural-network models that shed light on how development happens. These models open up new avenues for optimizing clinical interventions aimed at boosting the working memory abilities of at-risk infants.

In recent years, there has been a rise in the usage of functional near infra-red spectroscopy (fNIRS) in neuroimaging studies.^{1 – 5} fNIRS is a safe, non-invasive method of neuroimaging that measures the concentration of oxygenated and deoxygenated hemoglobin. fNIRS has become an important neuroimaging tool for developmental neuroscience and other fields where participant characteristics, naturalistic study paradigms, or costs make functional MRI difficult to use, such as in low-resource settings.^{6 – 8}

Infants, children and adults have been shown to track co-occurrence across ambiguous naming situations to infer the referents of new words. The extensive literature on this cross-situational word learning (CSWL) ability has produced support for two theoretical accounts—associative learning (AL) and hypothesis testing (HT)—but no comprehensive model of the behaviour. We propose WOLVES, an implementation-level account of CSWL grounded in real-time psychological processes of memory and attention that explicitly models the dynamics of looking at a moment-to-moment scale and learning across trials. We use WOLVES to capture data from 12 studies of CSWL with adults and children, thereby providing a comprehensive account of data purported to support both AL and HT accounts. Direct model comparison shows that WOLVES performs well relative to two competitor models. In particular, WOLVES captures more data than the competitor models (132 vs. 69 data values) and fits the data better than the competitor models (e.g., lower percent error scores for 12 of 17 conditions). Moreover, WOLVES generalizes more accurately to three ‘held-out’ experiments, although a model by Kachergis and colleagues (2012) fares better on another metric of generalization (AIC/BIC). Critically, we offer the first developmental account of CSWL, providing insights into how memory processes change from infancy through adulthood. WOLVES shows that visual exploration and selective attention in CSWL are both dependent on and indicative of learning within a task-specific context. Further, learning is driven by real-time synchrony of words and gaze and constrained by memory processes over multiple timescales.