Dataset Information

Eight-Month-Old Infants Meta-Learn by Downweighting Irrelevant Evidence.

ABSTRACT: Infants learn to navigate the complexity of the physical and social world at an outstanding pace, but how they accomplish this learning is still largely unknown. Recent advances in human and artificial intelligence research propose that a key feature to achieving quick and efficient learning is meta-learning, the ability to make use of prior experiences to learn how to learn better in the future. Here we show that 8-month-old infants successfully engage in meta-learning within very short timespans after being exposed to a new learning environment. We developed a Bayesian model that captures how infants attribute informativity to incoming events, and how this process is optimized by the meta-parameters of their hierarchical models over the task structure. We fitted the model with infants' gaze behavior during a learning task. Our results reveal how infants actively use past experiences to generate new inductive biases that allow future learning to proceed faster.

SUBMITTER: Poli F

PROVIDER: S-EPMC10320826 | biostudies-literature | 2023

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Eight-Month-Old Infants Meta-Learn by Downweighting Irrelevant Evidence.

Poli Francesco F Ghilardi Tommaso T Mars Rogier B RB Hinne Max M Hunnius Sabine S

Open mind : discoveries in cognitive science 20230601

Infants learn to navigate the complexity of the physical and social world at an outstanding pace, but how they accomplish this learning is still largely unknown. Recent advances in human and artificial intelligence research propose that a key feature to achieving quick and efficient learning is meta-learning, the ability to make use of prior experiences to learn how to learn better in the future. Here we show that 8-month-old infants successfully engage in meta-learning within very short timespa ...[more]

PMID: 37416070

Similar Datasets

Project description:BackgroundMetabolic programming of glucose homeostasis in the first 1,000 days of life may impact lifelong metabolic and cardiovascular health. Continuous glucose monitoring (CGM) devices may help measure the impact of dietary intake on glucose rhythms and metabolism in infants during the complementary feeding period.ObjectivesDemonstrate the feasibility of CGM to measure and quantify glucose variability in response to infant feeding and to evaluate associations between macronutrient meal composition and glucose variability.MethodsThe "FreeStyle Libre Pro®" device interstitial glucose meter was applied to the anterior thigh of 10 healthy 6-12-month-old infants. Parents recorded food intake, time of feeding, and used daily dairies to record sleep time and duration. Descriptive statistics were employed for food intake, sleep and key glycemic parameters over three full days. Mixed linear models were used to assess glycemic changes.ResultsMid-day, afternoon, and evening feeds contained >30 g carbohydrate and induced higher 2-h iAUC (3.42, 3.41, and 3.50 mmol/L*h respectively) compared to early and mid-morning feedings with ≤25 g carbohydrates (iAUC 2.72 and 2.81 mmol/L*h, p < 0.05). Early morning and evening milk feedings contained approximately 9 g of fat and induced a longer time to reach maximal glucose value (Tmax; 75 and 68 min, respectively) compared to lower fat feedings (2.9-5.9 g; Tmax range: 34-60 min; p < 0.05). Incremental glucose value at time of food intake (C0) increased significantly from 0.24 ± 0.39 mM in early morning to 1.07 ± 0.57 mM in the evening (p < 0.05). Over the day, 70% of glucose values remained within the normal range (3.5-5.5 mmol/L), 10% were between 5.5-10 mmol/L, and 20% were < 3.5 mmol/L.ConclusionOur data support the feasibility of using CGM to measure glucose in 6-12-month-old infants. The observation of possible diurnal glucose variability and typical glucose values may have implications for future studies investigating metabolic adaptation to nutritional intake in early life.

Dataset Information

Eight-Month-Old Infants Meta-Learn by Downweighting Irrelevant Evidence.

Publications

Eight-Month-Old Infants Meta-Learn by Downweighting Irrelevant Evidence.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets