Article summary with Making sense of the world: Infant learning from a predictive processing perspective by Köster a.o. - 2020

What is the predictive processing perspective?

Predictive processing (PP) is a theory that proposes that successful navigation in the environment relies on the organism´s ability to optimize predictions about how one´s own behavior will affect proprioceptive experiences and how social and physical entities in the outer world behave. Sensory inputs provide highly incomplete and variable information about our complex environment which changes with behavioral navigation. The brain needs to improve inferences on the basis of sensory inputs and minimizing prediction errors. As prediction errors are reduced, the accuracy of internal predictive models is increased.

What is the hierarchical system in predictive processing?

According to the PP framework, the brain consists of lower and higher level areas that are organized in a hierarchical system. The different levels continuously communicate with one another. Predictions are formed at every level of the hierarchy. Mismatches between what is predicted and what is perceived are learning opportunities. The prediction errors are sent back to higher levels where the prediction was made, to update existing predictions and improve predictive models.

How can the predictive processing perspective help explain cognitive development in early infancy?

The PP perspective states that the main purpose of learning processes is to minimize prediction errors. As infants grow they need to interpret sensory information and translate their experience into appropriate behavioral responses in increasingly sophisticated ways. The PP framework may provide a perspective on several phenomena of infant development and learning. Phenomena that fit into the PP framework include infants´ statistical learning, motor learning, proprioperception, and the emerging representations and expectations about the physical world.

How can statistical learning in infants be viewed in a PP framework?

The brain constantly computes the probability of events in the environment on the basis of incoming sensory information. Statistical regularities of the environment cause the brain to form probabilistic models of its environment. There is evidence that infants generate such probabilistic models to represent statistical regularities in their environment. For example, infants learn to segment artificial language on the basis of statistical information such as transitional probabilities between elements.

How can infants´ motor learning be viewed in a PP framework?

Infants´ early motor behavior serves the generation of internal models and maps movements to consequences. These internal models can then be used to predict the consequences of their own actions as well as the goals of another person´s actions. This allows for imitation learning which seems to be at least partially fueled by infants´ drive to minimize prediction error. Infants adjust their predictive models about their novel abilities (such as walking and falling down) to interact with the environment on the basis of their prior sensory experiences.

What is a comparator model?

Infants focus their attention selectively on new events and objects with which they are not familiar. They lose interest in perceptual stimuli that they have repeatedly encountered (habituation) and their attention revives for new stimuli (dishabituation). When orienting toward a stimulus, the infant compares the sensory information with an existing neuronal representation. If the stimulus deviates from the existing representation, an orienting response leads to increased attention and the formation or update of a neuronal representation of the stimulus.

How can infants´ basic understanding about the world be viewed in a PP framework?

The main purpose of processing sensory information is not simply representing the external world, but the generation of appropriate behavioral responses by making inferences on the consequences of behavioral responses for sensory inputs. Infants learning may then be conceptualized as the formation and refinement of predictive models about animate and physical entities in relation to the infant´s own body movements and actions.

What are violation-of-expectation paradigms?

In violation-of-expectation (VOE) paradigms, infants´ orienting response is taken as an indicator of infants´ basic concepts about the environment. It is basically the idea that infants will show surprise when witnessing an impossible event. The PP perspective can offer an explanation for infants´ VOE responses. VOE responses indicate infants´ processing of prediction errors, which require them to refine prior predictions. Infants actively seek to reduce their uncertainties, which may be especially important for objects that don´t comply with their existing predictive models.