Chapter 1: Infancy: the physical World
What kinds of knowledge are central to human cognitive development?
- Naïve physics:
An intuitive understanding we have about objects in the physical world, e.g. that objects that are dropped will fall, solid objects cannot pass through other solid objects etc.
- Naïve psychology
- Naïve biology
About 50 years ago it was thought that infants did not develop a full object concept until around 18 months of age. This assumption seems to be quite wrong.
At least three types of learning also appear to be functioning from very early in development:
1. Associative learning:
The ability to make connections between events that are reliably associated.
2. Learning by imitation
3. Explanation based learning
4. Learning by analogy
MEMORY
The capacity to retain information over short periods of time is often called “short-term memory” or “working memory”. Rose et al. reported that memory span increased with age: age 5/7 months almost 25% kept 3 or 4 items in working memory and by 12 months, almost 50%. Recency effects were found at all ages tested.
6-months-olds can also retain memories for events, and do so over very long time periods.
Reactivation paradigm: A procedure in which the participant usually an infant, is given a reminder of an earlier learned, but apparently forgotten, memory that enables this memory to become accessible again.
Conclusion: Very young infants can develop long-term memories for causal events, and memory retrieval appears to be governed by the same cues that determine retrieval in adults.
Delayed imitation: another way of examining infants’ long-term memory for causal events, which is imitation of a previously seen behavior after a delay.
Mandler argues that retaining causal relations provides one of the major ways of organizing material that is to be remembered in a coherent and meaningful fashion.
There are two types of memory system in human:
- Implicit/procedural memory: memory that is not available to verbal report.
- Explicit/declarative memory: memories for earlier experiences that can be readily brought to mind and thought about.’
A probably incorrect assumption about infants is that they are generally assumed not to encode explicit or declarative memories until they become verbally competent, which is known as ‘infantile amnesia’.
PERCEPTION AND ATTENTION
Adequate attentional mechanisms appear to be available shortly after birth.
Attention in infancy
One way to study when attentional mechanisms in infants come under volitional control is to study their expectations of visual events. The development of visual expectancies requires the volitional control of visual attention.
At least by the age of 3,5 months, babies can control their own perceptual (attentional) activity. Dit volgt uit het onderzoek (Haith, 1988) van het tonen van stimuli aan kinderen aan de linker of rechterkant. Bij het 1e onderdeel werden de plaatjes om-en-om getoond, waardoor er dus een bepaalde voorspelbaarheid was en bij het 2e onderdeel werden de plaatjes random getoond, waardoor er geen voorspelbaarheid was.
Gilmore and Johnson (1995) have shown that, by the age of 6 months, infants can also control their visual attention over delays of at least 3-5 seconds. Bij dit onderzoek werden er 3 hokjes getoond, de linker en rechter leeg en in het midden een opvallend plaatje, vervolgens werden alle hokjes leeg en in die tijd bleven de kinderen naar dit hokje kijken totdat er plaatjes in de linker en rechter hok kwamen.
Visual preference and habituation
Visual preference technique: Infants are shown pairs of stimuli and a preference for looking at one indicates the ability to discriminate between the two.
One way to find out whether infants can in fact distinguish two equally preferred visual stimuli is to use the habituation paradigm: infants are presented a stimulus, usually visual or auditory, until it no longer attracts attention: recovery of attention to a new stimulus (dishabituation) indicates discrimination between familiar and new.
Research with neonates by Slater and colleagues has shown that infants can indeed discriminate a cross from a circle.
Cross-modal perception
The ability to match perceptual information across modalities (cross-modal perception) also appears to be present from early in life. Voorbeeld van connecties tussen zicht en voelen: baby’s krijgen een speen met of zonder uitsteeksels in hun mond, terwijl ze dit niet kunnen zien. Later bij het zien van deze spenen, geven ze de voorkeur aan de speen die in hun mond zat, ook al hebben ze deze niet gezien.
Spelke found that infants prefer to look at the visual event that matched to auditory soundtrack. This suggests understanding of cross-modal equivalence. We have a strong perceptual preference for congruence across different perceptual modalities, and this preference is present from early in life.
Processing interrelations between features: the differentiation of prototypes
Prototype formation: the formation of an internal prototypical or generalized representation of a class of stimuli.
The use of more natural categories and real features to study prototype formation is important, as the correlational structure of objects in the real world is quite complex.
Prototypes and statistical learning in infancy
Younger’s cartoon-animal experiments demonstrated that infants could code the correlational structure between the different features being manipulated by the experimenters. This suggests a form of statistical learning: using the regularities in input to learn which features co-occur together.
THE PERCEPTUAL STRUCTURE OF THE VISUAL WORLD
The evidence for prototype formation shows that infants can code the perceptual structure of objects in terms of the relationships (covariations) between different features of these objects. Further, we have seen that they can track conditional probabilities between objects that follow each other in particular sequences.
One way of measuring infants’ ability to process and represent spatial, numerical, and causal relations is to introduce violations of typical regularities in the relations between objects, which then results in physically “impossible” events. This is known as the “violation of expectation paradigm”, and has been widely used to study infant cognition.
Violation of expectation paradigm: infants are shown a physical event and then on test trials shown events that are either imcompatible (thus, violating expectation) or compatible with the event. Longer looking at the possible event indicates that the infants understand the physical principle involved.
Representing spatial relations
One way to test whether infants are sensitive to spatial relations is to use habituation. Voorbeeld: baby’s van 3 maand oud testen of ze het verschil weten tussen boven en onder dmv een stip boven of onder een lijntje. Deze baby’s hebben na gewenning voor het ene, voorkeur voor het onbekende andere stipje.
There is evidence that infants show the same abilities with far more complex stimuli.
Very young babies appear to be able to represent spatial relationships such as relative height, at least in an occlusion paradigm. Denk hierbij aan voorbeeld van het muurtje en de grote en klein konijn die wel of niet zichtbaar zouden moeten zijn, blz 21.
Baillargeon, DeVos, and Graber (1989) went on to demonstrate that 8-month-old infants could retain these spatial memories for up to 70 seconds: So “out of sight” is not necessarily “out of mind” for infants.
Coding of spatial position: can occur either in relation to one’s own position in space (egocentric) or to external landmarks (allocentric).
Representing occlusion relations
When an object is occluded by a second object, we as adults believe that it still exists. Babies seems to make similar assumptions about the existence of occluded objects (object permanence).
Representing support relations
Infants younger than 6,5 months perceive any amount of contact between objects to be sufficient to ensure stability. They operate with a simpler causal rule that no contact = object falls, and partial contact = object is supported, even when the partial contact is very partial indeed. Kinderen ouder dan 6,5 maand begrijpen al dat als bijvoorbeeld een mok teveel op de rand staat, dat deze dan valt. Komt voornamelijk omdat ze dan al kunnen zitten in een stoel etc en er hierdoor meer mee geconfronteerd worden.
Representing containment relations
Continuity principle (Spelke): objects exist continuously in time and space.
Baillargeon and Wang suggested that infants treated containment events as distinct from occlusion events. They did not generalize their knowledge of a variable like height from one type of event to the other. Infants sort physical events into categories, and learn separately how each category operates. Hespos and Baillargeon reasoned that infants identify the variable height as relevant at about 4,5 months in occlusion paradigms and at about 7,5 months in containment paradigms. They therefore argued that the younger infants should search for the frog behind the tall occlude in the occlusion paradigm, but should not preferentially search inside the tall container in the containment paradigm. The older infants were expected to search for the frog successfully in both paradigms. This was exactly what they found. Infants’ physical reasoning systems are designed to acquire event-specific expectations rather than event-general principles.
What is measured in the violation of expectation paradigm?
