Psychology 475/575
Cognitive Development
Dr. Lou Moses

Midterm 2 Answer Key

Note: The information presented below is not intended to represent complete model answers to the questions. Rather it is intended to give you at least a rough sense of what was necessary to do well on the exam.

1) Text pp. 302-304

The constraints approach to semantic development is designed to answer the question of how children figure out the meanings of words, when those words can have an indefinitely large number of possible meanings. According to the constraints approach children bring to the word learning situation biases or tendencies to favor some hypotheses about word meanings over others.
Examples of constraints are the whole object assumption ( assume that a new word refers to the whole object rather than to a part or property), the taxonomic assumption (assume that the word refers to a category of things of like kind rather than to a thematic relation in which the thing participates), and the mutual exclusivity assumption (assume that things can only have one label; a new word must then refer to something for which the child does not already have a label).

2) Lecture by Dr. Baldwin, reader (Baldwin article).

The ability to draw inferences about the intentions of others helps infants to determine the intended referents of speakers’ utterances (are they talking about X or Y?). Baldwin’s research on “discrepant labeling” suggests that appreciating others’ intentions does affect language learning. In this research 18-month-old infants were focused on a different toy than that which the experimenter was targeting with a novel label.. nevertheless, they did not link the novel label to their own toy. Instead, they appropriately checked to see what the experimenter was looking at and linked the label to that object. In other words, they appeared to be taking into account the experimenter’s intentions in making correct word-object links.

3) Lecture; Wynn article in the reader; text (pp. 129-132).

Recent research suggests that infants can both perceive numerical information and calculate the results of simple arithmetic operations on small numbers of items. Evidence for the former ability comes from experiments showing that infants can discriminate between small numbers of items (e.g., 2 vs. 3) on the basis of number. Evidence for the latter ability comes from the Wynn experiments showing that infants apparently represent the number of items present behind a screen and then make inferences about how that number changes as items are either added or subtracted.

4) Lecture; Mischel article in reader

Delay of gratification is the ability to postpone a smaller, immediate reward in order to obtain a larger delayed one (e.g., one cookie now vs. two later). Factors that affect children's ability to delay gratification include age, keeping the object out of sight; engaging in distracting activity; think about abstract vs. arousing properties of the reward.

5) Lecture. 

Conceptual competence refers to having the basic concepts in a particular domain. Utilizational competence refers to knowing when to use these concepts. Research with Brazilian child street vendors suggests that these children have conceptual competence but not utilizational competence in the domain of numerical reasoning. They can do complex calculations when dealing with the familiar goods they sell everyday suggesting that they have the conceptual competence, but they cannot do the very same calculations when they are presented in an abstract context, suggesting that they do not have utilizational competence.

6) Lecture; text p 221

Self-recognition has been measured using the mirror task. Infants are first surreptitiously marked on the nose with rouge. They are then placed in front of a mirror. If infants then reach up and touch the mark on their nose (as opposed to touching the mark reflected in the mirror), they are said to have recognized themselves.
Self-recognition has also been measured by looking at whether infants recognize when they are being imitated (and hence see themselves reflected in the actions of another person). Infants are confronted with two experimenters, one of whom is imitating them and the other of whom is imitating a different baby (or in the 2nd experiment is merely performing actions that are in temporal synchrony with those of the current baby). Recognition of imitation is reflected by longer looking times with respect to the imitating experimenter.
The mirror task suggests that children begin to recognize themselves between about 15 and 24 months of age. The imitation task suggests that they recognize themselves as early as 14 months.
Self recognition does relate to other important developments. Examples include embarrassment (children who self recognize are more likely to show embarrassment in social situations such as being asked to dance by the experimenter), empathy (children who self recognize are more likely to show empathy, as in the Bischof-Kohler experiment in which they attempt to console the experimenter and/or the bear) and the development of standards (an awareness of success vs. failure at meeting standards emerges at about the same time as self recognition).

7) Lecture, text (pp. 102-105), DeLoache article in reader.

DeLoache believes it is a dual coding problem (children have difficulty in thinking of the scale model as both an object in its own right as well as a symbol). Alternative explanations and evidence against them are as follows. 1. Memory. Maybe children have forgotten where little snoopy was hidden. Against this, they can go back to find little snoopy after unsuccessfully failing to find big snoopy. 2. Language. Maybe children are confused by the language, especially the phrase “the same place”. Against this children do well when a photograph rather than a scale model is used, even though the language is very similar. 3. Perceptual Similarity. Perhaps children simply don’t recognize the correspondence between the scale model and the larger room because they look too different. Against this, when the rooms are made to be identical, 2-year-olds still find it difficult to locate big snoopy. Evidence positively supporting DeLoache’s explanation comes from (a) model manipulation experiments (playing with the model causes performance to deteriorate; putting the model behind a glass barrier enhances performance), (b) photograph experiments (children understand the relation between a photo and the larger space better than that between a scale model and the larger space) (c) comparable findings from pretense (2 1/2 -year-olds have difficulty pretending that an object with a known function could be something entirely different; they have less difficulty, however, pretending that an object with ambiguous or multiple functions could be something different) and (d) the Shrunken Room experiment (children have little difficulty when dual coding is not required. i.e., when the room is a shrunken version of the original rather than a representation of it).