Nov 11 ====== Jusczyk:1999 ------------ How do the people in the experiment measure how long a 7.5 month old listens to a word? Maybe I missed it, but do we know if the research is about British English or American English? Wouldn't these findings suggest that in languages with more regular prosodic features children can segment words faster/easier? The paper notes how infants learning English show a greater ability to recognise words with a strong/weak stress pattern and find it harder to correctly segment words with weak/strong stress pattern. It is noted that in English, the strong/weak stress pattern is more predominant. My question is how this would compare for learners of a language which does not have predominant strong/weak stress preference? Is the ability to recognise the strong/weak stress pattern related more to the frequency with which infants hear this pattern in continuous speech, or is there evidence to suggest that the strong/weak stress pattern is easier for an infant to recognise in general? Do infants that grow up with two or maybe even three languages have a similar development process as those just learning one? I noticed that this question is also asked at a later point in the paper but it's said that knowing regularities in one language is not helpful for another language. Does this mean that the development may be slower for all languages an infant is learning or do the infants have a "primary language" they develop faster in than the other(s)? What is the correlation between the time infants listen to words and their understanding of it? How it can be measured? (for me it wasn't very clear from the article) Does the process of word segmentation by infants significantly differ between different languages? For example some languages are spoken faster than others, would that make a difference? I was wondering, how exactly we can know whether infants learned certain features or not: What exactly is meant with "infants listen longer" to a certain sequence, what will they do if they don't listen / how do we know that they don't listen. And how do we know that this means that they learned the corresponding features? 1. How reliable is it to conclude that infants segment words in fluent speech only by looking at listening times? 2. English-learning 15-month-old infants showed response bias to target words in sentence-final position (Fernald et al.). If an explanation for the response bias of 15-month-old infants is that they need time to segment words and retrieve meanings, what would be an explanation that other studies found no response bias from 7.5-month-old infants? 3. In the paper, pauses were not mentioned, which are intuitively crucial to segmenting words. When adults talk to infants, adults speak sometimes deliberately slower such that there are "gaps" between syllables within a word. Do infants distinguish those gaps from actual pauses? If so, how? When encountering compounds like for example "smalltalk" or a more difficult german example: "Donaudampfschiffahrtsgesellschaftskapitän" does the word segmentation of infants improve when some or all of the words in the compound are known? Besides stress patterns and extracting meanings affecting the infant's ability to extract words, what other factors can affect this ability? What counts as the child listening to the learned words or being interested in them? How reliable is the data when using children as old as a few months in experiments? In this paper, the authors made several assumptions, for example, they assumed that an infant would listen to a passage which contains a learned word longer than a control passage. Are there other important assumptions for interpreting the experiment results? Do they actually hold?