Do well-connected words help us learn new associations?
Words with many connections in our mental lexicon make it easier to form and remember new associations.
Source
Evidence for preferential attachment: Words that are more well connected in semantic networks are better at acquiring new links in paired-associate learning
What they did
Researchers investigated whether words with higher semantic connectivity are better at acquiring new links using a paired-associate learning task. Experiments 1 and 2 tested 52 young adults who memorized 40 pairs of real English words under timed conditions. Experiment 3 simplified the design using 46 young adults who learned 80 real cue words paired with novel pseudowords to isolate semantic effects from word familiarity.
What they found
Across all three experiments, cue words with higher degree centrality significantly enhanced the recall and recognition of their paired response words. This memory advantage was unique to network connectivity and could not be explained by other factors like physical concreteness or word frequency. Furthermore, further exploratory analysis revealed that a word's incoming connections drove this memory boost more than its outgoing connections.
The limits
What it doesn't show
The study does not definitively determine the underlying cognitive mechanism, leaving it open whether memory enhancement stems from physical network proximity or from contextual flexibility. The experiments focused solely on younger adult samples, meaning the results may not generalize to child language learners or older adults. Lastly, the design cannot confirm if these results hold for the active production of new words, as Experiment 3 relied on a recognition task due to low initial recall scores.
Key terms
- paired-associate learning
- A classic memory task where individuals memorize pairs of items and are later asked to recall or recognize one item when prompted with its partner.
- degree centrality
- A measure of how well-connected a node is in a network, calculated as the sum of its incoming and outgoing links.
- preferential attachment
- A network growth process where highly connected nodes are more likely to acquire new connections than poorly connected ones.
- in-degree
- The number of times a word is generated as a response by participants in a free-association task.
- out-degree
- The number of unique responses a specific cue word triggers in a free-association task.
- pseudoword
- A fake but pronounceable letter string that follows language rules but has no real meaning, like 'arruity'.
Flashcards
Want these cards to stick?
Save the deck to NoteFren and study it with spaced repetition.
Quiz yourself
According to Steyvers and Tenenbaum (2005), what three structural properties characterize adult semantic networks?
Common questions
Why were pseudowords used in the third experiment?
Using novel pseudowords stripped away any pre-existing real-world associations or linguistic biases, allowing researchers to observe a clean measure of how real cue words help individuals learn completely new vocabulary.
Does a word's frequency in everyday speech explain these results?
No, because the researchers statistically controlled for word frequency in their models, showing that a word's network connectivity uniquely predicts memory recall independent of how common the word is.
What is the 'rich get richer' mechanism in this context?
It suggests that words that already have many connections in our mental lexicon are more efficient at latching onto and retaining new links during learning processes.
Which type of connection matters more: in-degree or out-degree?
Exploratory findings suggested that in-degree, which represents how often a word is triggered by other words, captured significantly more variance in memory accuracy than out-degree.
More on Learning